Content uploaded by Jean-Jacques Hublin
Author content
All content in this area was uploaded by Jean-Jacques Hublin on Mar 20, 2019
Content may be subject to copyright.
8 JUNE 2017 | VOL 546 | NATURE | 289
LETTER doi:10.1038/nature22336
New fossils from Jebel Irhoud, Morocco and the
pan-African origin of Homo sapiens
Jean-Jacques Hublin1,2, Abdelouahed Ben-Ncer3, Shara E. Bailey4, Sarah E. Freidline1, Simon Neubauer1, Matthew M. Skinner5,
Inga Bergmann1, Adeline Le Cabec1, Stefano Benazzi6, Katerina Harvati7 & Philipp Gunz1
Fossil evidence points to an African origin of Homo sapiens from a
group called either H. heidelbergensis or H. rhodesiensis. However,
the exact place and time of emergence of H. sapiens remain
obscure because the fossil record is scarce and the chronological
age of many key specimens remains uncertain. In particular, it is
unclear whether the present day ‘modern’ morphology rapidly
emerged approximately 200 thousand years ago (ka) among earlier
representatives of H. sapiens1 or evolved gradually over the last
400 thousand years2. Here we report newly discovered human
fossils from Jebel Irhoud, Morocco, and interpret the affinities of
the hominins from this site with other archaic and recent human
groups. We identified a mosaic of features including facial,
mandibular and dental morphology that aligns the Jebel Irhoud
material with early or recent anatomically modern humans and
more primitive neurocranial and endocranial morphology. In
combination with an age of 315 ± 34 thousand years (as determined
by thermoluminescence dating)3, this evidence makes Jebel Irhoud
the oldest and richest African Middle Stone Age hominin site that
documents early stages of the H. sapiens clade in which key features
of modern morphology were established. Furthermore, it shows
that the evolutionary processes behind the emergence of H. sapiens
involved the whole African continent.
In 1960, mining operations in the Jebel Irhoud massif 55 km south-
east of Safi, Morocco exposed a Palaeolithic site in the Pleistocene
filling of a karstic network. An almost complete skull (Irhoud 1) was
accidentally unearthed in 1961, prompting excavations that yielded
an adult braincase (Irhoud 2)4, an immature mandible (Irhoud 3)5,
an immature humeral shaft
6
, an immature ilium
7
and a fragment of a
mandible8, associated with abundant faunal remains and Levallois
stone-tool technology6. Although these human remains were all
reported to come from the bottom of the archaeological deposits, only
the precise location of the humeral shaft was recorded.
The interpretation of the Irhoud hominins has long been compli-
cated by persistent uncertainties surrounding their geological age.
They were initially considered to be around 40thousand years (kyr)
old and an African form of Neanderthals
9
. However, these affinities
have been challenged
5,10,11
and the faunal
8
and microfaunal
12
evidence
supported a Middle Pleistocene age for the site. An attempt to date one
of the hominins directly by uranium series combined with electron spin
resonance (U-series/ESR)3 suggested an age of 160 ± 16kyr (ref. 13).
Consistent with some genetic evidence
14
, fossils from Ethiopia (Omo
Kibish is considered to be as old as approximately 195kyr (ref. 15) and
Herto has been dated to approximately 160thousand years ago (ka)
16
)
are commonly regarded as the first early anatomically modern humans
(EMH). Notably, Omo Kibish 1 and the Herto specimens appear to be
more derived than the supposedly contemporaneous or even younger
Irhoud hominins. It has therefore been suggested that the archaic
features of the Irhoud fossils may indicate that north African H. sapiens
1Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6, Leipzig 04103, Germany. 2Chaire Internationale de Paléoanthropologie, Collège de France, Paris,
France. 3Institut National des Sciences de l’Archéologie et du Patrimoine, Rabat, Morocco. 4Department of Anthropology, Center for the Study of Human Origins, New York University, New York, New York
10003, USA. 5School of Anthropology and Conservation, University of Kent, Canterbury CT2 7NR, UK. 6Department of Cultural Heritage, University of Bologna, Ravenna 48121, Italy. 7Paleoanthropology,
Senckenberg Center for Human Evolution and Paleoenvironment, and DFG Center for Advanced Studies: “Words, Bones, Genes, Tools”, Eberhard Karls Universität, Tübingen, Germany.
a
b
Figure 1 | Facial reconstruction of Irhoud 10. a, b, Frontal (a) and basal
(b) views. This superimposition of Irhoud 10 (beige) and Irhoud 1 (light
blue) represents one possible alignment of the facial bones of Irhoud 10.
Nine alternative reconstructions were included in the statistical shape
analysis of the face (see Methods and Fig. 3). The maxilla, zygomatic bone
and supra-orbital area of Irhoud 10 are more robust than for Irhoud 1.
Scale bar, 20 mm.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter
reSeArCH
290 | NATURE | VOL 546 | 8 JUNE 2017
interbred with Neanderthals
17
, or that the Irhoud hominins represented
a north African, late surviving, archaic population18.
New excavations at Irhoud have enabled the recovery of in situ
archaeological material and the establishment of a precise chronology
of the deposits, which are much older than previously thought
3
. The
excavation yielded a new series of hominin remains, including an adult
skull (Irhoud 10) comprising a distorted braincase and fragments of
the face (Fig. 1), a nearly complete adult mandible (Irhoud 11) (Fig. 2),
one maxilla, several postcranial elements and abundant dental mate-
rial (Extended Data Table 1). These remains primarily come from a
single bone bed in the lower part of the archaeological deposits. This
concentration, stratigraphic observations made by previous excavators
and the anatomical similarity with earlier discoveries strongly suggest
that most, if not all, of the hominin remains from the site were accu-
mulated in a rather constrained window of time corresponding to the
formation of layer 7. This layer contains the remains of at least five
individuals (three adults, one adolescent and one immature individual,
around 7.5years old). The age of the site was redated to 315 ± 34kyr
(as determined by thermoluminescence dating)3, consistent with a
series of newly established U-series/ESR dates, which places the Irhoud
evidence in an entirely new perspective.
When compared to the large, robust and prognathic faces of the
Neanderthals or older Middle Pleistocene forms, the facial morphologies
of EMH and recent modern humans (RMH) are very distinctive.
The face is relatively short and retracted under the braincase. Facial
structures are coronally oriented and the infraorbital area is an ‘inflexion’
type, displaying curvatures along the horizontal, sagittal and coronal
profiles. This pattern, which may include some primitive retentions
19
,
strongly influences the morphology of the maxilla and zygomatic bone.
Our morphometric analysis (Fig. 3 and Methods) clearly distinguishes
archaic Middle Pleistocene humans and Neanderthals from RMH. By
contrast, all the possible reconstructions of the new facial remains of
Irhoud 10 fall well within RMH variation, as does Irhoud 1.
Another facial characteristic observed in RMH is the weakness of
their brow ridges. Some EMH from Africa and the Levant still have
protruding supraorbital structures, but they tend to be dissociated into
a medial superciliary arch and a lateral supraorbital arch. Among the
Irhoud hominins these structures are rather variable and this vari-
ability may be related to sexual dimorphism. Irhoud 1 has protruding
supraorbital structures and the arches are poorly separated. However,
in frontal view, the supraorbital buttress tends to form an inverted
V above each orbit. On Irhoud 2, the torus is less projecting and a
modern pattern can already be seen, with a clear sulcus separating the
two arches. On Irhoud 10, the preserved parts do not show projecting
supraorbital structures (Fig. 1). The new Irhoud 11 mandible is very
large overall (Fig. 2 and Extended Data Table 2). As in some EMH
from the Levant or north Africa, it has retained a vertical symphysis,
with a mental angle of 88.8° (Extended Data Fig. 1). The mandibular
body has a pattern typical of H. sapiens: its height strongly decreases
from the front to the back. This feature is also present in the immature
individual, Irhoud 3. Another modern aspect of Irhoud 11 is the rather
narrow section of the mandibular body expressed by the breadth/
height index at the level of the mental foramen (Extended Data Fig. 1).
The Irhoud mandibles also show some derived conditions in the mental
area (Extended Data Fig. 1). The symphyseal section of Irhoud 11
has a tear-shaped outline quite distinctive of H. sapiens. Although the
Irhoud mandibles lack a marked mandibular incurvation, the juve-
nile Irhoud 3 has a central keel between two depressions expanding
inferiorly into a thickened triangular eminence. This inverted T-shape,
typical of recent H. sapiens20, is incipient in the adult. Its inferior border
is somewhat distended and includes separated tubercles. Notably, this
modern pattern is still inconsistently present in Levantine EMH20. In
some aspects, Irhoud 11 is evocative of the Tabun C2 mandible, but it
is much more robust.
