1 s2.0 S0047248417300866 Main

Journal of Human Evolution 107 (2017) 107e133
Contents lists available at ScienceDirect
Journal of Human Evolution

journal homepage: www.elsevier.com/locate/jhevol
The afnities of Homo oresiensis based on phylogenetic analyses

of cranial, dental, and postcranial characters*
Debbie Argue a, *, Colin P. Groves a, Michael S.Y. Lee b, c, William L. Jungers d, e
a
School of Archaeology and Anthropology, College of Arts and Social Sciences, The Australian National University, Canberra, 0200, Australia
b
Earth Sciences Section, South Australian Museum, North Terrace, Adelaide, SA 5000, Australia
c
School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia
d
Department of Anatomical Sciences, Stony Brook University Medical Center, Stony Brook, NY, USA
e
Association Vahatra, BP 3972, Antananarivo 101, Madagascar
a r t i c l e i n f o a b s t r a c t
Article history: Although the diminutive Homo oresiensis has been known for a decade, its phylogenetic status remains
Received 4 February 2016 highly contentious. A broad range of potential explanations for the evolution of this species has been
Accepted 23 February 2017 explored. One view is that H. oresiensis is derived from Asian Homo erectus that arrived on Flores and
Available online 21 April 2017
subsequently evolved a smaller body size, perhaps to survive the constrained resources they faced in a
new island environment. Fossil remains of H. erectus, well known from Java, have not yet been discovered
Keywords:
on Flores. The second hypothesis is that H. oresiensis is directly descended from an early Homo lineage
Homo oresiensis
with roots in Africa, such as Homo habilis; the third is that it is Homo sapiens with pathology. We use
Homo habilis
Homo erectus
parsimony and Bayesian phylogenetic methods to test these hypotheses. Our phylogenetic data build
Australopithecus upon those characters previously presented in support of these hypotheses by broadening the range of
Cladistic analysis traits to include the crania, mandibles, dentition, and postcrania of Homo and Australopithecus. The new
Flores hominins data and analyses support the hypothesis that H. oresiensis is an early Homo lineage: H. oresiensis is
sister either to H. habilis alone or to a clade consisting of at least H. habilis, H. erectus, Homo ergaster, and
H. sapiens. A close phylogenetic relationship between H. oresiensis and H. erectus or H. sapiens can be
rejected; furthermore, most of the traits separating H. oresiensis from H. sapiens are not readily
attributable to pathology (e.g., Down syndrome). The results suggest H. oresiensis is a long-surviving
relict of an early (>1.75 Ma) hominin lineage and a hitherto unknown migration out of Africa, and not
a recent derivative of either H. erectus or H. sapiens.
2017 Elsevier Ltd. All rights reserved.
1. Introduction three broad ways: 1) that H. oresiensis is a distinct species and

derived from an early lineage of Homo (e.g., Brown et al., 2004; Falk
The taxonomic status of Homo oresiensis (Brown et al., 2004) et al., 2005; Morwood et al., 2005; Argue et al., 2006, 2009; Larson
has been controversial since the species was announced. The strata et al., 2007; Tocheri et al., 2007; Brown and Maeda, 2009; Jungers
in which the remains were found are dated to between 65 5 and et al., 2009b; Brown, 2012); 2) that H. oresiensis is a distinct spe-
89 7 thousand years ago (ka), while the ulna of Liang Bua 1 (LB1) cies, derived from Asian Homo erectus (Kaifu et al., 2011, 2015; Van
is dated to between 86.9 7.9 and 71.5 4.3 ka and the ulna of LB6 den Bergh et al., 2016); and 3) that it is not a valid species, but is
is slightly younger (Sutikna et al., 2016). Fragmentary material from instead a pathological population of anatomically modern humans
Mate Menge, Flores, dated at ~700 ka has also recently been (i.e., Homo sapiens) that had genetic or metabolic disorders (e.g.,
discovered and provisionally referred to the H. oresiensis lineage Henneberg and Thorne, 2004, Jacob et al., 2006; Martin et al., 2006;
(Brumm et al., 2016; Van den Bergh et al., 2016). Bones that have Richards, 2006; Hershkovitz et al., 2007; Obendorf et al., 2008;
been placed in the H. oresiensis hypodigm have been interpreted in Oxnard et al., 2010, 2012; Henneberg et al., 2014; for reviews to the
contrary, see Groves, 2007; Aiello, 2010; Van Heteren, 2013). White
*
(2010) mused that one might also want to consider the possibility
This paper is dedicated to the memory of our late friend and colleague,
Professor Mike Morwood.
