Introduction

PALEVO

1631-0683

Elsevier

S1631-0683(17)30137-9

10.1016/j.crpv.2017.11.003

Research article

Human Palaeontology and Prehistory

The Kocabaş hominin (Denizli Basin, Turkey) at the crossroads of Eurasia: New insights from morphometric and cladistic analyses

L’homininé de Kocabaş (bassin de Denizli, Turquie) au carrefour de l’Eurasie : nouvelles données obtenues à partir d’analyses morphométriques et cladistiques

Coppens

Yves

Vialet

Amélie

Vialet

Amélie

amelievialet@yahoo.fr ^a Prat

Sandrine

^b Wils

Patricia

^c Cihat Alçiçek

Mehmet

a UMR 7194, CNRS, Muséum national d’histoire naturelle, UPVD CERP de Tautavel, 1, rue René-Panhard, 75013 Paris, France a UMR 7194, CNRS, Muséum national d’histoire naturelle, UPVD CERP de Tautavel

1, rue René-Panhard

Paris

75013

France

b UMR 7194, CNRS, Muséum national d’Histoire naturelle, UPVD, Musée de l’Homme, 17, place du Trocadéro, 75116 Paris, France b UMR 7194, CNRS, Muséum national d’Histoire naturelle, UPVD, Musée de l’Homme

17, place du Trocadéro

Paris

75116

France

c UMS 2700, outils et méthodes de la systématique intégrative, Muséum national d’Histoire naturelle, 43, rue Buffon, 75005 Paris, France c UMS 2700, outils et méthodes de la systématique intégrative, Muséum national d’Histoire naturelle

43, rue Buffon

Paris

75005

France

d Department of Geology, Pamukkale University, 20070 Denizli, Turkey d Department of Geology, Pamukkale University Denizli

20070

Turkey

17 1-2

S1631-0683(18)X0003-7

Hominins and tools. Expansion from Africa towards Eurasia / Homininés et outils. Expansions depuis l’Afrique vers l’Eurasie

17 32

2017

Académie des sciences

Full (PDF)

The Kocabaş skullcap (Denizli Basin), dated between 1.2 and 1.6 Ma, is the only ancient hominin fossil from Turkey and is part of discussions focusing on the first settlement outside the African continent. Our morphometric study tends to link this specimen with the African fossils, Homo ergaster and early Homo erectus, and to distinguish it from the specimens from Dmanisi and Asian Homo erectus. These results are confirmed by a cladistic analysis, which shows a separation of Kocabaş from the Eurasian clade comprising the Dmanisi hominins and grouping it with the African fossils dated to around 1 Ma (KNM-OL 45500, Daka-Bouri BouVP2/66, Buia UA31). As in the Kocabaş fossil, the divergence of the frontal bone is not very marked on these latter fossils and the temporal lines are separated on the parietal bone. The Kocabaş skull seems to point to a different evolutionary history than that of the Dmanisi fossils, and could reflect a later “out-of-Africa” expansion.

Seul fossile d’homininé ancien en Turquie, la calotte crânienne de Kocabaş (bassin de Denizli), datée entre 1,2 et 1,6 Ma, s’inscrit dans la discussion sur les premiers peuplements en dehors du continent africain. Notre étude morphométrique tend à le rapprocher des fossiles africains, Homo ergaster et Homo erectus récents, et à le distinguer des spécimens de Dmanissi et des Homo erectus asiatiques. Ces résultats sont confirmés par une analyse cladistique qui montre une séparation du fossile de Kocabaş du clade eurasiatique qui inclut les homininés de Dmanissi, et le regroupe avec les fossiles africains datés autour d’1 Ma (KNM-OL 45500, Daka-Bouri BouVP2/66, Buia UA31). Ces derniers partagent avec Kocabaş une divergence de l’os frontal peu marquée et une séparation des lignes temporales sur l’os pariétal. Le crâne de Kocabaş semble témoigner d’une histoire évolutive différente de celle des fossiles de Dmanissi, qui pourrait correspondre à une expansion hors d’Afrique plus tardive.

Hominin, Turkey, Morphometry, Cladistics, Out of Africa

Homininé, Turquie, Morphométrie, Cladistique, Expansion hors d’Afrique

presented

Handled by Yves Coppens

1 Introduction

Several hominin fossils are known from Africa after 7 Ma, when the first early Homo is recorded at 2.8 Ma (Villmoare et al., 2015). There is a significant time lag of about 1 Ma between Africa and Eurasia, where the first human fossils are dated to 1.77 Ma at the site of Dmanisi in Georgia. This latter is the earliest known direct evidence of the “out-of-Africa” expansion. Based on lithic assemblages discovered in Asia, these first waves of settlement could be considerably older than 2 Ma (Prat, 2018, this issue). However, these discoveries remain scarce. In this context, the Kocabaş skull, from the Denizli Basin in Turkey, provides important evidence of early human presence in the Middle East, or an “out-of-Africa” expansion along the Levantine corridor, and possible dispersal toward Europe – via the Bosphorus Strait – and toward Asia – across mountainous territories such as the Taurus, Zagros, and the Caucasus. Although the Kocabaş skullcap is very fragmented, it fills a paleoanthropological gap between 1.6 and 1.2 Ma, not only in Africa but also in Eurasia. Indeed, there is no fossil between OH9 from Olduvai in Tanzania and the one-million-year-old Homo erectus-like hominins from East Africa (KNM-OL 45500, Daka-Bouri BouVP2/66, Buia UA 31). In Europe, apart from the deciduous tooth from the Orce Basin dated to 1.4 Ma (Toro-Moyano et al., 2013), the oldest human fossils are from the Atapuerca–Sima del Elefante site dated to 1.2 Ma (Carbonell et al., 2008).

The goal of this paper is, on one hand, to characterize morphometrically the Kocabaş fossil and, on the other hand, to test its link with other Pleistocene hominins from Africa and Asia using cladistic analysis. The main question is to establish whether the Turkish hominin, found at the crossroads of Africa and Asia between 1.6 and 1.2 Ma, is indeed closer to African or Asian specimens.

2 Background

The Kocabaş fossil was discovered in 2002 by one of us (M.C.A.) in one of the quarries in the Denizli Basin (Alçiçek and Alçiçek, 2014). The small skull was sliced by the blades used to cut blocks in this travertine quarry. Although the discovery of the skull was accidental, associated fauna is abundant in the find horizon (Alçiçek, 2014, Boulbes et al., 2014). This fossiliferous horizon from which the Kocabaş fossil comes from, is identified as the upper travertine (UT) (Lebatard et al., 2014a and Lebatard et al., 2014b), which was the only quarried travertine at the time of discovery in 2002, and the only unit to have yielded paleontological material up until now.

The stratigraphic and chronological context of the UT bearing the hominin skull and an abundant Upper Villafranchian fauna were comprehensively defined during the field missions in 2011 and 2012. This fieldwork has been focused on the deposits from the Faber Quarry where the succession reaches a depth of over 90 m and contains preserved levels comparable to those from which the fossils were extracted in 2002 within the UT. This UT unit is situated stratigraphically between two fluviatile levels. The results obtained from ²⁶Al/¹⁰Be cosmogenic nuclides analysis on the overlying and underlying fluviatile conglomerate levels (Lebatard et al., 2014a and Lebatard et al., 2014b) indicate an age of 1.1 Ma and 1.6 Ma respectively, setting a chronological bracket between these two dates for the Kocabaş fauna and the human fossil. The magnetostratigraphy for the whole sequence of the deposits (Khatib et al., 2014) allows a correlation of the base of the upper fluvial unit (normal polarity) with the Cobb Mountain excursion, dated to 1.22 Ma. Moreover, the biochronology of the fauna Archidiskodon merionalis meridionalis (Elephantidae) and Palaeotragus (Giraffidae) associated with the skull (Boulbes et al., 2014) confirms this chronological framework between 1.2 and 1.6 Ma.