The Irhoud teeth are generally very large (Extended Data Tables 3, 4).
However, their dental morphology is reminiscent of EMH in several
respects. The anterior teeth do not display the expansion observed in
non-sapiens Middle Pleistocene hominins and Neanderthals
21
and the
post-canine teeth are reduced compared to older hominins. The third
maxillary molar (M3) of Irhoud 21 is already smaller than in some
EMH. The crown morphology (Extended Data Table 5 and Extended
Data Fig. 2) also aligns the Irhoud specimens most closely with
H. sapiens, rather than with non-sapiens Middle Pleistocene hominins
and Neanderthals. They do not display expanded and protruding
first upper molar (M1) hypocones, lower molar middle trigonid crests
(especially at the enamel–dentine junction (EDJ)), or a second lower
premolar (P4) with a transverse crest, uninterrupted by a longitudinal
fissure. The molars are morphologically complex and similar to the
large teeth of African EMH, possessing accessory features such as
a cusp 6, cusp 7 and protostylid on the lower molars and cusp 5 on
the upper molars. The EDJ analysis demonstrates the retention of a
non-Neanderthal primitive pattern of the P4 (Extended Data Fig. 2b).
However, derived crown shapes shared with RMH are already seen in
the upper and lower molars, grouping Irhoud 11 with EMH from north
Africa and the Levant. The lower incisor and canine roots retain a large
size, but the shape is already within the range of the modern distribu-
tion (Extended Data Fig. 3). Mandibular molar roots are cynodont, that
is, modern-human-like. This mandibular root configuration of Irhoud
11 is similar to that observed in EMH from Qafzeh. Finally, Irhoud
3 shows a pattern of eruption and a period of dental development close
to recent H. sapiens13.
Figure 2 | Irhoud 11 mandible (lateral and occlusal views). See Methods
for the reconstruction. The bi-condylar breadth of the Irhoud 11 mandible
fits the width of the corresponding areas on the Irhoud 2 skull exactly.
Scale bar, 20 mm.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter reSeArCH
8 JUNE 2017 | VOL 546 | NATURE | 291
In contrast to their modern facial morphology, the Irhoud crania
retain a primitive overall shape of the braincase and endocast, that
is, unlike those of RMH, they are elongated and not globular10,18,22.
This results in a low outline of the occipital squama, elongated tem-
poral bones and a low convexity of the parietal11. However, the frontal
squama has a vertical orientation and a marked convexity when com-
pared to archaic Middle Pleistocene specimens. These derived condi-
tions are especially well expressed on Irhoud 2 (ref. 11). A geometric
morphometric analysis (Extended Data Fig. 4) of external vault shape
distinguishes Neanderthals and archaic Middle Pleistocene forms with
their primitive neurocranial shape from RMH and Upper Palaeolithic
humans. With regards to the first principal component (PC)1, Irhoud
1 and 2 are intermediate and group together with specimens such
as Laetoli H18 and Qafzeh, as well as Upper Palaeolithic individ-
uals from Mladeč or Zhoukoudian Upper Cave. To some degree all
of these specimens retained longer and lower braincase proportions
compared to RMH. The morphometric analysis of endocranial shape
(Fig. 3b), which is not affected by cranial superstructures, shows a clear
separation between H. erectus and the Neanderthal/archaic Middle
Pleistocene cluster along PC2. The latter have evolved larger neocor
-
tices but, in contrast to RMH, without a proportional increase in the
cerebellum (Extended Data Fig. 5). EMH and the Irhoud hominins also
display elongated endocranial profiles, but are intermediate between
H. erectus and the cluster of Neanderthals/archaic Middle Pleistocene
hominins along PC2. They range in rough agreement with their
geological age along PC1, in a morphological cline ending with the
extant globular brain shapes of RMH. Notably, Omo Kibish 2 falls
between Irhoud 1 and 2. This similarity continues the question of the
contemporaneity of Omo Kibish 1 and 2, two specimens with very
different braincase morphologies23.
The Irhoud fossils currently represent, to our knowledge, the most
securely dated evidence of the early phase of H. sapiens evolution
in Africa, and they do not simply appear as intermediate between
African archaic Middle Pleistocene forms and RMH. Even approx-
imately 300 ka ago their facial morphology is almost indistinguish-
able from that of RMH, corroborating the interpretation of the
fragmentary specimen from Florisbad (South Africa) as a primitive
H. sapiens tentatively dated to 260 ka (ref. 24). Mandibular and den-
tal morphology, as well as the pattern of dental development also
align the Irhoud fossils with EMH. Notably, the endocast analysis
suggests diverging evolutionary trajectories between early H. sapiens
and African archaic Middle Pleistocene forms. This anatomical evi-
dence and the chronological proximity between these two groups
25
reinforce the hypothesis of a rapid anatomical shift or even, as sug-
gested by some26, of a chronological overlap. The Irhoud evidence
supports a complex evolutionary history of H. sapiens invol ving
the whole African continent
25,27
. Like in the Neanderthal lineage
28
,
facial morphology was established early on, and evolution in the
last 300 ka primarily affected the braincase. This occurred together
with a series of genetic changes affecting brain connectivity29,
organization and development
22
. Through accretional changes, the
Irhoud morphology is directly evolvable into that of extant humans.
Delimiting clear-cut anatomical boundaries for a ‘modern’ grade
within the H. sapiens clade thus only depends on gaps in the fossil
record30.
Online Content Methods, along with any additional Extended Data display items and
Source Data, are available in the online version of the paper; references unique to
these sections appear only in the online paper.
Received 4 May 2016; accepted 6 April 2017.
1. Stringer, C. Modern human origins: progress and prospects. Phil. Trans. R. Soc.
B 357, 563–579 (2002).
2. Bräuer, G. The origin of modern anatomy: by speciation or intraspecic
evolution? Evol. Anthropol. 17, 22–37 (2008).
3. Richter, D. et al. The age of the Jebel Irhoud (Morocco) hominins and the
origins of the Middle Stone Age. Nature http://dx.doi.org/10.1038/
nature22335 (2017).
4. Ennouchi, E. Le deuxième crâne de l’homme d’Irhoud. Annales de
Paléontologie (Vértébrés) LIV, 117–128 (1968).
5. Hublin, J.-J. & Tillier, A. M. in Aspects of Human Evolution Vol. 21 Symposia of
the Society for the study of Human Biology, Volume XXI (ed. Stringer, C. B.)
167–186 (Taylor & Francis, 1981).
6. Hublin, J.-J., Tillier, A. M. & Tixier, J. L’humerus d’enfant moustérien (Homo 4)
du Jebel Irhoud (Maroc) dans son contexte archéologique. Bull. Mem. Soc.
Anthropol. Paris 4, 115–141 (1987).
7. Tixier, J., Brugal, J.-P., Tillier, A.-M., Bruzeket, J. & Hublin, J.-J. in Actes des 1ères
Journées Nationales d’Archéologie et du Patrimoine. Préhistoire Vol. 1 149–153
(Société Marocaine d’Archéologie et du Patrimoine, 2001).
–0.10–0.05 00.05
–0.05
0
0.05
PC2: 24%
Neanderthals
Amud
Gibr
Guatt
LF 1
LaCha
Mla1
CroM 1
RMH
RMH
Ir 1
Ir 2
Omo 2
LH Qa 6
Qa 9
Sk 5
3733
3883
WT 15000
Homo
erectus
Kabwe
Petr
b
SH 5
Arago Bodo
Kabwe
Petr
Middle
Pleistocene
archaic Homo LF 1
Qa 6
Qa 9 Sk 5
–0.10 –0.05 0 0.05
–0.05
0
0.05
PC2: 16%
a
PC1: 26% PC1: 28%
Brno 3
CroM 1
Mla 1
Pre 3
ZhUC 1
ZhUC 2
Neanderthals
Amud
Gibr
Guatt
LaCha
Shan 1
Shan 5
Ir 1 Ir 10
Figure 3 | Comparative shape analysis. a, Principal component analysis
(PCA) of the facial shape. EMH (black) and RMH (blue) are well separated
from Neanderthals and archaic Middle Pleistocene hominins. Irhoud
(Ir) 1 and all nine alternative reconstructions of Irhoud 10 (pink stars
and pink 99% confidence ellipse, see Methods) fall within the RMH
variation. b, PCA of the endocranial shape. RMH (blue), Neanderthals
(red) and Homo erectus (green) are separated. Archaic Middle Pleistocene
hominins (orange) plot with Neanderthals. Irhoud 1 and 2 (pink stars)
and some EMH (black) fall outside the RMH variation. Shape differences
are visualized in Extended Data Fig. 5a. Sample compositions and
abbreviations can be found in the Methods.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter
reSeArCH
292 | NATURE | VOL 546 | 8 JUNE 2017
8. Amani, F. & Geraads, D. Le gisement moustérien du Djebel Irhoud, Maroc:
précisions sur la faune et la biochronologie, et description d’un nouveau reste
humain. Comptes Rendus à l'Académie des Sciences de Paris 316, 847–852 (1993).