that the unique anatomy and small body size might be the unusual
* Corresponding author. result of founder effect and rapid genetic drift in a small population
E-mail address: debbie.argue@anu.edu.au (D. Argue). of modern humans colonizing Flores. However, modern humans
http://dx.doi.org/10.1016/j.jhevol.2017.02.006
0047-2484/ 2017 Elsevier Ltd. All rights reserved.
108 D. Argue et al. / Journal of Human Evolution 107 (2017) 107e133
have dwarfed repeatedly around the world and none converge on Kaifu et al. (2011) undertook a detailed morphometric and
the unique proportions and anatomy of H. oresiensis (Jungers et al., morphological comparison of the LB1 cranium with H. habilis sensu
2016). lato (comprising KNM-ER 1470, KNM-ER 1590, KNM-ER 1805,
The hypothesis that H. oresiensis represents a modern human KNM-ER 1813, KNM-ER 3732, KNM-ER 3735, KNM-ER 7330, OH 24),
population with a genetic disorder has not withstood testing H. erectus sensu stricto (Sangiran and Trinil crania), and the Dmanisi
(Argue et al., 2006, 2009; Falk et al., 2007, 2009; Jungers et al., group (putative Homo georgicus). Using 67 cranial characters that
2009b; Aiello, 2010; Brown, 2012; Baab et al., 2013, 2016; Van they observed on H. oresiensis, they tested four hypotheses: that
Heteren, 2013). The most recent iteration of these kinds of hy- H. oresiensis was descended from H. habilis, that it descended from
potheses proposes that LB1 had Down syndrome (Henneberg and the Dmanisi group, that it originated from something similar to
Thorne, 2004). Baab et al. (2016), however, found minimal early Indonesian H. erectus and subsequently dramatically dwarfed
congruence between the phenotype of LB1 and clinical descriptions in body and brain size, and that H. oresiensis was not related to any
of Down syndrome, and that H. oresiensis remains a phenotypi- of these species. Assessing the number of cranial characters that
cally unique, valid species with its roots in Plio-Pleistocene Homo exclusively supported each hypothesis, they found that only three
taxa. We also test whether the phylogenetic position of traits support a relationship between H. oresiensis and H. habilis;
H. oresiensis we obtained is attributable to Down syndrome, using no characters exclusively support a hypothesis for a relationship
similar reasoning to Dembo et al. (2015), who note that if Liang Bua between H. oresiensis and Dmanisi; and 17 characters supported,
1 had Down syndrome, it should have characters diagnostic of this or are compatible with, the hypothesis that H. oresiensis was
syndrome and characters that align it with H. sapiens. derived from early Javanese H. erectus. More recently, Kaifu et al.
Insular dwarng has been invoked to explain the morphology of (2015) reiterated this hypothesis based upon a comparative study
H. oresiensis from a H. erectus-like predecessor. The island rule for of the dentition. A re-evaluation of the endocranial value of 417 cc
insular dwarng (Foster, 1964; Alcover, 1976; Sondaar, 1977; Dayan for LB1 (Falk et al., 2005) to 426 cc by Kubo and colleagues (2013),
and Simberloff, 1998; Ko hler and Moya -Sola
, 2004) stipulates that based upon micro-CT scans, led them to also prefer the view that
body size of mammals alters when a founder population reaches an H. oresiensis was an island dwarf form of H. erectus, while
island, becomes reproductively separated from its mainland group, acknowledging the possibility that H. oresiensis may have dwarfed
and faces an environment different from those of its mainland from a H. habilis-like species.