No lithic artefacts are strictly associated with the human and animal fossils from the UT of the Denizli Basin. Archaeological prospection in the surrounding terraces has just started recently (Maddy et al., 2015; Aytek comm. pers.), leading to the discovery of lithic artefacts, mostly on the surface. More generally, few Lower Palaeolithic sites are known in Turkey (Dinçer, 2016) and none are as old as the Kocabaş locality. We have no evidence for the moment to ascertain whether at 1.2/1.6 Ma, the technological mode could be related to Oldowan or Acheulean techno-complexes. However, the Dursunlu site (Güleç et al., 2009) and Yarimburgaz Cave (Howell et al., 1996) have yielded stone flakes. The open-air site of Kaletepe Deresi 3 attested to the presence of well-shaped handaxes made of obsidian (Slimak et al., 2004 and Slimak et al., 2008) dated to the middle Pleistocene (Tryon et al., 2009). Some handaxes were discovered in the lower levels of the Karain E Cave (Taşkıran, 2008, Taşkıran, 2018 this issue). At this site, the Mousterian industry is well represented in the upper levels in association with 20 fragmentary Neandertal-like human remains (Chevalier et al., 2015) older than 125 ka (Otte et al., 1998).

3 Anthropological settings

The Kocabaş skull was attributed to Homo erectus (Kappelman et al., 2008). It is composed of 3 fragments separated at the cranial sutures, which were not completely fused due to the young age of the specimen (Fig. 1). The three cranial fragments were scanned using the Philips helical scanner at the Pamukkale teaching hospital, in Denizli, on September 14th, 2009; slice thickness was 0.80 mm, the space between slices was 0.4 mm (field of view: 20 cm, matrix: 512 × 512, power: 175 mA, intensity: 120 kV).

A first virtual reconstruction was carried out based on the CT data (Vialet et al., 2011 and Vialet et al., 2012). This paper concerns the second virtual reconstruction with Geomagic Studio 12 software by A.V. and P.W. (technical platform ASIM, at the MNHN). In this reconstruction, called “Kocabaş 2”, the right part of the frontal bone is more accurately situated. Indeed, there is a gap of a few millimeters between the parietal and frontal sutures which prevents the direct anatomical connection between the two bones, as was previously thought and consequently applied to the first reconstruction. This reconstruction proceeded as follows. First, the two parietal fragments (right and left) were rearticulated following the congruence of their sagittal suture. Such rearticulation could be done using different angles. To ensure that the rearticulated bones were not out of the initial volume, we confirmed that they are situated between the two reference planes (i.e. the cutting sections made by the blades of the machine shown by the exposure of diploic tissue at the central part of the frontal and posterior part of the parietals). Then, using the mirror-imaging technique, we duplicated the right part of the frontal bone to reconstruct the left one. Finally, the anterior block (made by the completed frontal) was connected to the posterior block (made by the rearticulated parietals) using as guidelines the sagittal suture and the frontal crest (Fig. 2).

To conclude, the two reconstructions are different. This is the reason why the measurements taken for each one are not exactly the same, resulting in a slightly larger skull for the Kocabaş 2 reconstruction (Table 1).

Based on several morphological changes of various size observed on the endocranial surface of the right frontal fragment, Kappelman and co-workers (Kappelman et al., 2008) argued for the presence of a Leptomeningitis tuberculosa (tuberculous meningitis). However, the validity of this diagnosis was refuted and critical comments were made by Roberts and co-workers (Roberts et al., 2009).

Complementary studies employing new macromorphological and micro-CT analysis on the endocranial lesions are ongoing in order to clarify their morphological aspect, especially by focusing on soft tissue imprints and their possible inner extension.

4 Material

Casts of the main fossil hominin skulls from the lower to upper Pleistocene in Africa, Asia, and Europe were considered in this study (Table 2). They are stored at the National Museum of Natural History in Paris. 28 skulls in total were included in the 2D morphometric analysis and among them, 20 were used for cladistic analysis (including Sts5 as the outgroup specimen).

5 Methods 5.1 2D morphometry

Standard metrical measurements from Martin and Saller (1957) were used to characterize the development of the frontal bone (Table 3). Principal component analysis was generated by Past 3.1. software (Hammer et al., 2001).

5.2 Cladistic analysis

To investigate the phylogenetic relationships of the Kocabaş hominin, we carried out a cladistic analysis using the characters defined in Table 4. Quotation was done on the originals and casts of the fossil skulls and from the literature for KNM-ER42700 (Spoor et al., 2007), Buia UA 31 (Abbate et al., 1998, Bruner et al., 2016), Daka-Bouri BouVP 2/66 (Asfaw et al., 2002 and Asfaw et al., 2008) and KNM-OL 45500 (Potts et al., 2004).

Eighteen characters of the frontal bone were used; 14 are morphological characters and 4 are metric characters. As there is no consensus on the taxonomic classification of numerous hominin specimens, we define the Operational Taxonomic Unit (OTU) by the specimen or group of specimens rather than by the species (as often used). This approach has been used by Cameron et al., 2004, Caparros, 1997, Gilbert et al., 2003, Mounier and Caparros, 2015, Mounier et al., 2016, Prat, 2004, Prat, 2005 and Zeitoun, 2000 and Zeitoun et al. (2016). Character polarity has been determined by rooting the outgroup. The polymorphism is coded as multiple states (0&1, 0&2, 1&2, 0&1&2) with the polymorphism option of the Paup 4.01 software (Swofford, 1998). The quantitative characters were coded using the method proposed by Thiele (1993) (after a logarithmic transformation of data, these latter being standardized using the formula x _s = ((x–min)/(max–min)) × n; n = maximum number of ordered states allowable by the algorithm used (32 for PAUP). There has been considerable debate concerning methods of coding quantitative characters. The method of Thiele (1993) was used because it allows the coding of all characters in a similar manner. The data were computed in a non-arbitrary way to avoid any preconceived phylogenetic hypothesis. Most of these characters have been taken from Zeitoun (2000) and Prat, 2000, Prat, 2004 and Prat, 2005. Some of them have been defined in this study (Table 4). The characters have been coded on ten OTU and the allocation of the character state for each one is given in Table 5. The OTU are as follows: Australopithecus africanus (Sts5), which is the outgroup in the analysis; early Homo (which comprises some specimens commonly allocated to Homo habilis and Homo rudolfensis: OH 24, KNM-ER 1470; KNM-ER 1813; Stw 53); early African Homo erectus/Homo ergaster (KNM-ER 3733, KNM-ER 3883; OH 9, KNM-ER 42700); Homo georgicus (D2280 and D2282); late African Homo erectus (Buia UA 31, Daka-Bouri BouVP2/66, KNM-OL 45500); Kocabaş 2; Indonesian Homo erectus (Sangiran 17); Chinese Homo erectus (ZKD 3, 11,12; Nankin 1). Except the individuals from this latter group, Asian and European middle and upper pleistocene specimens were not considered, as our analysis is more focused on the relationships between the Kocabaş fossil and lower and early middle Pleistocene hominins.

To bring to light the homoplasies and synapomorphies, we performed a parsimony analysis following the protocol developed by Miguel Caparros in Caparros (1997) and Mounier and Caparros (2015). A first low-level analysis is conducted with a branch and bound search in order to ascertain the phylogenetic information content of the character by its retention index. The retention index (RI) is calculated by subtracting the observed tree-length from the maximum possible tree-length and then dividing this value by the difference between the maximum and the minimum lengths (Archie, 1989 and Farris, 1989). All characters are unordered and all have equal weight. In a second step, 18 characters were reanalyzed after reweighting with the rescale consistency index (RC = CI (consistency index) × RI (retention index)). The consistency index (CI) is calculated as the minimum possible tree-length divided by the observed tree-length (Farris, 1989 and Kluge and Farris, 1969). If there is no homoplasy in a tree, then its observed length equals the minimum tree-length and the CI equals one. If homoplasy is present, then the CI is less than one. We use the ACCTRAN method (which maximizes reversals) in order to resolve the ambiguities of the character changes at the hypothetical ancestor nodes, as it provides more character support to clades.

The most parsimonious trees were obtained using the heuristic search algorithm with the branch-swapping algorithm (TBR). We use PAUP version 4.01 (Swofford, 1998), and MacClade (v. 4.06) in order to illustrate the supporting characters at the hypothetical ancestor's nodes.

6 Results 6.1 Anthropological study–2D morphometry

Metrics are presented in Table 6.

The PCA was computed based on 4 variables (M9, M10, M43, M29.2) measured on 28 fossils. On the first axis (which accounts for 82.5% of the total variance), individuals are spread following their global size (Fig. 3). Kocabaş 2 is among the Homo erectus s.l. from Africa and Asia. The Dmanisi fossils are smaller and the Homo heidelbergensis, late Homo erectus and Neandertals are bigger.