9. Ennouchi, E. Un Néanderthalien: l’homme du Jebel Irhoud (Maroc).
Anthropologie 66, 279–299 (1962).
10. Stringer, C. B in Recent advances in Primatology. (eds Chivers, D. J. &
Joysey, K. A.) 395–418 (Academic, 1978).
11. Hublin, J.-J. Recent human evolution in northwestern Africa. Phil. Trans. R. Soc.
B 337, 185–191 (1992).
12. Geraads, D. et al. The rodents from the late middle Pleistocene hominid-
bearing site of J’bel Irhoud, Morocco, and their chronological and
paleoenvironmental implications. Quat. Res. 80, 552–561 (2013).
13. Smith, T. M. et al. Earliest evidence of modern human life history in North
African early Homo sapiens. Proc. Natl Acad. Sci. USA 104, 6128–6133 (2007).
14. Gonder, M. K., Mortensen, H. M., Reed, F. A., de Sousa, A. & Tishko, S. A.
Whole-mtDNA genome sequence analysis of ancient African lineages.
Mol. Biol. Evol. 24, 757–768 (2007).
15. McDougall, I., Brown, F. H. & Fleagle, J. G. Stratigraphic placement and age of
modern humans from Kibish, Ethiopia. Nature 433, 733–736 (2005).
16. White, T. D. et al. Pleistocene Homo sapiens from Middle Awash, Ethiopia.
Nature 423, 742–747 (2003).
17. Smith, F. H. in Continuity or Replacement, Controversies in Homo sapiens
evolution (eds Bräuer, G. & Smith, F. H.) 145–156 (A. A. Balkema, Zagreb,
1992).
18. Bruner, E. & Pearson, O. Neurocranial evolution in modern humans: the case
of Jebel Irhoud 1. Anthropol. Sci. 121, 31–41 (2013).
19. Bermúdez de Castro, J. M. & Martinón-Torres, M. Evolutionary interpretation of
the modern human-like facial morphology of the Atapuerca Gran Dolina-TD6
hominins. Anthropol. Sci. 122, 149–155 (2014).
20. Schwartz, J. H. & Tattersall, I. The human chin revisited: what is it and who has
it? J. Hum. Evol. 38, 367–409 (2000).
21. Trinkaus, E. Dental remains from the Shanidar adult Neanderthals. J. Hum.
Evol. 7, 369–382 (1978).
22. Gunz, P. et al. A uniquely modern human pattern of endocranial development.
Insights from a new cranial reconstruction of the Neandertal newborn from
Mezmaiskaya. J. Hum. Evol. 62, 300–313 (2012).
23. Klein, R. G. The Human Career: Human Biological and Cultural Origins 3rd edn
(Chicago Univ. Press, 2009).
24. Grün, R. et al. Direct dating of Florisbad hominid. Nature 382, 500–501
(1996).
Acknowledgements The research program at Jebel Irhoud is jointly conducted
and supported by the Moroccan Institut National des Sciences de l’Archéologie
et du Patrimoine and the Department of Human Evolution of the Max Planck
Institute for Evolutionary Anthropology. We are grateful to the many curators
and colleagues who, over the years, gave us access to recent and fossil hominin
specimens for computed tomography scanning or analysis, to E. Trinkaus
for providing comparative data and to C. Kiarie, M. Lui, C. Piot, D. Plotzki,
A. Buchenau and H. Temming for their technical assistance.
Author Contributions The study was conceived by J.-J.H., A.B.-N. and P.G.
Cranial metrical and non-metrical data were compiled and analysed by J.-J.H.,
A.B.-N., S.E.F., S.N., K.H. and P.G. Mandibular metrical and non-metrical data
were compiled and analysed by J.-J.H. and I.B. Dental metrical and non-metrical
data were compiled and analysed by S.E.B., M.M.S., A.L.C. and S.B. J.-J.H. and
P.G. wrote the manuscript with contributions from all other authors.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial
interests. Readers are welcome to comment on the online version of the
paper. Publisher’s note: Springer Nature remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.
Correspondence and requests for materials should be addressed to
J.-J.H. (hublin@eva.mpg.de) or P.G. (gunz@eva.mpg.de).
Reviewer Information Nature thanks R. G. Klein, C. Stringer and the other
anonymous reviewer(s) for their contribution to the peer review of this work.
25. Stringer, C. The Origin of Our Species. (Allen Lane, Penguin, 2011).
26. Harvati, K. et al. The later Stone Age calvaria from Iwo Eleru, Nigeria:
morphology and chronology. PLoS ONE 6, e24024 (2011).
27. Gunz, P. et al. Early modern human diversity suggests subdivided population
structure and a complex out-of-Africa scenario. Proc. Natl Acad. Sci. USA 106,
6094–6098 (2009).
28. Arsuaga, J. L. et al. Neandertal roots: cranial and chronological evidence from
Sima de los Huesos. Science 344, 1358–1363 (2014).
29. Meyer, M. et al. A high-coverage genome sequence from an archaic Denisovan
individual. Science 338, 222–226 (2012).
30. Weaver, T. D. Did a discrete event 200,000–100,000 years ago produce
modern humans? J. Hum. Evol. 63, 121–126 (2012).
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter reSeArCH
METHODS
Data reporting. No statistical methods were used to predetermine sample size.
The experiments were not randomized and the investigators were not blinded to
allocation during experiments and outcome assessment.
Computed tomography. The original fossil specimens were scanned using a BIR
ARCTIS 225/300 industrial micro-computed tomography scanner, at the Max
Planck Institute for Evolutionary Anthropology (MPI EVA), Leipzig, Germany.
The non-dental material was scanned with an isotropic voxel size ranging from
27.4 to 91.4 μ m (130 kV, 100–150 μ A, 0.25–2.0 mm brass filter, 0.144° rotation steps,
2–3 frames averaging, 360° rotation). The dental material was scanned with an
isotropic voxel size ranging from 12.8 to 32.8 μ m (130 kV, 100 μ A, 0.25–0.5 mm
brass filter, 0.144° rotation steps, 3 frames averaging, 360° rotation). Segmentation
of the micro-computed tomography volume was performed in Avizo (Visualization
Sciences Group). The comparative dental sample was scanned with an isotropic
voxel size ranging from 11.6 to 39.1 μ m at the MPI EVA on a BIR ARCTIS 225/300
micro-computed tomography scanner (130–180 kV, 100–150 μ A, 0.25–2.0 mm
brass filter, 0.096–0.144° rotation steps, 2–4 frames averaging, 360° rotation) or
on a Skyscan 1172 micro-computed tomography scanner (100 kV, 100 μ A, 0.5 mm
aluminium and 0.04 mm copper filters, 0.10–1.24° rotation steps, 360° rotation,
2–4 frames averaging). The micro-computed tomography slices were filtered using
a median filter followed by a mean-of-least-variance filter (each with a kernel size
of three) to reduce the background noise while preserving and enhancing edges31.
Virtual reconstruction. Using Avizo, nine reconstructions of the Jebel Irhoud 10
face were made on the basis of the segmented surfaces of its preserved parts consist-
ing of a left supraorbital torus, two left maxillary fragments and a nearly complete
left zygomatic bone. First, we used several RMH from diverse geographical regions
(for example, Africa, North America and Australia) and Irhoud 1 as a reference
to align the two left maxillary bones. Since a large portion of the dental arcade of
Irhoud 10 is preserved, the range of possible ‘anatomically correct’ alignments in
the palate was limited (Fig. 1b). On the basis of this maxillary alignment, each of
the subsequent reconstructions differed by several millimetres in the following
ways: broadening the palate; increasing the facial height; increasing the orbital
height; or rotating the zygomatic bones anteriorly or posteriorly in a parasagittal
direction. Additionally, we aligned one reconstruction to match the facial pro
-
portions and orientation of a ‘classic’ Neanderthal (La Ferrassie 1). In doing so,
the zygomatic bone was rotated parasagittally and moved posteriorly (> 5 mm).