sister species. For example, a smaller body size would be expected The island rule relates to changes in overall body mass and does
as a response to a limited food supply, and a larger size to the not necessarily predict whether specic body regions would be
absence of predation (Foster, 1964). Studies that focus on this most affected. Changes in body mass may be achieved in different
phenomenon have disputed the universality of this rule (Sondaar, ways, such as brain, gut, and kidney being traded-off against one
1977; Heaney, 1978; Wassersug et al., 1979; Lawlor, 1982; Melton, another. Nor can we predict specic changes in brain volume or
1982; Libois et al., 1993; Dayan and Simberloff, 1998; Meiri et al., limb morphology after isolation on an island, because this will
2008), but it remains the case that rapid insular changes in body depend on the selective pressures involved (Niven, 2007). This
size are very common (Millien, 2006). notwithstanding, the question of allometric scaling of H. erectus in
Van Heteren (2008) argued that the island dwarng process in relation to H. oresiensis has been investigated and this explanation
large animals typically produces paedomorphic characters, i.e. was abandoned early on by Morwood et al. (2005). Falk et al. (2005)
retention of juvenile characteristics. Van Heteren (2008) therefore showed that the brain:body size of H. oresiensis scales with aus-
tested for paedomorphic features in H. oresiensis and found that its tralopith species, even though its endocast shape is similar to
orbital index falls within the range of H. erectus children but not Chinese H. erectus (Indonesian H. erectus was not examined).
H. erectus adults: in particular she proposed that the high orbital, Gordon et al. (2008) scaled six cranial vault measurements to
dental, and brachial indices, low humeral and low tibial torsions, identify the best t between the LB1 cranium and fossil hominin
high gonial angle, and shortened lower limbs in H. oresiensis may groups, nding that H. oresiensis scaled with H. habilis, Dmanisi
be explained by paedomorphosis occurring in the dwarng process hominin D2700, and Koobi Fora cranium KNM-ER 3733 (Homo
of H. erectus, but that additional adaptations also occurred, such as ergaster), and not with H. erectus (Sangiran 17). Baab and McNulty
the relatively small brain (to lower the daily energy requirements), (2009) explored the relationship between cranial size and shape,
while the relatively short legs and broad pelvis were adaptations to and showed that the morphology of the LB1 cranium is consistent
low gear locomotion and to increase stability on uneven ground with the expected shape for a very small specimen of archaic Homo
(Van Heteren, 2012). There are numerous problems with this het- and quite distinct from the modern human sample. The high
erochronic explanation, including its generality. Paedomorphic humerofemoral index (limb proportions) for H. oresiensis has also
hominids should have relatively large crania (Godfrey and been supposed to be a consequence of its extreme small body size,
Sutherland, 1996), and LB1's skull is remarkably small relative to on the assumption that there exists a negative correlation between
any measure of body size. Moreover, no member of the genus Homo overall body size and limb length (e.g., Richards, 2006; Holliday and
at any age of development has limb proportions comparable to LB1 Franciscus, 2009). To test this hypothesis, Jungers (2009) examined
(Jungers, 2013). limb proportions of a range of modern humans, including pyg-
Lyras et al. (2009) performed a principal components analysis moids, and Australopithecus afarensis (AL 288-1), H. ergaster (KNM-
and cluster analysis of shape as represented by digitized landmarks WT 15000), and H. oresiensis. He discovered that intraspecic
of a stereolithographic replica of the LB1 cranium and of cranial covariation between limb proportions and body size in H. sapiens
casts of H. erectus Sangiran 17, Homo habilis KNM-ER 1813, and was statistically insignicant, so there is no evidence for a biological
Australopithecus africanus STS 5; they calculated Euclidean dis- rule relating body size to limb proportions; and that A. afarensis
tances among the specimens. Their results (Lyras et al., 2009: Fig. 3) and H. oresiensis fell well outside the range of all human samples,
show that H. oresiensis clusters with H. habilis on Function 1 including the small-bodied people. He tested this result in a
(contrary to their interpretation that H. oresiensis clusters with number of ways; the results led him to conclude that the unusual
H. erectus), while Function 2 and Function 3 do not discriminate limb proportions of H. oresiensis are not simply size-related al-
between H. sapiens, H. erectus, H. oresiensis, or A. africanus. In their lometries that follow a scaling model for modern humans (which in
weighted pair group method of arithmetic means (Euclidean dis- any case had little support), but can be explained as primitive re-
tances), however, H. erectus and H. oresiensis cluster. tentions (also see Jungers et al., 2016).