On the second axis (11.3% of the total variance), the length of the frontal bone separates the fossils. Kocabaş 2 is distinct not only from the Zhoukoudian Homo erectus but also from Sangiran 17, OH9 and KNM-ER3883. It is closer to African fossils such as KNM-ER3733 and Daka-Bouri BouVP2/66 (Fig. 3) which show a relatively shorter frontal scale.

The distribution of the fossils across the third axis (4% of the total variance) is linked with the width of the supra-orbital torus (Fig. 4). Again, Kocabaş 2 is in an intermediate position between the Asian Homo erectus (Zhoukoudian with a gracile supra-orbital torus) and the African ones (OH9, Daka-Bouri BouVP2/66 with a strong supra-orbital torus). For this feature, the Turkish specimen is closer to the Dmanisi and the East Turkana (KNM-ER3733 and ER3883) specimens.

The weak divergence of the frontal bone (proportion between minimal-M9 and maximal-M10 frontal widths) is expressed on the fourth axis (2.1%), on which the Kocabaş 2 skull is far from the Asian and African Homo erectus, except OH9 (Fig. 5).

The results of this study allow us: •

to link Kocabaş 2 with early and late African Homo erectus;

•

to distinguish it from the early Homo specimens and from the small Georgian hominins;

•

to distinguish it from the Asian Homo erectus.

In comparison to the latter, Kocabaş 2 shows a shorter and posteriorly narrower frontal bone and a more prominent supra-orbital torus. With its small size, the Turkish skull is clearly distinct from the middle and upper Pleistocene hominins. Such results confirm the previous studies based on anatomical features and 3D morphometrics (Vialet et al., 2012 and Vialet et al., 2014).

6.2 Cladistic analysis

Concerning the phylogenetic analysis based on 18 characters and 7 specimens or group of specimens in the ingroup, three trees have been obtained on the first step of the analysis. The tree length is 31, consistency index (CI) = 0.7097; retention index (RI) = 0.5714; and rescaled consistency index (RC) = 0.4055.

A heuristic search based on 18 frontal features (Table 7), entered with equal weights, identified 9 morphological characters that contain phylogenetic information (RI ≥ 0.5). We identify 6 true synapomorphies (CI, RI, ad RC = 1, HI = 0); characters # 5 Projection of the temporal band on the frontal bone; # 6 Bregmatic eminence overlapped on the parietal bone; # 7 Supratrigonal depression; # 14 Sagittal keel on the anterior half of the sagittal suture; # 15 Postcoronal depression; # 17 Temporal lines separated on the parietal bone.

For the second step, we used a branch and bound algorithm to analyze the reweighted characters by respective RCs. Three trees have been obtained. The strict consensus tree is shown in Fig. 6. The tree length is 13.66, retention index (RI) = 0.8790; consistency index (CI) = 0.9051 and rescaled consistency index (RC) = 0.7956.

At Node A, the hypothetical ancestor of the group (Buia UA 31, Daka-Bouri BouVP2/66, OL 45500 and Kocabaş) is well supported by one synapomorphy (#17: superior and inferior temporal lines changing from fused to separated on the parietal bone, they do not form a strong ridge). This node is also supported by one informative homoplasy (# 12) frontal bone changing from divergent to less divergent.

Kocabaş exhibits the following informative characters (RI ≥ 0.5): •

# 5–Projection of the temporal band on the frontal bone (RI = 1): changes from flat to prominent and strongly prominent, which is shared with the group Buia UA 31, Daka-Bouri BouVP2/66, KNM-OL45500;

•

# 7–Supratrigonal depression (RI = 1): changes from absent to present, which is shared with the group Buia UA 31, Daka-Bouri BouVP2/66, KNM-OL45500;

•

# 8–Height of the lateral part of the supra-orbital torus (RI = 0.5): changes from weak to developed, which is shared with the group (KNM-ER 3733, 3883, 42700, OH9) and the group (Buia UA 31, Daka-Bouri BouVP2/66, KNM-OL45500) and a convergence with the hypothetical ancestor of the group (Zhoukoudian 3, 11, 12, Nankin 1) and Sangiran 17;

•

# 11–Postorbital constriction (RI = 0.5): changes from developed to less developed which is a convergence with the hypothetical ancestor of Zhoukoudian 3, 11, 12, Nankin 1 and Sangiran 17. The decrease in postorbital constriction can be associated with increased flexion of the anterior cranial base (Cameron et al., 2004);

•

# 12–Frontal bone divergence (RI = 0.5): changes from divergent to less divergent; this is shared with the group (Buia UA 31, Daka-Bouri BouVP2/66, KNM-OL45500) and a convergence with Sangiran 17;

•

# 17–Temporal lines separated on the parietal bone (RI = 1): superior and inferior temporal lines separated on the parietal bone, they do not form a strong ridge, changing from fused to separated, which is shared with the group (Buia UA 31, Daka-Bouri BouVP2/66, KNM-OL45500).

Furthermore, the Kocabaş specimen shows a conical orbital roof (#18, CI = 0,5; RI = 0), which is also observed in the group (OH 24, KNM-ER 1813, 1470, Stw 53).

7 Discussion

Our multivariate analysis, based on the main measurements of the frontal bone, places the Kocabaş fossil in the Homo erectus s. l. group (i.e. Zhoukoudian and Nankin 1 from China, KNM-ER3733, KNM-ER3883 from East Africa and the late African Homo erectus such as OH9 and Daka-Bouri BouVP2/66). Its overall size is greater than that of early Homo and the Dmanisi specimens and smaller than Middle and Upper Pleistocene hominins. However, the Turkish fossil differs from the Asian Homo erectus by the shortness of its frontal scale which is close to the African fossils, KNM-ER3733 and Daka-Bouri BouVP2/66. In our opinion, this difference, dealing with the proportion of the frontal bone and its elongation, is a very significant feature linked with endocranial development. Our conclusions differ from Aytek and Harvati's, who argued for a closeness between the Kocabaş fossil and, not only the Chinese Zhoukoudian III specimen, but also European hominins such as Ceprano and Arago (Aytek and Harvati, 2016). It is important to recall that their analysis only focuses on the supra-orbital torus which is a strongly dimorphic area. However, it is true that the Kocabaş and Zhoukoudian III specimens are morphologically close for this anatomical region and they are not far from the European fossils, such as the Ceprano and Arago 21 fossils which have a primitive supra-orbital torus. Regarding this area, we must mention another divergence from previous studies. Indeed, we consider that Kappelman et al. (2008) over-estimated the supra-orbital torus breadth (124 mm vs 118 mm in this study) and height (min 16–max 19 vs max 13 in our own measurements), which led to overemphasis on the closeness of the Turkish fossil with the Middle Pleistocene hominins. However, we must keep in mind that only the lateral half of the supra-orbital torus on the Kocabaş fossil is preserved which means that it is not possible to know how developed the medial part was. For the width of the supra-orbital torus, our results show an intermediate position for the Kocabaş hominin between on the one hand, a gracile group composed of the Asian Homo erectus and most of the Dmanisi and early Homo specimens, and on the other hand, the more robust OH9 and Daka-Bouri BouVP2/66 skulls.

Finally, Kocabaş shares a weak frontal divergence with the one-million-year-old Homo erectus-like hominins from East Africa (OH9, Daka-Bouri BouVP2/66). This character distinguishes them from earlier hominins including the Dmanisi fossils. These results are corroborated by the cladistic analysis which considers morphological as well as metrical features. Our analysis clearly separates two clades: an Asian one (including the Dmanisi specimens) and an African one (including the Kocabaş fossil). It enables us to consider an evolutionary history for the Kocabaş fossil, filling a gap in this fossil record, which is distinct from that of the Dmanisi group and linked with the African Homo erectus known between 1.65 Ma (KNM-ER3733), 1.5 Ma (OH9) and around 1 Ma (KNM-OL 45,500, Daka-Bouri BouVP2/66, Buia UA 31). The Turkish specimen is the first representative of such a cluster outside Africa. Another group is represented by the Asian Homo erectus (Zhoukoudian and Nankin 1), which are in a same clade with the Dmanisi fossils showing an expansion toward the East, rooted at, at least, 1.77 Ma. However, a less marked postorbital constriction (#11) is a common feature between Kocabaş and the ancestor of the Asian Homo erectus group showing that they have reached a similar developmental stage. In the same way, the frontal bone divergence (#12) is less pronounced in Sangiran 17 and the group including Kocabaş, Buia UA 31, Daka-Bouri BouVP 2/66, and KNM-OL45500 which is a convergent trait due to endocranial development. But among these fossils, Kocabaş shows a convexity of the frontal wall below the temporal lines (#13) which is vertical in the others. In summary, even though the Asian Homo erectus, the one-million-old African specimens and the Turkish fossil belong to the same stage of cranial development, Kocabaş seems different to some extent. Moreover, in this specimen, the orbital roof forms, at the back, a vertical wall (#18) which suggests the orbital cavity to be short and globular, instead of long and conical as observed in general on fossil hominins and Homo sapiens. This pattern, observed on KNM-ER1813 and on Stw53 as far as this area is preserved, may be linked with the overall small size of these individuals and/or its anatomical age.