Correspondingly, the brow ridge was realigned postero-superiorly by several mm,
and the maxillary bones were moved inferiorly by several mm to increase its facial
height. For each reconstruction, each bone was mirror-imaged along the mid-
sagittal plane of Irhoud 1 and then the right and left sides were merged to form one
surface model. The reconstruction of the Irhoud 11 mandible was conducted by
mirroring the left side of the mandible, which was best preserved and minimally
distorted, onto the right side, apart from the condyle, which was only preserved
on the right side and mirrored onto the left side. The left side of the mandible was
represented by three main fragments. Before mirroring, the sediment filling the
cracks between the main fragments was virtually removed, the fragments were
re-fitted and the broken crown of the left canine was reset on its root. Note that the
position of the condyles in the reconstruction is only indicative.
Shape analysis of the face, endocast and cranial vault. Geometric morphometric
methods (GMM) were used to analyse different aspects of the morphology of the
Irhoud fossils in a comparative context. To this end we digitized 3D landmarks and
sliding semilandmarks
32–34
to separately analyse the shape of the face, the endo-
cranial profile and the external vault. On the face (Fig. 3a), 3D coordinates of ana-
tomical landmarks, as well as the curve and surface semilandmarks (n = 791) were
digitized using Landmark Editor35 either on computed tomography scans (BIR
ACTIS 225/300 and Toshiba Aquilion), or surface scans (Minolta Vivid 910 and
Breuckmann optoTOP-HE) of recent modern human and fossil crania (n = 267)
following previously published protocols36. Whenever possible, measurements
were taken on scans of the original fossil; landmarks on some fossil specimens
were measured on scans of research-quality casts. Avizo was used to extract surface
files from the computed tomography scans; data from surface scanners were pre-
processed using Geomagic Studio (Geomagic Inc.) and OptoCat (Breuckmann).
On the endocast (Fig. 3b), landmarks and semilandmarks (n = 31) along the
internal midsagittal profile of the braincase were digitized on computed tomo-
graphy scans of the original specimens (n = 86) in Avizo (Visualization Sciences
Group) following the measurement protocol described in ref. 37, and converted to
2D data by projecting them onto a least squares plane in Mathematica (Wolfram
Research).
On the external vault (Extended Data Fig. 4), coordinate measurements of 97
anatomical landmarks and curve semilandmarks (along the external midsagittal
profile from glabella to inion, the coronal and lambdoid sutures, and along the
upper margin of the supraorbital torus) were captured using a Microscribe 3DX
(Immersion Corp.) portable digitizer on recent and fossil braincases (n = 296)
following the measurement protocol described in ref. 38. The points along sutures
were later resampled automatically in Mathematica to ensure the same semiland-
mark count on every specimen.
H. erectus samples include KNM ER 3733 (3733), KNM ER 3883 (3883), KNM
WT 15000. Archaic Middle Pleistocene samples include Petralona (Petr), Arago,
Sima de los Huesos H5 (SH5), Saldanha, Kabwe, Bodo. Neanderthal samples
include La Chapelle-aux-Saints 1 (LaCha), Guattari 1 (Guatt), La Ferrassie 1
(LF 1), Forbes’ Quarry 1 (Gibr), Feldhofer (Feld), La Quina 5 (LQ 5), Spy 1 and 2
(Sp 1 and Sp 2), Amud 1 (Amud), Shanidar 1 and 5 (Shan 1 and Shan 5). Primitive
H. sapiens and EMH specimens include Laetoli H18 (LH), Omo Kibish 2 (Omo 2),
Singa (Si), Qafzeh 6 and 9 (Qa 6 and Qa 9), Skhul 5 (Sk 5). Upper Palaeolithic modern
human specimens include Cro-Magnon 1 and 3 (CroM 1 and CroM 3), Mladeč
1 and 5 (Mla 1 and Mla 5), Brno 3, Předmostí 3 and 4 (Pre 3 and Pre 4), Abri
Pataud (AbP), Cioclovina (Ci), Zhoukoudian Upper Cave 1 and 2 (ZhUC 1 and
ZhUC 2). The RMH samples are composed of individuals of diverse geographical
origins (n = 232 individuals in Fig. 3a, n = 55 individuals in Fig. 3b, n = 263 indi-
viduals in Extended Data Fig. 4).
Crown outline analysis. The crown outline analysis (Extended Data Fig. 3a)
of Irhoud 10 and Irhoud 21 left M
1
follows previously described protocols
39,40
.
For Irhoud 10, computed tomography images were virtually segmented using a
semi-automatic threshold-based approach in Avizo to reconstruct a 3D digital
model of the tooth, which was then imported in Rapidform XOR2 (INUS
Technology, Inc.) to compute the cervical plane. The tooth was aligned with the
cervical plane parallel to the x–y plane of the Cartesian coordinate system and
rotated around the z axis with the lingual side parallel to the x axis. The crown out-
line corresponds to the silhouette of the oriented crown as seen in occlusal view and
projected onto the cervical plane. For Irhoud 21, an occlusal image of the crown
was taken with a Nikon D700 digital camera and a Micro-Nikkor 60 mm lens. The
tooth was oriented so that the cervical border was perpendicular to the optical axis
of the camera lens. The image was imported into the Rhino 4.0 Beta CAD environ-
ment (Robert McNeel & Associates) and aligned to the x–y plane of the Cartesian
coordinate system. The crown outline was digitized manually using the spline
function, and then oriented with the lingual side parallel to the x axis. Both crown
outlines
41
were first centred superimposing the centroids of their area according to
the M1 sample from ref. 40, but combined with 10 additional late early and Middle
Pleistocene Homo M
1
specimens (that is, Arago-31, AT-406, ATD6-11, ATD6-
69, ATD6-103, Bilzingsleben-76-530, Petralona, Steinheim, Rabat, Thomas 3).
Then, the outlines were represented by 24 pseudolandmarks obtained by equi-
angularly spaced radial vectors out of the centroid (the first radius is directed
buccally and parallel to the y axis of the Cartesian coordinate system), and scaled
to unit centroid size
39,41
. Late Early and Middle Pleistocene archaic samples include
Arago 31 (Ar 31), Atapuerca Gran Dolina 6-11, 6-69, 6-103 (ATD6-11, ATD6-69,
ATD6-103), Atapuerca Sima de los Huesos 406 (AT-406), Bilzingsleben-76-530
(Bil76-530), Petralona (Petr), Steinheim (Stein), Rabat (Rab), Thomas 3 (Tho 3).
The Neanderthal sample includes Arcy-sur-Cure 39, Cova Negra, Krapina (KDP
1, KDP 3, KDP 22, D101, D171, Max C, Max D), La Ferrassie 8, La Quina H18,
Le Fate XIII, Le Moustier 1, Monsempron 1953-1, Obi Rakhmat, Petit Puymoyen,
Roc de Marsal, Saint-Césaire 1. EMH specimens include Dar es-Soltan II-NN
and II-H6 (DSII-NN and DSII-H6), Qafzeh 10 and 15 (Qa 10 and Qa 15), Skhul
1 (Skh 1), Contrebandiers H7 (CT H7). Upper Palaeolithic modern human sam-
ples include Abri Pataud, Fontéchevade, Gough’s Cave (Magdalenian), Grotta del
Fossellone, Kostenki 15, Lagar Velho, Laugerie-Basse, La Madeleine, Les Rois 19,
Les Rois unnumbered, Mladeč (1 and 2), Peskő Barlang, St Germain (2, B6, B7),
Sunghir (2, 3), Veyrier 1. The RMH samples are composed of individuals of diverse
geographical origins (n = 80).
Molar and premolar EDJ shape analysis. Enamel and dentine tissues (Extended
Data Fig. 3b) of lower second molars and second premolars were segmented using
the 3D voxel value histogram and its distribution of greyscale values42,43. After
the segmentation the EDJ was reconstructed as a triangle-based surface model
using Avizo (using unconstrained smoothing). Small EDJ defects were corrected
digitally using the ‘fill holes’ module of Geomagic Studio. We then used Avizo to
digitize 3D landmarks and curve-semilandmarks on these EDJ surfaces
42,43
. For
the molars, anatomical landmarks were placed on the tip of the dentine horn of the
protoconid, metaconid, entoconid and hypoconid. For the premolars anatomical
landmarks were placed on protoconid and metaconid dentine horns. Moreover, we
placed a sequence of landmarks along the marginal ridge connecting the dentine
horns beginning at the top of the protoconid moving in lingual direction; the points
along this ridge curve were then later resampled to the same point count on every
specimen using Mathematica. Likewise, we digitized and resampled a cur ve along
the cemento–enamel junction as a closed curve starting and ending below the
protoconid horn and the mesiobuccal corner of the cervix. The resampled points
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter
reSeArCH
along the two ridge curves were subsequently treated as sliding curve semiland-
marks and analysed using GMM together with the four anatomical landmarks.