D. Argue et al. / Journal of Human Evolution 107 (2017) 107e133 109
Argue et al. (2009) used cladistic analyses to test alternative 2.1. Character and character state selection
hypotheses that H. oresiensis was a late-surviving member of an
early hominin lineage or that it was a dwarfed descendant of Our characters and character states are derived from:
H. erectus. Their sample included A. africanus, A. afarensis, H. habilis,
Homo rudolfensis, H. erectus s.s., H. ergaster, and a sample of Dmanisi i) publications that described characters that were potentially
hominins (putative H. georgicus). Using mostly cranial (and only a useful for inferring hominin phylogeny. This includes explicit
few mandibular and postcranial) characters, they presented two phylogenetic analyses focusing on Homo and Australopithecus,
equally parsimonious phylogenetic trees, each of which placed but also largely descriptive studies.
H. oresiensis in the early Homo section of the tree, close to H. habilis ii) direct observation of characters of H. oresiensis, some of which
and H. rudolfensis. They recognized that the hypothesis that have not been mentioned in the literature.
H. oresiensis is the result of a dwarng process in H. erectus entails
a close relationship between these two species, and they therefore Most of our character state determinations are from observa-
tested for a sister-taxon relationship between H. oresiensis and tions made on the original fossil material and scored according to
H. erectus. This alternative placement entailed a markedly longer the criteria in Table 2. Character states that are based upon metrics
phylogenetic tree than the most parsimonious topology. Their re- are derived from our measurements or from the literature as
sults, then, led them to hypothesize that H. oresiensis derived from denoted in Table 2. The data for H. georgicus and H. naledi are ob-
a more archaic lineage than the erectines and that this lineage likely tained from the literature (Table 1). In cases where we could not
existed at the same time as H. habilis and H. rudolfensis in Africa (i.e., make direct observations (for example, premolar root morphology),
2.0e1.4 million years ago); they therefore proposed that we referred to the literature; our sources for these data are also
H. oresiensis represents a very early migration out of Africa that listed in Table 2. Character states for each species are presented in
arrived on Flores at an unknown but potentially very early time. Supplementary Online Material (SOM) Appendix 1.
Dembo et al. (2015) applied an extensive suite of craniodental In all but one set of analyses, we have not excluded characters
data from the literature to test a range of phylogenetic relationships that are poorly sampled across taxa, hence there remains a
among several hominin species, including H. oresiensis. Using considerable amount of missing data. Poe and Wiens (2000) argued
Bayesian methods, they presented a phylogeny in which that characters with high levels of missing data are harmless to
H. oresiensis is sister taxon to a branch comprising H. rudolfensis, benecial; on simulated phylogenies, adding characters for which
H. ergaster, H. erectus, Homo antecessor, Homo heidelbergensis, Homo data are missing from many terminal taxa signicantly increased
neanderthalensis, and H. sapiens. H. habilis lies outside this branch phylogenetic accuracy, and they surmised that these poorly-
and forms a clade with Australopithecus sediba. sampled characters nevertheless contained extra phylogenetic in-
Most recently, Van den Bergh et al. (2016) identied mandibular formation (when combined with better-sampled characters) and
fragments and isolated teeth from Mata Menge in Flores dated at rejected the common practice of automatically excluding charac-
~700 Ma that they ascribed provisionally to the H. oresiensis ters with high levels of missing data.