An expansion of the Kocabaş group representatives toward the west could be expected based on their geographical location. Key specimens to test this question are the fossils from the Atapuerca-Sima del Elefante site dated to 1.2 Ma (Bermúdez de Castro et al., 2011 and Carbonell et al., 2008) and Orce dated to 1.4 Ma (Toro-Moyano et al., 2013). Unfortunately, the Turkish (frontal bone) and Spanish fossils (mandible, hand phalanx, fragment of femur for Sima del Elefante, deciduous tooth for Orce) are not strictly comparable in terms of skeletal representation. More suitable are the frontal and the parietal bones discovered in the Aurora stratum (TD6.2) in the Atapuerca-Gran Dolina site, dated to 0.9 Ma (Moreno et al., 2015). Indeed, ATD6-15 consists of a large portion of the frontal bone (better preserved on its right side) but showing post-mortem distortion (Arsuaga et al., 1999). As far as they can be compared, the Turkish and Spanish fossils show some similarities. On both, we can observe the same convexity of the orbital roof (character #18) as a globular/short roof (vs conical/long one). As we have seen previously, this feature appears interesting to highlight the peculiarity of Kocabaş in comparison with the other specimens even if the RI is low (< 0.5). Maybe this similarity between the Turkish and Spanish hominins is due to the young age of both individuals, as the former is considered a young adult (Vialet et al., 2012) and the latter close to puberty (Arsuaga et al., 1999). The parietal bone (ATD6-100/168) is also from a young individual between 8 and 10 years old when Homo sapiens is used as the reference (Bruner et al., 2017). That is the reason why the comparison is limited due to considerations of growth patterns and the state of preservation (i.e. in Kocabaş, only the anterior half of the parietal is preserved).

8 Conclusion

The results of the anthropological analysis highlight that the reconstructed Kocabaş partial skull (Kocabaş 2) shows a conformation of its frontal bone which is distinct from the early Homo pattern and intermediate between that of Asian Homo erectus and African Homo erectus. Cladistically speaking, the Turkish fossil is clearly separated from the Dmanisi fossils and the Asian Homo erectus, which are grouped in the same clade. It is linked not only to the African Homo erectus bracketed between 1.49 and 1.65 Ma but also with the one-million-year-old African hominins such as the KNM-OL45500, Daka-Bouri BouVP2/66 and Buia UA 31 skulls. Due to the lack of consensus to name the latter specimens and, generally speaking, to define Homo erectus, we maintain the attribution of the Kocabaş hominin to this species in the broadest sense, while we can suggest that the Turkish fossil shares an evolutionary history more with the African groups than the Asian ones. Filling a paleoanthropological gap outside Africa, between 1.6 and 1.2 Ma, it may provide evidence of human dispersal through south-western Europe and the entire Mediterranean area.

The authors want to warmely thank Pr. Yves Coppens for his invitation in this special issue of C.R. Palevol dedicated to the “Hominins and tools. Expansions from Africa towards Eurasia”. Most of the contributions of this issue have benefited from the support of the Agence Universitaire de la Francophonie in the context of the session B52 of the XVIIe Congrès de l’Union Internationale des Sciences Préhistoriques et Protohistoriques- UISPP (2014, Burgos, Spain). This research is an emanation of the “First Homo in Turkey”, a scientific project led by A. Vialet and supported by the CNRS (PICS 2016-2018). We are grateful to Louise Byrne and Phil Glauberman for linguistic assistance. MCA is supported to the GEBİP grant (The Outstanding Young Scientist Award) given by the Turkish Academy of Sciences (TÜBA).

Abbate et al., 1998

Abbate

Albianelli

Azzaroli

Benvenuti

Tesfamariam

Bruni

Cipriani

Clarke

R.J.

Ficcarelli

Macchiarelli

Napoleone

Papini

Rook

Sagri

Medhin Tecle

Torre

Villa

A one-million-year-old Homo cranium from the Danakil (Afar) depression of Eritrea

Nature 393 1998

458–460

Alçiçek, 2014

Alçiçek

M.C.

Historique de la découverte et des recherches sur la calotte crânienne d’Homo erectus archaïque de Kocabaş, bassin de Denizli, Anatolie, Turquie

L’Anthropologie 118 2014

11–15

Alçiçek and Alçiçek, 2014

Alçiçek

M.C.

Contexte géographique et géologique du site de Kocabaş, Bassin de Denizli, Anatolie, Turquie

L’Anthropologie 118 2014

8–10

Archie, 1989

Archie

Homoplasy excess ratios: new indices for measuring levels of homoplasy in phylogenetic systematics and a critique of consistency index

Syst. Zool. 38 1989

239–252

Arsuaga et al., 1999

Arsuaga

J.L.

Martinez

Lorenzo

Gracia

Muñoz

Alonso

Gallego

The human cranial remains from Gran Dolina lower Pleistocene site (Sierra de Atapuerca, Spain)

J. Hum. Evol. 37 1999

431–457

Asfaw et al., 2002

Asfaw

Gilbert

W.E.

Beyene

Hart

W.K.

Renne

P.R.

WoldeGabriel

Vrba

E.S.

White

Remains of Homo erectus from Bouri, Middle Awash

Ethiopia. Nature 416 2002

317–320

Asfaw et al., 2008

Asfaw

Gilbert

W.H.

Richards

G.D.

Homo erectus cranial anatomy Gilbert

W.H.

Asfaw

Homo erectus Pleistocene Evidence from the Middle Awash, Ethiopia

2008

University of California Press

(458 pp.)

Aytek and Harvati, 2016

Aytek

A.I.

Harvati

The human fossil record from Turkey Harvati

Roksandic

Paleoanthropology of the Balkans and Anatolia. Human evolution and its context

2016

Springer

79–91

Bahain et al., 2017

Bahain

J.J.

Shao

Han

Sun

Voinchet

Liu

Yin

Falguères

Contribution des méthodes ESR et ESR/U-Th à la datation de quelques gisements pléistocènes de Chine

L’Anthropologie 121 3 2017

215–233

Bermúdez de Castro et al., 2011

Bermúdez de Castro

J.M.

Martinón-Torres

Gómez-Robles

Prado-Simón

Martín-Francés

Lapresa

Olejniczak

Carbonell

Early Pleistocene human mandible from Sima del Elefante (TE) cave site in Sierra de Atapuerca (Spain): a comparative morphological study

J. Hum. Evol. 61 2011

1–11

Boulbes et al., 2014

Boulbes

Mayda

Titov

V.V.

Alçiçek

M.C.

Les grands mammifères du Villafranchien supérieur des travertins du Bassin de Denizli (Sud-Ouest Anatolie, Turquie)

L’Anthropologie 118 2014

44–73

Bruner et al., 2016

Bruner

Bondioli

Coppa

Frayer

D.W.

Holloway

R.L.

Libsekal

Medin

Rook

Macchiarelli

The Endocast of the One-Million-Year-Old Human Cranium from Buia (UA 31), Danakil Eritrea

Am. J. Phys. Anthropol. 160 2016

458–468

Bruner et al., 2017

Bruner

Píšová

Martín-Francésa

Martinón-Torres

Arsuaga

J.L.

Carbonell

Bermúdez de Castro

J.M.