H. erectus specimens includes KNM-ER 992 second lower molar and second lower
premolar (M2 and P4), S1b (M2 and P4), S5, S6a. We also included the H. habilis44
specimen KNM-ER-1802 to establish trait polarity. Archaic Middle Pleistocene
samples include Mauer (M
2
and P
4
), Balanica BH-1 (Bal) and KNM-BK 67. The
Neanderthal sample includes Abri Suard S36, Combe Grenal (29, IV, VIII), El
Sidron (303, 540, 755, 763a), Krapina (53, 54, 55, 57, 59, D1, D6, D9, D35, D50, D80,
D86, D105, D107), La Quina H9, Le Moustier 1 (M2 and P4), Le Regourdou 1 (M2
and P
4
), Scladina I-4A (M
2
and P
4
), Vindija 11-39. EMH samples include Dar es-
Soltan II H4 (DS II-H4), El Harhoura (El H; M2 and P4), Irhoud 11 (Ir 11; M2 and P4),
Irhoud 3 (Ir 3; M
2
and P
4
), Qafzeh 9 (M
2
and P
4
), Qafzeh 10, Qafzeh 11 (M
2
and P
4
),
Qafzeh 15, Contrebandiers 1 (CT; M2 and P4). The RMH samples are composed
of individuals of diverse geographical origins (M2 sample, n = 8; P4 sample, n = 8).
Tooth root shape analysis. Analysis is shown in Extended Data Fig. 3. Dental
tissues (enamel, dentine and pulp) of the anterior dentition were first segmented
semiautomatically using a region growing tool, and when possible using the water-
shed principle
45
; this segmentation was edited manually to correct for cracks. Each
tooth was then virtually divided into crown and root by cutting the 3D models
at the cervical plane defined by a least-square-fit plane between the landmarks
set at the points of the greatest curvature on the labial and lingual sides of the
cement–enamel junction. Following the protocol described in ref. 46, we analysed
dental root shape: using Avizo, a landmark was digitized at the root apex and a
sequence of 3D landmark coordinates was recorded along the cement–enamel
junction. Using Mathematica, this curve was then resampled to 50 equidistant
curve-semilandmarks. The shape of the root surface, delimited by the cervical
semilandmarks and the apical landmark, was quantified using 499 surface–semi-
landmarks
46
: a mesh of 499 landmarks was digitized manually on a template spec-
imen, then warped to each specimen using a thin-plate spline interpolation and
lofted onto the segmented root surface by projecting to closest surface vertex.
These landmarks and semilandmarks were then analysed using GMM. H. erectus
is represented by KNM-WT 15000 (WT 15000). The Neanderthal samples include
Krapina (Krp53, Krp 54, Krp 55, Krp 58, Krp 59), Saint-Césaire 1 (SC), Abri
Bourgeois-Delaunay 1 (BD1), Kebara 2 and 28 (Keb 2, KMH 28). EMH samples
include Contrebandiers 1 (Tem) Qafzeh 8 and 9 (Qa 8, Qa 9) and Tabun C2
(Tab C2). Upper Palaeolithic and Mesolithic modern samples include individuals
from Oberkassel (Ob), Nahal-Oren (NO 8, NO 14), Hayonim (Ha 8, Ha 19, Ha 20),
Kebara (Keb A5) and Combe-Capelle (CC). The RMH samples include individuals
of diverse geographical origins (n = 47).
Statistical analysis. 3D landmark and semilandmark data were analysed using
GMM functions in Mathematica
34,47
. Curves and surfaces were quantified using
sliding semilandmarks on the basis of minimizing the thin-plate spline bending
energy
32
between each specimen and the sample mean shape
33,34
. Missing land-
marks or semilandmarks were estimated using a thin-plate spline interpolation on
the basis of the sample mean shape during the sliding process
48
. After sliding, all
landmarks and semilandmarks were converted to shape variables using generalized
least-squares Procrustes superimposition
49
; these data were then analysed using
PCA, and between group PCA
50
. For the M
1
crown outlines analysis, the shape
variables of the outlines were projected into the shape–space obtained from a PCA
of the comparative M1 sample. The data were processed and analysed through
software routines written in R51.
Mandibular metric data. Data are shown in (Extended Data Table 2 and Extended
Data Fig. 1c). Linear measurements were taken on 3D surface models generated
from micro-computed tomography data in Avizo. They were complemented by
measurements of the original specimens taken by E. Trinkaus (Extended Data
Fig. 1c) and by comparative data taken from the literature52–98. The African and
European archaic Middle Pleistocene samples include KNM-BK 67, KNM-BK 8518,
Sidi Abderrahmane 2, Thomas Quarry I, Thomas Quarry Gh 10717, Tighenif (1, 2,
3), Arago (I, XIII), Mauer, Montmaurin 1, Sima de los Huesos (XIX, XXI, XXVIII),
AT 1, AT 75, AT 300, AT 605, AT 607. The Asian Neanderthal specimens include
Amud 1, Chagyrskaya 6, Kebara 2, Shanidar (1, 2, 4) and Tabun C1. The European
Neanderthal specimens include Arcy II, Banyoles, El Sidrón (1, 2, 3), Guattari
(2, 3), Hortus 4, Krapina (57, 58, 59), Suard S 36, Bourgeois Delaunay 1, La
Ferrassie 1, La Quina 5, La Naulette 1, Le Regourdou 1, Saint-Césaire 1, Sima de
las Palomas (1, 6, 23, 59), Spy (1, 3), Subalyuk 1, Vindija (206, 226, 231, 250, 11.39,
11.40, 11.45), Weimar-Ehringsdorf F1009 and Zafarraya. The EMH specimens
include Dar es-Soltan II-H5, El Harhoura 1, Dire Dawa, Klasies River (KRM 13400,
KRM 14695, KRM 16424, KRM 21776, KRM 41815), Qafzeh (9, 25), Skhul (IV, V),
Tabun C2 and Contrebandiers 1. The Upper Palaeolithic and Epipalaeolithic
sample includes individuals from Abri Pataud 1, Arene Candide (2, 18), Asselar,
Barma del Caviglione, Chancelade, Cro Magnon (1, 3), Dar es-Soltan (II-H2,
II-H3), Dolni Věstonice (3, 13, 14, 15, 16), El Mirón, Grotte des Enfants 4, Hayonim
(8, 17, 19, 20, 25, 27, 29 and 29a), Isturitz (106 and 115), Le Roc (1, 2), Minat 1, Moh
Khiew, Muierii 1, Nahal Oren (6, 8, 14, 18), Nazlet Khater 2, Oase 1, Oberkassel
(1, 2), Ohalo II (1, 2), Pavlov 1, Předmostí (3, 21), Sunghir (1, 6), Villabruna 1 and
Zhoukoudian Upper Cave (101, 104, 108).