lineage. This material is critically important in demonstrating the Intraspecic variation (polymorphism) was also present in
antiquity, long-term stability (e.g., in body size), and thus non- many characters. Again, as with poorly-sampled characters, char-
pathological nature, of the H. oresiensis lineage. acters with extensive intraspecic variation have been demon-
Here we present the results of new parsimony and Bayesian strated to contain potential phylogenetic information (Wiens,
phylogenetic analyses of an expanded morphological dataset that 1995). Two approaches were used here to deal with intraspecic
tests the competing hypotheses for the origin of H. oresiensis. The variation. In dataset A (Table 2), species exhibiting such variation
present study uses a range of specimens for Australopithecus and were treated as exhibiting a distinct intermediate state between
Homo species, and employs characters sampled widely from the two alternative xed states. Thus, three taxadxed for state A,
cranium, mandible, dentition, and postcranium. This has not pre- polymorphic for states A and B, and xed for state Bdwere coded
viously been attempted and affords a new database to apply to the as exhibiting states 1, 2, and 3 respectively in an ordered 3-state
problem of the phylogenetic position of H. oresiensis. character (1, 2, 3). This approach makes the assumption that a
species evolving from one xed state to another should pass
2. Materials and methods through a polymorphic intermediate stage. Dataset B (Table 2)
codes taxa with intraspecic variation as polymorphic and exhib-
Our sample comprises 133 cranial, postcranial, mandibular, and iting all observed states, which is the more widespread approach in
dental characters collected from A. afarensis, A. africanus, A. sediba, phylogenetics.
H. habilis, H. rudolfensis, H. ergaster, H. erectus (Sangiran and Trinil), Parsimony analyses were performed on datasets A and B us-
H. georgicus, Homo naledi, H. oresiensis, and H. sapiens (Table 1). ing PAUP* (Swofford, 2003). Our preliminary analyses were
Our comparative sample does not include Australopithecus garhi or performed on the full 14-taxon dataset. When testing the sta-
Australopithecus anamensis. A. garhi is relatively poorly represented bility of our tree using RogueNaRok (Aberer et al., 2013), how-
skeletally, missing 83% of data in our dataset. In comparison, most ever, we found that H. rudolfensis, H. naledi, and H. georgicus were
taxa included in our comparative sample have much less missing most unstable. By omitting the species with the least data,
data; A. afarensis is missing only 4.5%, A. africanus 7.5%, A. sediba H. rudolfensis, relative stability was restored for H. naledi and
37.6%, H. habilis 19.5%, H. oresiensis 8.3%, H. erectus 29.3%, H. georgicus. We therefore performed another set of analyses in
H. georgicus 36.8%, H. ergaster 4.5%, and H. sapiens 0.8%, while which H. rudolfensis was omitted. Results from both sets of an-
H. naledi is missing 51.1% and H. rudolfensis is missing 72.2%. alyses are presented.
Possible close phylogenetic relationships between H. oresiensis Parsimony analyses treated all character changes as equally-
and Kenyanthropus, Paranthropus, H. antecessor, Homo rhodesiensis, weighted, and changes within multistate terminal taxa were not
H. neanderthalensis, and H. heidelbergensis have not been proposed counted (when calculating tree metrics). Character states are or-
and these are not included in our analyses. dered where there is a clear linear morphocline (e.g., small, me-
Outgroups are required in parsimony and (non-clock) Bayesian dium, large; see also discussion above regarding treatment of
analyses to root the tree; we used male and female Pan, Gorilla, and polymorphism). Branch-and-bound searches, guaranteed to nd
Pongo as three successively distant outgroups (Table 1). the shortest trees, could be employed because we had only a small
Table 1
List of species included in the comparative sample, the institution in which specimens are curated, the skeletal components studied, and name of hominin. The two sets of
specimens listed for the H. habilis postcranium refer to the composition of specimens used in alternative analyses. See text for details.