A human parietal fragment from the late Early Pleistocene Gran Dolina-TD6 cave site, Sierra de Atapuerca, Spain

C. R. Palevol 16 2017

71–81

Cameron et al., 2004

Cameron

Patnaik

Sahni

The phylogenetic significance of the Middle Pleistocene Narmada hominin cranium from central India

Int. J. Osteoarchaeol. 14 2004

419–447

Caparros, 1997

Caparros

Homo sapiens archaïques : un ou plusieurs (taxons) espèces ? Analyse cladistique et analyse morphométrique (Thèse) 1997

Museum National d’Histoire Naturelle de Paris

Carbonell et al., 2008

Carbonell

Bermúdez de Castro

J.M.

Parés

J.M.

Pérez-González

Cuenca-Bescós

García

Granger

D.E.

Huguet

van der Made

Martinón-Torres

Rodríguez

X.P.

Rosas

Sala

Stock

G.M.

Vallverdú

Vergés

J.M.

Allué

Benito

Burjachs

Cáceres

Canals

Díez

J.C.

Lozano

Mateos

Navazo

Rodríguez

Rosell

Arsuaga

J.L.

The first hominin of Europe

Nature 452 2008

465–470

Chevalier et al., 2015

Chevalier

Özçelik

de Lumley

M.A.

Kosem

de Lumley

Yalçinkaya

Taşkıran

The endostructural pattern of a middle Pleistocene human femoral diaphysis from the Karain E site (southern Anatolia, Turkey)

Am. J. Phys. Anthropol. 157 2015

648–658

Dinçer, 2016

Dinçer

The Lower Paleolithic in Turkey: Anatolia and Hominin dispersal out of Africa Harvati

Roksandic

Paleoanthropology of the Balkans and Anatolia. Human evolution and its context

2016

Springer

213–228

Falguères et al., 2015

Falguères

Shao

Han

Bahain

J.-J.

Richard

Perrenoud

Moigne

A.-M.

de Lumley

New ESR and U-series dating at Caune de l’Arago, France: a key-site for European Middle Pleistocene

Quat. Geochronol. 30 2015

547–553

Farris, 1989

Farris

J.S.

The retention index and the rescaled consistency index

Cladistics 5 1989

417–419

Feibel et al., 1989

Feibel

C.S.

Brown

McDougall

Stratigraphic context of fossil hominids from the Omo group deposits: northern Turkana Basin, Kenya and Ethiopia

Am. J. Phys. Anthrop. 78 1989

595–622

Feibel et al., 2009

Feibel

C.S.

Lepre

C.J.

Quinn

R.L.

Stratigraphy, correlation, and age estimates for fossils from Area 123, Koobi Fora

J. Hum. Evol. 57 2009

112–122

Gilbert et al., 2003

Gilbert

W.H.

White

T.D.

Asfaw

Homo erectus, Homo ergaster, Homo “cepranensis” and the Daka cranium

J. Hum. Evol. 45 2003

255–259

Grün, 1996

Grün

A re-analysis of electron spin resonance dating results associated with the Petralona hominid

J. Hum. Evol. 30 1996

227–241

Grün et al., 1998

Grün

Huang

Dermott

F.M.

Thorne

Stringer

C.B.

Yan

ESR and U-series analyses of teeth from the palaeoanthropological site of Hexian, Anhui Province, China

J. Hum. Evol. 34 1998

555–564

Güleç et al., 2009

Güleç

White

Kuhn

Özer

Saǧır

Yilmaz

Howell

F.C.

The Lower Pleistocene lithic assemblage from Dursunlu (Konya), central Anatolia, Turkey

Antiquity 83 2009

11–22

Guérin et al., 2015

Guérin

Frouin

Talamo

Aldeias

Bruxelles

Chiotti

Dibble

H.L.

Goldberg

Hublin

J.-J.

Jain

Lahaye

Ch.

Madelaine

Maureille

McPherron

S.P.

Mercier

Murray

A.S.

Sandgathe

Steele

T.S.

Thomsen

K.J.

Turq

A multi-method luminescence dating of the Palaeolithic sequence of La Ferrassie based on new excavations adjacent to the La Ferrassie 1 and 2 skeletons

J. Archaeol. Sci. 58 2015

147–166

Hammer et al., 2001

Hammer

Ø.

Harper

D.A.T.

Ryan

P.D.

PAST: paleontological statistics software package for education and data analysis

Paleontologia Electronica 4 1 2001

1–9

Hay, 1976

Hay

R.L.

The Geology of Olduvai Gorge Hay

R.L.

The setting of the gorge and history of the geological investigations

1976

University of California Press

Berkeley

25–28

Herries and Shaw, 2011

Herries

A.I.R.

Shaw

Palaeomagnetic analysis of the Sterkfontein palaeocave deposits: implications for the age of the hominin fossils and stone tool industries

J. Hum. Evol. 60 2011

523–539

Howell et al., 1996

Howell

F.C.

Arsebük

Kuhn

The Middle Pleistocene lithic assemblage from Yarimburgaz cave Turkey

Paléorient 22 1 1996

31–49

Hyodo et al., 2011

Hyodo

Matsu’ura

Kamishima

Kondo

Takeshita

Kitaba

Danhara

Aziz

Kurniawan

Kumai

High-resolution record of the Matuyama-Brunhes transition. Constrains the age of javanese Homo erectus in the Sangiran dome, Indonesia

Proc. Natl. Acad. Sci. USA 108 2011

19563–19568

Indriati et al., 2011

Indriati

Swisher

C.C.

III Lepre

Quinn

R.L.

Suriyanto

R.A.

Hascaryo

A.T.

Grün

Feibel

C.S.

Pobiner

Aubert

Lees

Anton

S.C.

The age of the 20 meter Solo river terrace, Java, Indonesia and the survival of Homo erectus in Asia

PLoS One 6 6 2011

e21562

Joordens et al., 2013

Joordens

J.C.A.

Dupont-Nivet

Feibel

C.S.

Spoor

Sier

M.J.

van der Lubbe

J.H.J.L.

Trine Kellberg Nielsen

T.K.

Knul

M.V.

Davies

G.R.

Vonhof

H.B.

Improved age control on early Homo fossils from the upper Burgi Member at Koobi Fora, Kenya

J. Hum. Evol. 65 2013

731–745

Kappelman et al., 2008

Kappelman

Alçiçek

M.C.

Kazanci

Schultz

Özkul

Şen

Ş.

First Homo erectus from Turkey and implications for migrations into temperate Eurasia

Am. J. Phys. Anthropol. 135 2008

110–116

Khatib et al., 2014

Khatib

Rochette

Alçiçek

M.C.

Lebatard

A.E.

Demory

Saos

Études stratigraphique, sédimentologique et paléomagnétique des travertins de Kocabaş, Bassin de Denizli, Anatolie, Turquie, contenant des restes fossiles quaternaires

L’Anthropologie 118 2014

16–33

Kluge and Farris, 1969

Kluge

A.C.

Farris

J.S.

Quantitative phyletics and the evolution of the anurans

Syst. Zool. 18 1969

1–32

Kuman and Clarke, 2000

Kuman

Clarke

R.J.

Stratigraphy, artefact industries and hominid associations for Sterkfontein, Member 5

J. Hum. Evol 38 2000

827–847

Lebatard et al., 2014a

Lebatard

A.-E.

Alçiçek

M.C.

Rochette

Khatib

Vialet

Boulbes

Bourlès

D.L.

Demory

Guipert

Mayda

Titov

V.V.

Vidal

de Lumley

Dating the Homo erectus bearing travertine from Kocabaş (Denizli, Turkey) at least 1.1 Ma

Earth Planet. Sci. Lett. 390 2014

8–18

Lebatard et al., 2014b

Lebatard

A.E.

Bourlès

D.L.

Alçiçek

M.C.

Datation des travertins de Kocabaş par la méthode des nucléides cosmogéniques ²⁶Al/¹⁰Be

L’Anthropologie 118 2014

34–43

Lordkipanidze et al., 2007

Lordkipanidze

Jashashvili

Vekua

Ponce de León

M.S.

Zollikofer

C.P.E.

Rightmire

G.P.

Pontzer

Ferring

Oms

Tappen

Bukhsianidze

Agusti

Kahlke

Kiladze

Martinez-Navarro

Mouskhelishvili

Nioradze

Rook

Postcranial evidence from early Homo from Dmanisi, Georgia

Nature 442 2007

305–310

Lumley and Tianyuan, 2008

de Lumley

Tianyuan

Le site de l’Homme de Yunxian 2008

CNRS Éditions et Editions Recherche sur les civilisations

Quyuanhekou, Quingqu, Yunxian, Province du Hubei, Chine

(587 p.)