Dental metric and non-metric data. Crown metric and non-metric data
(Extended Data Fig. 3 and Extended Data Tables 3–5) were collected from casts
or originals with a few exceptions taken from the literature. The latter include:
Mumba XII (ref. 99), Eyasi100, Kapthurin101, Olduvai102, Sima de los Huesos103 and
some Sangiran metric data104. Root metric data were taken on 3D models generated
from micro-computed tomographic data105. Crown measurements were taken
using Mitituyo digital callipers. Non-metric trait expressions were scored using
the Arizona State University Dental Anthropology System
106
where applicable (for
lower dentition: P
4
lingual cusps, cusp 6, cusp 7, M groove pattern, protostylid;
for upper dentition: shovelling, tuberculum dentale, canine distal accessory ridge,
cusp 5, Carabelli’s trait, parastyle, metacone and hypocone reduction), and ref. 107
for all others. The RMH sample includes individuals from south, west and east
Africa, western and central Europe, northeast Asia, west Asia, India, Australia, New
Guinea and Andaman Islands. For root metrics (Extended Data Fig. 3) the sample
composition can be found in table 1 of ref. 105. In Extended Data Tables 3– 5,
H. erectus includes individuals from Zhoukoudian, Sangiran, West Turkana, East
Rudolf, Olduvai and Dmanisi. Middle Pleistocene African archaics (MPAf) include
individuals from Thomas Quarries, Salé, Rabat, Hoedijiespunt, Cave of Hearths,
Olduvai, Kapthurin, Eyasi, Broken Hill and Sidi Abderrahmane. Middle Pleistocene
European archaics (MPE) include individuals from Mauer, Arago, Sima de los
Huesos, Fontana Ranuccio. Neanderthal samples include individuals from Amud,
Arcy sur Cure, Chateauneuf, Combe Grenal, Cova Negra, Ehringdorf, Feldhofer,
Grotta Guattari, Grotta Taddeo, Hortus, Kalamakia, Krapina, Kebara, Kulna, La
Quina, La Fate, La Ferrassie, Le Moustier, Marillac, Melpignano, Mongaudier,
Monsempron, Monte Fenera, Malarnaud, Montmaurin, Obi-Rakhmat, Ochoz,
Pech-de-l’Azé, Petit Puymoyen, Pontnewydd, Rozhok, Regourdou, Roc-de-Marsal,
Saccopastore, Saint-Césaire, Spy, Subalyuk, Taubach, Tabun and Vindija. EMH
samples include individuals from Die Kelders, Equus Cave, Klasies River Mouth,
Sea Harvest, Mumba, Haua Fteah, Dar es-Soltan, Contrebandiers, El Harhoura,
Qafzeh and Skhul.
Data availability. The data that support the findings of this study are available
from the corresponding authors upon reasonable request.
31. Wollny, G. et al. MIA - a free and open source software for gray scale medical
image analysis. Source Code Biol. Med. 8, 20 (2013).
32. Bookstein, F. L. Landmark methods for forms without landmarks:
morphometrics of group dierences in outline shape. Med. Image Anal. 1,
225–243 (1997).
33. Gunz, P., Mitteroecker, P. & Bookstein, F. L. in Modern Morphometrics in Physical
Anthropology. (ed. Slice, D.E.) 73–98 (Kluwer Academic/Plenum Publishers,
2005).
34. Gunz, P. & Mitteroecker, P. Semilandmarks: a method for quantifying curves
and surfaces. Hystrix 24, 103–109 (2013).
35. Wiley, D. F. et al. Evolutionary Morphing. In 16th IEEE Visualization Conference
(VIS 2005) 55, 431–438 (American Journal of Physical Anthropology, 2005).
36. Freidline, S. E., Gunz, P., Harvati, K. & Hublin, J.-J. Middle Pleistocene human
facial morphology in an evolutionary and developmental context. J. Hum. Evol.
63, 723–740 (2012).
37. Neubauer, S., Gunz, P. & Hublin, J.-J. The pattern of endocranial ontogenetic
shape changes in humans. J. Anat. 215, 240–255 (2009).
38. Harvati, K., Gunz, P. & Grigorescu, D. Cioclovina (Romania): anities of an
early modern European. J. Hum. Evol. 53, 732–746 (2007).
39. Benazzi, S. et al. Early dispersal of modern humans in Europe and
implications for Neanderthal behaviour. Nature 479, 525–528 (2011).
40. Bailey, S. E., Benazzi, S. & Hublin, J.-J. Allometry, merism, and tooth shape of
the upper deciduous M2 and permanent M1. Am. J. Phys. Anthropol. 154,
104–114 (2014).
41. Benazzi, S. et al. Cervical and crown outline analysis of worn Neanderthal and
modern human lower second deciduous molars. Am. J. Phys. Anthropol. 149,
537–546 (2012).
42. Skinner, M. M., Gunz, P., Wood, B. A. & Hublin, J.-J. Enamel–dentine junction
(EDJ) morphology distinguishes the lower molars of Australopithecus africanus
and Paranthropus robustus. J. Hum. Evol. 55, 979–988 (2008).
43. Skinner, M. M., Gunz, P., Wood, B. A., Boesch, C. & Hublin, J.-J.
Discrimination of extant Pan species and subspecies using the enamel–
dentine junction morphology of lower molars. Am. J. Phys. Anthropol. 140,
234–243 (2009).
44. Spoor, F. et al. Reconstructed Homo habilis type OH 7 suggests deep-rooted
species diversity in early Homo. Nature 519, 83–86 (2015).
45. Beucher, S. & Lantuéjoul, C. in International Workshop on Image Processing:
Real-Time Edge and Motion Detection 2.1–2.12 (Rennes, France, 1979).
46. Le Cabec, A., Gunz, P., Kupczik, K., Braga, J. & Hublin, J.-J. Anterior tooth root
morphology and size in Neanderthals: taxonomic and functional implications.
J. Hum. Evol. 64, 169–193 (2013).
47. Mitteroecker, P. & Gunz, P. Advances in geometric morphometrics. Evol. Biol.
36, 235–247 (2009).
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter reSeArCH
48. Gunz, P., Mitteroecker, P., Neubauer, S., Weber, G. W. & Bookstein, F. L.
Principles for the virtual reconstruction of hominin crania. J. Hum. Evol. 57,
48–62 (2009).
49. Rohlf, F. J. & Slice, D. Extensions of the procrustes method for the optimal
superimposition of landmarks. Syst. Zool. 39, 40–59 (1990).
50. Mitteroecker, P. & Bookstein, F. Linear discrimination, ordination, and the
visualization of selection gradients in modern morphometrics. Evol. Biol. 38,
100–114 (2011).
51. R Development Core Team. R: a language and environment for statistical
computing http://www.r-project.org (R Foundation for Statistical Computing,
Vienna, Austria, 2012).
52. Arambourg, C. & Biberson, P. The fossil human remains from the Paleolithic
site of Sidi Abderrahman (Morocco). Am. J. Phys. Anthropol. 14, 467–489
(1956).
53. Sausse, F. La mandibule atlanthropienne de la carrière Thomas I (Casablanca).
Anthropologie 79, 81–112 (1975).
54. Rightmire, G. P. Comparative studies of Late Pleistocene human remains from
Klasies River Mouth, South Africa. J. Hum. Evol. 20, 131–156 (1991).
55. Stewart, T. D. in Annual Report of the Smithsonian Institution 521–533
(US Government Printing Oce, Washington, 1962).
56. Tillier, A.-M. in Le Squelette moustérien de Kébara 2 97–112 (Centre National
de la Recherche Scientique, Paris, 1991).
57. Stewart, T. D. The Neanderthal skeletal remains from Shanidar Cave, Iraq:
a summary of ndings to date. Proc. Am. Phil. Soc. 121, 121–165 (1977).
58. Trinkaus, E. The Shanidar Neanderthals (Academic, 1983).
59. Leroi-Gourhan, A. Étude des Restes Humains Fossiles provenant des Grottes
d’Arcy-sur-Cure (Masson et Cie, 1958).
60. Daura, J. et al. A Neandertal mandible from the Cova del Gegant (Sitges,
Barcelona, Spain). J. Hum. Evol. 49, 56–70 (2005).
61. Topinard, P. Les caracteres simiens de la machoire de la Naulette.
Rev. Antropol. 15, 385–431 (1886).
62. Blake, C. C. On a human jaw from the cave of La Naulette, near Dinant,
Belgium. Anthropol. Rev. 5, 294–303 (1867).
63. Leguebe, A. & Toussaint, M. La Mandibule et les Cubitus de la Naulette:
Morphologie et Morphométrie 15 (Editions du Centre National de la Recherche
Scientique, 1988).
64. Heim, J. L. Les hommes fossiles de La Ferrassie (Dordogne) et le problem de la
denition des Neandertaliens classiques. III. Squelette céphalique.
Anthropologie 78, 321–378 (1974).
65. de Lumley, M.-A. Les Néandertaliens de la grotte de l’Hortus. Etudes
Quaternaires 1, 375–385 (1972).
66. Condemi, S. et al. Possible interbreeding in late Italian Neanderthals? New
data from the Mezzena jaw (Monti Lessini, Verona, Italy). PLoS ONE 8, e59781
(2013).
67. Corrain, C. Resti scheletrici umani del ‘Riparo Mezzena’. Memorie del Museo
civico di Storia naturale di Verona 16, 97–101 (1968).
68. Walker, M. J., Lombardi, A. V., Zapata, J. & Trinkaus, E. Neandertal mandibles
from the Sima de las Palomas del Cabezo Gordo, Murcia, southeastern Spain.
Am. J. Phys. Anthropol. 142, 261–272 (2010).