Species and institution Skeletal component Specimen
Pan troglodytes Cranium, cranial part M6057 (male), M27547 (female), M32977 (female)
Australian National University, Canberra; Australian
Museum, Sydney
Gorilla Cranium, cranial part M37431 (female), M43538 (male), M36149 (male), M36148
Australian National University, Canberra; Australian (male)
Museum, Sydney
Pongo pygmaeus Cranium, cranial part M43539 (male), M3480 (female), M39946 (mature male but
Australian National University, Canberra; Australian no crest), M39947 (male), M8608 (young adult male), S1639
Museum, Sydney (male)
H. oresiensis Cranium, cranial part, postcranium LB1
Indonesian National Centre for Archaeology (ARKENAS) Mandible, lower dentition LB1, LB6
H. erectus Cranium, cranial part Grogan Wetan; Trinil 1; Sangiran 2,3,4,12,17;
Sangiran Museum and Centre for Geological Survey, Sambungmacan 1
Indonesia;
Senckenberg Museum, Frankfurt; Naturalis, Leiden
Mandible/lower dentition Sangiran 1b, 9, 22; SB 8013; BK 7095; Hanoman 13
Postcranium Femur II, III, IV, V; Kresna II femur
H. habilis Cranium, cranial part KNM-ER 1813 OH 13, OH 16, OH 24
Kenya National Museums, Nairobi, Kenya; National
Museum and House of Culture, Dar es Salaam,
Tanzania
Mandible/lower dentition OH 7, KNM-ER 1501, KNM-ER 1502, KNM-ER 1507
Postcranium 1) OH 62,
2) OH 62, OH 20, OH 34, KNM-ER 1472, KNM-ER 1481
H. ergaster Cranium, cranial part KNM-ER 730, KNM-ER 1808, KNM-ER 3883, KNM-WT
Kenya National Museums, Nairobi, Kenya 15000
Mandible/lower dentition KNM-ER 730, KNM-ER 992, KNM-ER 1808, KNM-WT 15000
Postcranium KNM-WT 15000
A. afarensis Cranium, cranial part AL 58-22, AL 162-28, AL198-1, AL 199-1, AL 200-1, AL 333-1,
National Museum of Ethiopia, Addis Ababa, Ethiopia AL 333-2, AL 333-105, AL 438-2, LH 29, BEL VP 1/1, EP 2400/
00, EP 1500/01, Garusi
Mandible/lower dentition AL 128-23, AL 145-35, AL 188-1, AL 207-13, AL 266-1, AL
277-1, AL 333-43, AL 333-45, AL 74, AL 333-w-1, AL 333-w-
12, AL 333-w-60, AL400-1A, AL 438-1, LH 2, LH 4
Postcranium AL 128-1, AL 129-1, AL 288-1 (cast), LH 21
A. africanus Cranium, cranial part STS 5, STS 17, STS 52, STS 53, STW 13, STW 73, STW 151,
University of Witwatersrand, Ditsong Museum, South STW 183, STW 505, TM 1511, TM 1512
Africa
Mandible/lower dentition STS 7, STS 36, STS 52, STW 151, STW 327, STW 384, STW
404, STW 498
Postcranium STS 14, STW 431, STW 992/100/101
A. sediba Cranium, cranial part MH 1
University of Witwatersrand, Johannesburg, South
Africa
Mandible/lower dentition MH 1, MH 2
Postcranium MH 1, MH 2
H. sapiens Crania, mandibles/dentition/postcrania 37 male and female specimens from Africa (6, labeled
Australian National University, Canberra; Australian Negroid), China (1), Europe (8), Indonesia (incl. Liang Toge;
Museum, Sydney 2), Jordan (3), Japan (10, incl. 2 Ainu), Melanesian (1), New
Ireland (1), Papua (2), Tahiti (1), Vietnam (1), Qafzeh skull II
(cast)
H. georgicus Crania, mandibles, dentition, postcranium Cranium and associated mandible:
Data obtained from the literature D2282 D211
Brauer and Schultz, 1996
Rosas and de Castro, 1998
Rightmire et al., 2006
Van Arsdale, 2006
D2700 D2735
Van Arsdale, 2006
D3444 D3900
Lordkipanidze et al., 2006
Van Arsdale, 2006
D4500 D2600
Van Arsdale, 2006
Cranium:
D2280
Gabunia et al., 2000
Table 1 (continued )
Species and institution Skeletal component Specimen
Dentition:
Rosas and de Castro, 1998; dentition from mandible D211
Rightmire et al., 2006; dentition from mandible D2735
Van Arsdale, 2006; dentition from mandibles D2600, D211,
D2735 Martino n-Torres et al., 2008; dentition from
mandibles D211, D2735, D2600
Postcranium:
Pontzer et al., 2010
H. rudolfensis Cranium KNM-ER 1470, KNM-ER 1590
Kenya National Museums, Nairobi, Kenya
H. naledi Cranium, mandible, postcranium, dentition Berger et al., 2015; online electronic scans (for characters
Data obtained from the literature 14, 18, 23, 24, 28, 42, 44, 45, 47). The specimen numbers are:
DH1, DH2, DH3, DH5.