Mc Dougall et al., 2012

Mc Dougall

Brown

F.H.

Casconcelos

P.M.

Cohen

B.E.

Thiede

D.S.

Buchanan

M.J.

New single crystal ⁴⁰Ar/³⁹Ar ages improve time scale for deposition of the Omo Group, Omo–Turkana Basin, East Africa

J. Geol. Soc. 169 2012

213–226

Maddy et al., 2015

Maddy

Schreve

Demir

Veldkamp

Wijbrans

J.R.

van Gorp

van Hinsbergen

D.J.J.

Dekkers

M.J.

Scaife

Schoorl

J.M.

Stemerdink

van der Schriek

The earliest securely-dated hominin artefact in Anatolia?

Quat. Sci. Rev. 109 2015

68–75

Martin and Saller, 1957

Martin

Saller

Lehrbuch der Anthropologie vol. 1/7 1957

Gustav Fischer Verlag, Stuttgart

(661 pp.)

Moreno et al., 2015

Moreno

Falguères

Perez-Gonzalez

Voinchet

Ghaleb

Despriée

Bahain

J.J.

Sala

Carbonell

Bermúdez de Castro

J.M.

Arsuaga

J.L.

New radiometric dates on the lowest stratigraphical section (TD1 to TD6) of Gran Dolina site (Atapuerca, Spain)

Quat. Geochronol. 30 2015

535–540

Mounier and Caparros, 2015

Mounier

Caparros

The phylogenetic status of Homo heidelbergensis – a cladistic study of Middle Pleistocene hominins

Bull. Mem. Soc. Anthropologie Paris 27 2015

110–134

Mounier et al., 2016

Mounier

Balzeau

Caparros

Grimaud-Hervé

Brain, calvarium, cladistics: a new approach to an old question, who are modern humans and Neandertals?

J. Hum. Evol. 92 2016

22–36

Otte et al., 1998

Otte

Yalçinkaya

Kozlowski

Bar-Yosef

López Bayón

Taşkıran

Long-term technical evolution and human remains in the Anatolian Palaeolithic

J. Hum. Evol. 34 1998

413–431

Potts et al., 2004

Potts

Behrensmeyer

A.K.

Deino

Ditchfield

Clark

Small Mid-Pleistocene Hominin Associated with East African Acheulean Technology

Science 305 2004

75–78

Prat, 2000

Prat

Origine et taxinomie des premiers représentants du genre Homo (Thèse) 2000

Université de Bordeaux I,

Talence, France

(Vol. 1, 587 p., Vol. 2, 100 p.)

Prat, 2004

Prat

Les premiers représentants du genre Homo, en quête d’une identité. Apports de l’étude morphologique et de l’analyse cladistique

Bull. Mem. Soc. Anthropol. Paris 16 1–2 2004

17–35

Prat, 2005

Prat

Characterising early Homo From tools to

symbols.

In: Backwell

D’errico

From hominids to modern humans

2005

Witwatersrand University Press

Johannesburg

198–228

Prat, 2017

Prat

First settlements out of Africa: Tempo and dispersal mode: review and perspectives

C. R. Palevol 2017

10.1016/j.crpv.2016.04.009

(in press)

Rendu et al., 2014

Rendu

Beauval

Crevecoeur

Bayle

Balzeau

Bismuth

Bourguignon

Delfour

Faivre

J.P.

Lacrampe-Cuyaubère

Tavorminac

Todisco

Turq

Maureille

Evidence supporting an intentional Neandertal burial at La Chapelle-aux-Saints

Proc. Natl. Acad. Sci. USA 111 2014

81–86

Roberts et al., 2009

Roberts

C.A.

Pfister

L.A.

Mays

Letter to the Editor: Was Tuberculosis present in Homo erectus in Turkey?

Am. J. Phys. Anthropol. 139 2009

442–444

Sémah et al., 2010

Sémah

A.M.

Sémah

Djubiantono

Brasseur

Landscapes and hominids’ environments: changes between the Lower and the early Middle Pleistocene in Java (Indonesia)

Quat. Int. 223-224 2010

451–454

Schwartz et al., 1994

Schwartz

J.H.

Grün

Tobias

P.V.

ESR dating studies of the Australopithecine site of Sterkfontein, South Africa

J. Hum. Evol. 26 1994

175–181

Schwartz and Tattersall, 2003

Schwartz

J.H.

Tattersall

The Human Fossil Record Craniodental Morphology of Genus Homo (Africa and Asia) Vol. 2 2003

Wiley-Liss

New York

(603 p.)

Shen et al., 2009

Shen

Gao

Granger

D.E.

Age of Zhoukoudian Homo erectus determined with ²⁶Al/¹⁰Be burial dating

Nature 458 2009

198–200

Slimak et al., 2004

Slimak

Roche

Mouralis

Buitenhuis

Balkan-Atli

Binder

Kuzucuoğlu

Grenet

Kaletepe Deresi 3 (Turquie), aspects archéologiques, chronologiques et paléontologiques d’une séquence pléistocène en Anatolie centrale

C. R. Palevol 3 2004

411–420

Slimak et al., 2008

Slimak

Kuh

S.L.

Roche

Mouralis

Buitenhuis

Balkan-Atlı

Binder

Kuzucuoğlu

Guillou

Kaletepe Deresi 3 (Turkey): archaeological evidence for early human settlement in Central Anatolia

J. Hum. Evol. 54 2008

99–111

Spoor et al., 2007

Spoor

Leakey

M.G.

Gathogo

P.N.

Brown

F.H.

Anton

S.C.

McDougall

Kiarie

Manthi

F.K.

Leakey

L.N.

Implications of new early Homo fossils from Ileret, east of Lake Turkana, Kenya

Nature 448 2007

688–691

Swofford, 1998

Swofford

D.L.

PAUP phylogenetic analysis using parsimony, version 4.01.s Computer program 1998

Illinois Natural History Survey

Champaign, Illinois, USA

Taşkıran, 2008

Taşkıran

Réflexions sur l’Acheuléen d’Anatolie

L’Anthropologie 112 2008

140–158

Taşkıran, 2017

Taşkıran

The distribution of Acheulean culture and its possible routes in Turkey

C. R. Palevol. 17 2018

10.1016/j.crpv.2016.12.005

(this issue)

Thiele, 1993

Thiele

The holy of the perfect character: the cladistics treatment of morphometric data

Cladistics 9 1993

275–304

Toro-Moyano et al., 2013

Toro-Moyano

Martínez-Navarro

Agustí

Souday

Bermúdez de Castro

J.M.

Martinón-Torres

Fajardo

Duval

Falguères

Oms

Parés

J.M.

Anadon

Julia

Garcia Aguilar

J.M.

Moigne

A.M.

Espigares

M.P.

Rosmontoya

Palmqvist

The oldest human fossil in Europe, from Orce (Spain)

J. Hum. Evol. 65 1 2013

1–9

Tryon et al., 2009

Tryon

C.A.

Logan

M.A.V.

Mouralis

Kuhn

Slimak

Balkan-Atlı

Building a tephrostratigraphic framework for the Paleolithic of Central Anatolia, Turkey

J. Archaeol. Sci. 36 3 2009

637–652

Vialet et al., 2011

Vialet

Guipert

Alçiçek

M.C.

Reconstitution 3D et étude de l’Homo erectus de Kocabaş, en Turquie

Bull. Mem. Soc. Anthropol. Paris 23 2011

S38

Vialet et al., 2012

Vialet

Guipert

Alçiçek

M.C.

Homo erectus found still further west: reconstruction of the Kocabaş cranium (Denizli, Turkey)

C. R. Palevol 11 2012

89–95

Vialet et al., 2014

Vialet

Guipert

Alçiçek

M.C.

de Lumley

M.-A.

La calotte crânienne de l’Homo erectus de Kocabaş (Bassin de Denizli, Turquie)

L’Anthropologie 118 2014

74–107

Villmoare et al., 2015

Villmoare

Kimbel

W.H.

Seyoum

Campisano

C.J.

Dimaggio

E.N.

Rowan

Braun

D.R.

Arrowsmith

J.R.

Reed

K.E.