69. Martin, H. Machoire humaine moustérienne trouvée dans la station de La
Quina. L’homme préhistorique 13, 3–21 (1926).
70. Martin, H. Position stratigraphique des Ossements humains recueillis dans le
Moustérien de La Quina de 1908 à 1912. Bull. Soc. Préhistorique 9, 700–709
(1912).
71. Martin, H. L’Homme fossile de la Quina. (Libraire Octave Doin, 1923).
72. Pap, I., Tillier, A. M., Arensburg, B. & Chech, M. The Subalyuk Neanderthal
remains (Hungary): a re-examination. Ann. Hist. Nat. Mus. Natl. Hung. 88,
233–270 (1996).
73. Sanchez, F. Comparative biometrical study of the Mousterian mandible from
Cueva del Boquete de Zafarraya (Málaga, Spain). Hum. Evol. 14, 125–138
(1999).
74. Vlcˇek, E. Fossile Menschenfunde von Weimar-Ehringsdorf, Weimarer
Monographien zur Ur- und Frühgeschichte Vol. 30 (Landesamt für
Archäologische Denkmalpege, 1993).
75. Condemi, S. Les néandertaliens de La Chaise: abri Bourgeois-Delaunay. Comité
des travaux historiques et scientiques (CTHS, 2001).
76. Bar-Yosef, O. & Vandermeersch, B. (eds) Le squelette moustérien de Kébara 2.
(Editions du Centre National de la Recherche Scientique, 1991).
77. Arambourg, C. & Hostetter, R. Le gisement de Ternine Vol. 1. (Masson, 1963).
78. Hooton, E.A., Hencken, H.O & Snow, Ch. E. The ancient Palestinian: Skhu¯ l V
reconstruction. 17, 5–10 (American School of Prehistoric Research, 1953).
79. Sollas, W. J. The Chancelade skull. J. R. Anthropol. Inst. 57, 89–122 (1927).
80. Martin, H. Caractères des squelettes humains quaternaires de la vallée du Roc
(Charente). Bull. Mem. Soc. Anthropol. Paris 8, 103–129 (1927).
81. Vercellotti, G., Alciati, G., Richards, M. P. & Formicola, V. The Late Upper
Paleolithic skeleton Villabruna 1 (Italy): a source of data on biology and
behavior of a 14,000 year-old hunter. J. Anthropol. Sci. 86, 143–163 (2008).
82. Formicola, V. Una mandibola umana dal deposito dell’Epigravettiano nale
delle Arene Candite (scavi 197O). Rev. Antropol. 64, 271–278 (1986).
83. Odano, A. M. & Riquet, R. Le gisement préhistorique de Dar-es-Soltane 2.
Champ de tir de El Menzeh à Rabat (Maroc). Note préliminaire. 2-Étude
anthropologique des restes post-atériens. Bull. d’Archéologie Marocaine 11,
25–63 (1978).
84. Debénath, A. Nouveaux restes humains atériens du Maroc. CR Acad. Sci. Paris
290, 851–852 (1980).
85. Crognier, E. & Dupouy-Madre, M. Les Natouens du Nahal Oren (Ouadi Fallah)
Etude anthropologique. Paéorient 2, 103–121 (1974).
86. Henke, W. Vergleichend-morphologische Kennzeichnung der Jungpaläolithiker
von Oberkassel bei Bonn. Z. Morphol. Anthropol. 75, 27–44 (1984).
87. Hershkovitz, I. et al. Ohalo II H2: a 19,000-year-old skeleton from a water-
logged site at the Sea of Galilee, Israel. Am. J. Phys. Anthropol. 96, 215–234
(1995).
88. Socaru, A., Dobos, A. & Trinkaus, E. Early modern humans from the Pestera
Muierii, Baia de Fier, Romania. Proc. Natl Acad. Sci. USA 103, 17196–17201
(2006).
89. Crevecoeur, I. Etude anthropologique des restes humains de Nazlet Khater
(Paléolithique Superieur, Egypte). PhD Thesis, Université Sciences et
Technologies Bordeaux I (2006).
90. Thoma, A. Morphology and anities of the Nazlet Khater man. J. Hum. Evol. 13,
287–296 (1984).
91. Anderson, J. E. In The Prehistory of Nubia Vol. 2 (ed. Wendorf, F.) 996–1040
(Southern Methodist Univ. Press,1968).
92. Crevecoeur, I. From the Nile to the Danube: a comparison of the Nazlet Khater
2 and Oase 1 early modern human mandibles. Anthropologie 42, 203–213
(2004).
93. Trinkaus, E. et al. An early modern human from the Pes¸ tera cu Oase, Romania.
Proc. Natl Acad. Sci. USA 100, 11231–11236 (2003).
94. Trinkaus, E. & Svoboda, J. (eds) Early Modern Human Evolution in Central Europe.
The People of Dolní Veˇ stonice and Pavlov. The Dolni Vestonice Studies Vol. 12
(Oxford Univ. Press, 2006).
95. Sládek, V., Trinkaus, E., Hillson, S. W. & Holliday, T. W. The People of the
Pavlovian. Skeletal Catalogue and Osteometrics of the Gravettian Fossil Hominids
from Dolni Vestonice and Pavlov. The Dolni Vestonice Studies Vol. 5 1–244
(BRNO, 2000).
96. Drozdová, E. The evaluation of a rediscovered fragment of human lower jaw,
No 21 from Prˇedmostí u Prˇerova. Archeologické rozhledy 53, 452–460 (2001).
97. Dutour, O. Palimpseste paléoanthropologique sur l’«Homme fossile d’Asselar»
(Sahara). Travaux du Laboratoire d’Anthropologie et de Préhistoire des Pays de la
Méditerranée Occidentale 1, 73–83 (1992).
98. Gambier, D. Vestiges humains du gisement d’Isturitz (Pyrénées—Atlantiques):
étude anthropologique et analyse des traces d’action humaine intentionelle.
Antiquités nationales 22–23, 9–26 (1990).
99. Bräuer, G. & Mehlman, M. J. Hominid molars from a middle Stone Age level at
the Mumba Rock Shelter, Tanzania. Am. J. Phys. Anthropol. 75, 69–76 (1988).
100. Protsch, R. R. R. Di Archäologischen und Anthropologischen Ergebnisse der
Kohl-Larsen-Expeditionen in Nord-Tanzania 1933–1939. (Institut für
Urgeschichte der Universität Tübingen, 1981).
101. Wood, B. A. & Van Noten, F. L. Preliminary observations on the BK 8518
mandible from Baringo, Kenya. Am. J. Phys. Anthropol. 69, 117–127 (1986).
102. Rightmire, G. P. Middle Pleistocene hominids from Olduvai Gorge, northern
Tanzania. Am. J. Phys. Anthropol. 53, 225–241 (1980).
103. Bermúdez de Castro, J. M. Dental remains from Atapuerca (Spain) I. Metrics.
J. Hum. Evol. 15, 265–287 (1986).
104. Grine, F. E. & Franzen, J. L. Fossil hominid teeth from the Sangiran Dome (Java,
Indonesia). Courier Forschungsinstitut Senckenberg 171, 75–103 (1994).
105. Le Cabec, A., Gunz, P., Kupczik, K., Braga, J. & Hublin, J.-J. Anterior tooth root
morphology and size in Neanderthals: taxonomic and functional implications.
J. Hum. Evol. 64, 169–193 (2013).
106. Turner, C. G. II, Nichol, C. R. & Scott, G. R. Advances in Dental Anthropology
(eds Kelley, M., & Larsen, C.) 13–31 (Wiley Liss, 1991).
107. Bailey, S. E. Neandertal Dental Morphology: Implications for modern human
origins. PhD thesis, Arizona State Univ. (2002).
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter
reSeArCH
Neanderthals
n=40
Middle Pleistocene
archaic Homo
n=19
Recent modern humans
n=46
Early modern humans
n=12
Mandibular Corpus Height at Mental Foramen (mm)
22 24 26 28 30 32 34 36 38 40
42
10
12
14
16
20
18
22
Mandibular Corpus Breadth at Mental Foramen (mm)
Ir 11
c)
Extended Data Figure 1 | Mandibular morphology. a, Symphyseal
section of the Irhoud 11 mandible showing the mental angle. b, Mental
area of Irhoud 11 before virtual reconstruction (top) and Irhoud 3
(bottom). Both images are surface models generated from micro-
computed tomography data. c, Bivariate plot of mandibular corpus breadth
versus height at the mental foramen. Irhoud 11 (pink star) falls within the
EMH distribution and has one of the largest corpus heights among Middle
to Late Pleistocene hominins. Values are in mm. n indicates sample size.