Early Homo at 2.8 Ma from Ledi-Geraru, Afar, Ethiopia

Science 347 2015

1352–1355

Xiao et al., 2002

Xiao

Jin

Zhu

Age of the fossil Dali man in northcentral China deduced from chronostratigraphy of the loess paleosol sequence

Quat. Sci. Rev. 21 2002

2191–2198

Yokoyama et al., 2008

Yokoyama

Falguères

Sémah

Jacob

Grün

Gamma-ray spectrometric dating of late Homo erectus skulls from Ngandong and Sambungmacan, Central Java, Indonesia

J. Hum. Evol. 55 2008

274–277

Zeitoun, 2000

Zeitoun

Révision de l’espèce Homo erectus (Dubois, 1893). Utilisation des données morphologiques et métriques en cladistique, reconsidération du cas Homo erectus

Bull. Mem. Soc. Anthropologie Paris 12 ns 1–2 2000

1–200

Zeitoun et al., 2016

Zeitoun

Barriel

Widoanto

Phylogenetic analysis of the calvaria of Homo floresiensis

C.R. Palevol 15 2016

555–568

Fig. 1

Superior view of the three isolated fragments composing the Kocabaş fossil. a: the right part of the frontal bone; b: the left fronto-parietal part; c: the right parietal fragmentary bone.

Vue supérieure des trois fragments osseux isolés qui composent le fossile de Kocabaş. a : la partie droite de l’os frontal ; b : la partie fronto-pariétale gauche ; c : l’os pariétal fragmentaire droit.

Fig. 2

The virtual reconstruction of Kocabaş 2. From left to right: the 3 steps of the process: 1, putting together the two parietals; 2, mirroring the right part of the frontal to reconstruct the overall bone; 3, connecting the frontal and parietal parts; 4, to obtain the most complete reconstruction (Kocabaş 2).

La reconstitution virtuelle de Kocabaş 2. De gauche à droite : les trois étapes du processus. 1, mise en connexion des deux pariétaux ; 2, reconstitution de l’ensemble du frontal, par image miroir de la partie droite ; 3, mise en connexion des parties frontale et pariétale ; 4, pour obtenir la reconstitution la plus complète (Kocabaş 2).

Fig. 3

PCA on 4 measurements on the frontal bone (M9, M10, M43, M29.2) from 28 hominins. Axis 1 accounts for 82.5% of the total variance and axis 2 for 11.3% of the total variance. Black squares: early Homo and African Homo erectus, blue dots: Asian Homo erectus, purple crosses: middle Pleistocene European hominins and Neandertals, red star: the Kocabaş hominin.

ACP sur quatre mesures de l’os frontal (M9, M10, M43, M29,2) sur 28 homininés. L’axe 1 compte pour 82,5 % de la variance totale et l’axe 2 pour 11,3 % de la variance totale. Carrés noirs : premiers Homo et Homo erectus africains ; points bleus : Homo erectus asiatiques ; croix violettes : homininés européens du Pléistocène moyen et Néandertaliens ; étoile rouge : l’homininé de Kocabaş.

Fig. 4

PCA on 4 measurements on the frontal bone (M9, M10, M43, M29.2) from 28 fossils. Axis 1 accounts for 82.5% of the total variance and axis 3 for 4%. Black squares: early Homo and African Homo erectus, blue dots: Asian Homo erectus, purple crosses: middle Pleistocene European hominins and Neandertals, red star: the Kocabaş hominin.

ACP sur quatre mesures de l’os frontal (M9, M10, M43, M29,2) sur 28 homininés. L’axe 1 compte pour 82,5 % de la variance totale et l’axe 3 pour 4 %. Carrés noirs : premiers Homo et Homo erectus africains ; points bleus : Homo erectus asiatiques ; croix violettes : homininés européens du Pléistocène moyen et Néandertaliens, étoile rouge : l’homininé de Kocabaş.

Fig. 5

PCA on 4 frontal bone measurements (M9, M10, M43, M29.2) from 28 fossils. Axis 1 accounts for 82.5% of the total variance and axis 4 for 2,1% of the total variance. Black squares: early Homo and African Homo erectus, blue dots: Asian Homo erectus, purple crosses: middle Pleistocene European hominins and Neandertals, red star: the Kocabaş hominin.

ACP sur quatre mesures de l’os frontal (M9, M19, M43, M29,2) sur 28 homininés. L’axe 1 compte pour 82,5 % de la variance totale et l’axe 4 pour 2,1 % de la variance totale. Carrés noirs : premiers Homo et Homo erectus africains ; points bleus : Homo erectus asiatiques ; croix violettes : homininés européens du Pléistocène moyen et Néandertaliens ; étoile rouge : l’homininé de Kocabaş.

Fig. 6

Most Parsimonious Tree (MPT). Overall Retention Index RI = 0.8790; and Consistency Index CI = 0.9051.

Arbre le plus parcimonieux. Indice de rétention général RI = 0,8790 ; indice de robustesse CI = 0,9051.

Table 1

Comparison of the measurements of the two reconstitutions of the Kocabaş skull.

Comparaison des mesures des deux reconstitutions du crâne de Kocabaş.

Table 1

Reconstitution M10 M9 M43 Kocabaş 1 From Vialet et al., 2012 102 85 116 Kocabaş 2 From Vialet et al., 2014 106 88 118

Table 2

List of the hominin fossils used in this study.

Liste des homininés fossiles utilisés dans cette étude.

Table 2

Provenience N Name 2D Clad Dating Reference Africa South Africa 2 Sterfontein Sts 5 X 2.8–2.4 Schwartz et al., 1994 2.16–2.05 Herries and Shaw, 2011 Stw53 X 2.6–2 Ma Kuman and Clarke, 2000 Kenya 6 Koobi Fora KNM-ER1470 X X 2.06 Ma Joordens et al., 2013 KNM-ER1813 X X 1.78 Ma Feibel et al., 2009 KNM-ER3733 X X 1.65 Ma Mc Dougall et al., 2012 KNM-ER3883 X X 1.49–1.65 Ma Feibel et al., 1989 KNM-ER42700 X 1.55 Ma Spoor et al., 2007 KNM-OL45500 X 0.97–0.90 Ma Potts et al., 2004 Tanzania 2 Olduvai OH24 X 1.88 Ma Hay, 1976 OH9 X X 1.5–1.4 Ma Schwartz and Tattersall, 2003 Ethiopia 1 Daka-Bouri BouVP2/66 X X 1 Ma Asfaw et al., 2002 Eritrea 1 Buia UA31 X 1 Ma Abbate et al., 1998 Zambia 1 Kabwe X > 125 Ka Schwartz and Tattersall, 2003 Georgia 4 Dmanisi D2280 X X 1.77 Ma Lordkipanidze et al., 2007 D2282 X D2700 X X D3444 X Turkey 1 Denizli Basin Asia Kocabaş 2 X X 1.6–1.2 Ma Lebatard et al., 2014a and Lebatard et al., 2014b Indonesia 5 Sangiran S17 X X 0.8 Ma Sémah et al., 2010, Hyodo et al., 2011 Sambungmacan Sm3 X 0.5 Ka Yokoyama et al., 2008 Ngandong Ng 5, 10, 12 X 150–27 Ka Indriati et al., 2011 China 7 Yunxian Yunxian IIR X 936 Ka Lumley and Tianyuan, 2008 Zhoukoudian ZKD3, 11, 12 X X 780 ± 80 Ka Shen et al., 2009 Hulu cave Nankin 1 X X 641 ± 39 Ka Bahain et al., 2017 Hexian 412 ± 25 Ka Grün et al., 1998 Dali 270 Ka Xiao et al., 2002 Europe Greece 1 Petralona X 150–250 Ka Grün, 1996 France 3 Arago 21-47 X 438 ± 31 Ka Falguères et al., 2015 LCAS–La Chappelle aux saints X MIS3-MIS4 Rendu et al., 2014 LF1–La Ferrassie 1 X 54 ± 3 to 40 ± 2 Ka Guérin et al., 2015 Total 34 28 19

Table 3

Conventional measurements used in this study. M9.1 is only used for the cladistic analysis.

Mesures conventionnelles utilisées dans cette étude. M9.1 a été utilisé uniquement dans l’analyse cladistique.

Table 3

Conventional measurementsfrom Martin and Sallers 1957 Definitions M9–Minimal frontal width Between the two frontotemporals (ft) M9.1–Postorbital breadth Minimum postorbital breadth M10–Maximal frontal width On the coronal suture, between the two coronion (co) M43–Superior facial width Between the two fronto-malar-temporals (fmt) Frontal bone lengthM29.2 chord of the cerebral part of the frontal bone From supra-glabellar (sg) to bregma (b)

Table 4

Features used for the cladistic analysis.

Caractères utilisés pour l’analyse cladistique.

Table 4

Feature number Description References 1 Continuity of the postorbital groove 0: absent, 1: present but incomplete, 2: present continuously Zeitoun, 2000 (feature 4 p54) 2 Shape of the frontal superior border in superior view 0: straight, 1: convex Zeitoun, 2000 (feature 8 p56) 3 Height of the lateral part of the supra-orbital torus (c) in comparison with its central part (mid-orbital position) (b) 0: c > b, 1: c < b, 2: c = b Zeitoun, 2000 (modified from feature 12 p59) 4 Width of the temporal band (space between temporal lines) on the frontal bone 0: narrow, 1: large Zeitoun, 2000 (feature 13 p60) 5 Projection of the temporal band (space between temporal lines) on the frontal bone 0: flat, 1: prominent and strongly prominent Zeitoun, 2000 (feature 14 p60) 6 Bregmatic eminence overlapped on the parietal bone 0: absent, 1: present Zeitoun, 2000 (modified from feature 21 p62) 7 Supratrigonal depression (on the antero-medial border of the temporal line) 0: absent, 1: present Zeitoun, 2000 (feature 25 p65) 8 Height of the lateral part of the supra-orbital torus (metric) 0: weak, 1: developed, 2: strong Prat, 2000, Prat, 2004 and Prat, 2005 (modified from feature 6 p165) 9 Development of the zygomatic process of the frontal bone (which forms the lateral border of the orbit) 0: short, 1: long Feature exclusive to this study 10 Proportion of the frontal scale (M29.2/M9.1) (metric) 0: short, 1: long, 2: very long Feature exclusive to this study 11 Postorbital constriction (M9/M43) (metric) 0: strong, 1: developed, 2: less developed Feature exclusive to this study 12 Frontal bone divergence (M9.1/M10) (metric) 0: very divergent, 1 : divergent, 2: less divergent Feature exclusive to this study 13 Convexity of the external lateral wall of the frontal bone below the temporal lines 0: vertical, 1: convex Feature exclusive to this study 14 Sagittal keel on the anterior half of the sagittal suture 0: absent, 1: present Zeitoun, 2000 (modified from feature 27 p65) 15 Postcoronal depression 0: absent, 1: present Zeitoun, 2000 (feature 31 p66) 16 Discontinuity of the temporal lines at Stephanion 0: absent, 1: present Zeitoun, 2000 (feature 34 p67) 17 Temporal lines (superior and inferior) separated on the parietal bone 0: separated, 1: fused Feature exclusive to this study 18 Shape of the orbital roof 0: globular (short roof), 1: conical (long) Feature exclusive to this study

Table 5

Character states coding by Operational Taxonomic Unit.

Codage des états de caractères par unité taxinomique opérationnelle (OTU).

Table 5

Character numbers Kocabaş D2700, D2280 OH24, KNM-ER1813, KNM-ER1470, Stw53 KNM-ER37333, KNM-ER3883, OH9, KNM-ER42700 Buia UA31, Daka Bouri BOUVP2/66, KNM-OL45500 Zhoukoudian 3, 11, 12, Nankin S17 Sts 5 1 2 1 0&2 0&1&2 1&2 2 0 0 2 1 0&1 0&1 1 0&1 0&1 1 0 3 0 0&1 0&1 0&1&2 1 0&1&2 0 0 4 0 0 0&1 0&1 1 0 0 0 5 1 0 0&1 0&1 1 0 0 0 6 0 1 0 0 ? 1 1 0 7 1 0 0&1 0&1 1 0&1 0 0 8 1 0 0 1 ? 1 2 0 9 0 0 0&1 0&1 0 1 0 0 10 1 2 2 1 0 2 1 0 11 2 1 1 1 1 2 2 1 12 2 1 1 1 2 1 2 0 13 1 1 0&1 1 0 1 0 1 14 0 1 0 0&1 0 1 ? 0 15 0 1 0 0 ? 1 1 0 16 0 0 0 1 1 0 0 0 17 0 0&1 1 1 0 0&1 1 1 18 0 1 0 1 1 1 1 ?

Table 6

Measurements on the frontal bones of 28 fossil hominins.

Mesures sur les os frontaux de 28 homininés fossiles.

Table 6

Locality No. specimen Measurements M9 M10 M43 M29.2 Denizli – Turkey Kocabaş 2 88 106 118 80 Kenya – KNM ER 1813 66 88 93 64 ER 1470 70 90 106 78 ER3733 83 109 116 77 ER3883 81 108 115 87 Tanzania – Olduvai OH9 84 105 130 84 Georgia – Dmanisi D3444 70 94 105 75 D2280 74 106 114 86 D2282 67 91 105 78 D2700 67 90 97 71 Ethiopia – Daka-Bouri BouVP2/66 89 105 124 75 China – Yunxian YII Reconst. 102 136 130 74 China – Hulu cave Nankin 83 101 110 80 China – Zhoukoudian ZKD 3 81 105 109 90 ZKD 11 84 106 111 90 ZKD 12 91 110 119 97 Java – Sangiran Sang 17 96 115 119 94 China Hexian 96 116 111 76 France – Arago Arago 21 Reconst. 109.1 113.1 123 92 Greece Petralona 108 120 130 92 Zambia Kabwe 98 117 133 105 China Dali 103 121 123 101 Java – Sambungmacan Sm 3 99 110 112 93 Java – Ngandong Ng 5 100 121 116 97 Ng 10 102 121 121 99 Ng 12 103 117 123 95 France – La Chapelle-aux-Saints LCAS 106 129 118 89 France – La Ferrassie LF1 108 126 120 92

Table 7

Character states coding.

Codage des états de caractères.

Table 7

Morhological features Range Minsteps Treesteps Maxsteps CI RI RC HI 1_Continuity of the postorbital groove 2 2 3 3 0.667 0.000 0.000 0.333 2_Shape of the frontal superior border in superior view 1 1 1 1 1.000 0/0 0/0 0.000 3_Height of the lateral part of the supra-orbital torus (c) in comparison with its central part (mid-orbital position) (b) 1 1 1 1 1.000 0/0 0/0 0.000 4_Width of the temporal band (space between temporal lines) on the frontal bone 1 1 1 1 1.000 0/0 0/0 0.000 5_Projection of the temporal band (space between temporal lines) on the frontal bone 1 1 1 2 1.000 1.000 1.000 0.000 6_Bregmatic eminence overlapped on the parietal bone 1 1 1 3 1.000 1.000 1.000 0.000 7_Supratrigonal depression (on the antero-medial border of the temporal line) 1 1 1 2 1.000 1.000 1.000 0.000 8_Height of the lateral part of the supra-orbital torus (metric) 2 2 3 4 0.667 0.500 0.333 0.333 9_Development of the zygomatic process of the frontal bone (which forms the lateral border of the orbit) 1 1 1 1 1.000 0/0 0/0 0.000 10_Proportion of the frontal scale (M29.2/M9.1) 2 2 4 5 0.500 0.333 0.167 0.500 11_Postorbital constriction (M9/M43) 1 1 2 3 0.500 0.500 0.250 0.500 12_Frontal bone divergence (M9.1/M10) 2 2 3 4 0.667 0.500 0.333 0.333 13_Convexity of the external lateral wall of the frontal bone below the temporal lines 1 1 2 2 0.500 0.000 0.000 0.500 14_Sagittal keel on the anterior half of the sagittal suture 1 1 1 2 1.000 1.000 1.000 0.000 15_Postcoronal depression 1 1 1 3 1.000 1.000 1.000 0.000 16_Discontinuity of the temporal lines at Stephanion 1 1 2 2 0.500 0.000 0.000 0.500 17_Temporal lines (superior and inferior) separated on the parietal bone 1 1 1 2 1.000 1.000 1.000 0.000 18_Shape of the orbital roof 1 1 2 2 0.500 0.000 0.000 0.500