Data sources and sample compositions can be found in the Methods. Scale
bar, 20 mm.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter reSeArCH
Extended Data Figure 2 | Dental morphology. a, Shape–space PCA
plot of Late Early and Middle Pleistocene archaic Homo, Neanderthals and
RMH M1 crown outlines. The deformed mean crown outlines in
the four directions of the PCs are drawn at the extremity of each axis.
Sample compositions and abbreviations can be found in the Methods.
b, EDJ morphology of the M2 and P4. Top left, the PCA analysis of the
EDJ shape of the M2 places Irhoud 11 intermediate between H. erectus
and RMH (along with other north Africa fossil humans) and distinct from
Neanderthals. Surface models illustrate EDJ shape changes along PC1
(bottom left) and PC2 (top right); the former separating H. erectus from
RMH, Neanderthals and north African EMH and the latter separating
Neanderthals from RMH and north African EMH. Bottom right, a PCA
analysis of the EDJ shape of the P4 groups Irhoud 11 with modern and
fossil humans.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter
reSeArCH
Extended Data Figure 3 | Shape analysis of I2 roots. A between-group
PCA shows a complete separation between Neanderthals and a worldwide
sample of recent modern humans based on subtle shape differences.
Irhoud 11 (pink star) plots at the fringes of RMH, close to the EMH from
Contrebandiers 1 (Tem). Colour-coded Procrustes group mean shapes
are plotted in the same orientation as the I2 root surface of Irhoud 11.
Although Irhoud 11 is more similar, overall, to Neanderthals in terms
of root size, its root shape is clearly modern. The H. erectus specimen
KNM-WT 15000 and hypothetical EMH Tabun C2 have incisor root
shapes similar to Neanderthals, suggesting that roots that are labially
more convex than in RMH represent a conserved primitive condition with
limited taxonomical value. Sample compositions and abbreviations can be
found in the Methods.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter reSeArCH
Extended Data Figure 4 | Shape analysis of the external vault. a, PC
scores of PC1 and PC2 of external braincase shape in H. erectus, archaic
Middle Pleistocene Homo, geographically diverse RMH and Neanderthals.
Results are consistent with the analysis of endocranial shape (Fig. 3a).
However, several EMH and Upper Palaeolithic specimens fall outside the
RMH variation. This is probably owing to the projecting supraorbital tori
in these specimens. b, Shape changes associated with PC1 (two standard
deviations in either direction) shown as thin-plate spline deformation
grids in lateral and oblique view. PC1 captures a contrast between
elongated braincases with projecting supraorbital tori (low scores, in
black) and a more globular braincase with gracile supraorbital tori (high
scores, in red). Sample compositions and abbreviations can be found in the
Methods.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter
reSeArCH
Extended Data Figure 5 | Facial and endocranial shape differences
among Homo groups. Visualizations of GMM shape analyses in Fig. 3.
a, Average endocranial shape differences between H. erectus, recent
H. sapiens and Neanderthals. Thin-plate spline grids are exaggerated.
b, Visualization of shape changes along PC1 in Fig. 3b in frontal, lateral
and superior view; two standard deviations in either direction from the
mean shape (grey, negative; black, positive). c, Shape changes along PC2.
All recent and fossil modern humans (low scores along PC2) share smaller,
orthognathic faces, that differ from the larger, robust and prognathic faces
of the Middle Pleistocene humans and Neanderthals (high scores along
PC1). Arrow length is colour-coded (short, blue; long, red). As these
visualizations are affected by the Procrustes superimposition, we also show
grids for the maxilla and the supraorbital area. The arrow points to the
plane of the maxillary thin-plate spline (red) in the template configuration.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter reSeArCH
Extended Data Table 1 | List of hominin specimens
Starting with the 2004 excavation, specimens were given identication numbers from the project catalogue. Layer 18 of the excavation in ref. 6 corresponds to Layer 7 of the 2004–2011 excavation.
MPAf, archaic Middle Pleistocene African specimens; MPE, archaic Middle Pleistocene European specimens.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter
reSeArCH
Extended Data Table 2 | Measurements of the Irhoud 11 mandible after reconstruction
The mandibles are compared to those of ve groups of fossil hominins. Values are in mm.
x
is the mean, σ is the standard deviation, n indicates sample size. The value with a ? is an estimate. Bi-M,
Bimolar. Data sources and sample compositions can be found in the Methods
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter reSeArCH
Extended Data Table 3 | Dental measurements (upper dentition)
Teeth are identied by letters: C, canine; M, molar; P, premolar. BL, bucco-lingual width; MD, mesiodistal length. Values are in mm.
x
is the mean; minimum and maximum values are between square
brackets; σ is the standard deviation; n indicates sample size. Values in parentheses represent uncorrected measurements on worn or cracked teeth. Data sources and sample compositions are in the
Methods.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter
reSeArCH
Extended Data Table 4 | Dental measurements (lower dentition)
Teeth are identied by letters: C, canine; I, incisor; M, molar; P, premolar. BL,bucco-lingual width; MD, mesiodistal length; RL, root length. All values are in mm.
x
is the mean; minimum and maximum
values are between square brackets; σ is the standard deviation; n indicates sample size. Values in parentheses represent uncorrected measurements on worn or cracked teeth. Data sources and
sample compositions are in the Methods.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
letter reSeArCH
Extended Data Table 5 | Morphological dental trait comparison
Teeth are identied by letters: C, canine; I, incisor; M, molar; P, premolar. Numbers given are trait frequency scores at the enamel surface. Pres., present. Sample sizes are in brackets. Data sources and
sample compositions are in the Methods.
© 2017 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
CORRECTION
https://doi.org/10.1038/s41586-018-0166-3
Author Correction: New fossils
from Jebel Irhoud, Morocco and
the pan-African origin of Homo
sapiens
Jean-Jacques Hublin, Abdelouahed Ben-Ncer, Shara E. Bailey,
Sarah E. Freidline, Simon Neubauer, Matthew M. Skinner,
Inga Bergmann, Adeline Le Cabec, Stefano Benazzi,
Katerina Harvati & Philipp Gunz
Correction to: Nature https://doi.org/10.1038/nature22336,
published online 7 June 2017.
In the originally published version of this Letter, the x axis in Fig. 3a
should have been: ‘PC1: 26%’ rather than ‘PC1: 46%’, and the yaxis
should have been: ‘PC2: 16%’ rather than ‘PC2: 29%’. We also noticed
an error in the numbering of the fossils from Qafzeh: Qafzeh 27 should
be removed, and Qafzeh 26 is actually Qafzeh 25, following Tillier
(2014)1 and Schuh et al. (2017)2 and personal communication with
B. Vandermeersch and M. D. Garralda. The correct enumeration of
Qafzeh samples in the ‘Mandibular metric data’ section of the Methods
is therefore: ‘Qafzeh (9, 25)’ rather than ‘Qafzeh (9, 26, 27)’. Owing to
the removal of Qafzeh 27, the convex hull of early modern humans
changes slightly in Extended Data Fig. 1c. The sample sizes in Extended
Data Fig. 1c should have read: Middle Pleistocene archaic Homo n=19
(instead of 11), Neanderthals n=40 (instead of 41), early modern
humans n=12 (instead of 7), and recent modern humans n=46
(instead of 48). In Extended Data Table 2, the mean and standard devi-
ation of corpus height and breadth at mental foramen for early modern
humans should have been:
x
=33.15, σ=3.26 for height (rather than
x
=34.23, σ=4.57); and
x
=16.25, σ=1.28 for breadth (rather than
x
=16.04, σ=1.75). Accordingly, n=12 (rather than n= 13) for both
breadth and height. These errors have been corrected in the Letter
online (the original Extended Data Fig. 1 is shown in Supplementary
Information to this Amendment). These changes do not alter any infer-
ences drawn from the data.
Supplementary Information is available in the online version of this Amendment.
CORRECTIONS & AMENDMENTS
1. Tillier, A.-M. New middle palaeolithic hominin dental remains from Qafzeh,
Israel. Paléorient 40.1, 13–24 (2014).
2. Schuh, A. et al. La mandibule de l’adulte Qafzeh 25 (Paléolithique moyen,
basse galilée) reconstruction virtuelle 3D et analyse morphométrique.
Paléorient 43.1, 49–59 (2017).
NATURE| www.nature.com/nature
© 2018 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.