Elen Feuerriegel., pers comm. (for characters 102, 103, 104,
10), the specimen number is U.W-1301
number of taxa (14). Branch (Bremer) support (Bremer, 1994) for The posterior probabilities (sampled frequencies) of all clades
each node on the shortest tree (or the strict consensus of shortest on the majority rule consensus were assessed, along with those
trees) was calculated using reverse constraints, in PAUP batch les alternative topological hypotheses pertaining to H. oresiensis
generated with the aid of TreeRot (Sorenson and Franzosa, 2007), (discussed above): the post-burn-in trees were loaded into PAUP
but edited to employ branch-and-bound searches. Nonparametric and the lter command used to identify the fraction of (Bayesian)
bootstrapping (Felsenstein, 1985; Hillis and Bull, 1993), based on trees consistent with each topological hypothesis.
1000 replicates, was also used to evaluate topological support un- Model testing was performed using stepping-stone analysis to
der parsimony. The bootstrap frequencies of each clade on the most estimate relative marginal likelihoods (Xie et al., 2011). These
parsimonious tree, as well as the majority-rule bootstrap suggested inclusion of the gamma parameter for across-character
consensus, were recorded. Fossil hominins vary greatly in rate heterogeneity (Bayes factor, calculated as twice the logn dif-
completeness; PAUP* deals with missing characters by assigning to ference in marginal likelihoods between the tested models, was
each taxon the character state(s) that would be most parsimonious >30 for both datasets A and B). The stepping-stone analyses used
given its phylogenetic position on the tree (as determined by non- the same settings as above and employed 40 steps each of one
missing characters). million generations (200 samples). These same stepping-stone
We also evaluated the support for alternative hypotheses of the settings were used for evaluating the position of H. oresiensis
afnities of H. oresiensis under parsimony. Further searches were proposed in Dembo et al. (2015).
performed with H. oresiensis constrained to be sister to each Analyses were also performed after exclusion of characters with
hominin species in turn, and as a sister group to a clade that in- relatively poor sampling (unknown in 4 taxa; identied with
cludes all core Homo (H. erectus, H. sapiens, H. ergaster, H. habilis). asterisk in Table 2). This reduced the character list from 133 to 78
The length of the shortest tree(s) consistent with each pairing was characters; most of the postcranial characters are thereby excluded.
ascertained. The nonparametric test of Templeton (1983), as Exclusion of poorly sampled characters has been strongly argued to
implemented in PAUP*, was also used to compare these alternative be inadvisable (see Wiens, 2000) and any threshold for exclusion is
topologies. We further examined the frequency in which all these largely arbitrary. Nevertheless, we repeated the parsimony and
alternative clades appeared in bootstrap analysis (all sampled Bayesian analyses of datasets A and B using the above exclusion
clades and their frequencies are displayed in the default PAUP threshold.
bootstrap output). Finally, analyses were also performed after recoding H. habilis by
Bayesian analyses were performed on datasets A and B using excluding equivocally-attributed postcranial remains (see above).
MrBayes 3.2.5 (Ronquist et al., 2012). It should be noted, however, With the exception of the femora of H. georgicus, H. ergaster WT-
that MrBayes treats polymorphic cells in a phylogenetic matrix as 15000, the pathological KNM-ER 1808, and H. habilis OH 62, early
totally unknown to improve computational efciency; thus, the Pleistocene femora have not been discovered in association with
results for Dataset B (variable taxa coded as polymorphic) are crania. It is, then, difcult to establish to which species isolated
potentially highly conservative. femoral bones may belong. This has resulted in each of the femora
We employed the appropriate evolutionary model selected by being referred to, or treated as, different species (H. habilis,
stepping-stone analysis (see below). All Bayesian analyses H. rudolfensis, H. erectus [Africa], and Paranthropus) in different
employed four replicate runs, each with one cold and three studies, or not included in studies. All the femora in our analyses
incrementally-heated chains (temperature 0.2); run length was have been placed in H. habilis by one or more researchers, while
50 million generations with sampling every 5000 (i.e., 10,000 some would place only one femur, OH 62, in that species. We,
samples). As all characters except for 56 and 106 were variable therefore, treat the ve femora in two ways: i) as if they are
(i.e., invariant characters were largely not sampled), the appro- H. habilis and ii) by restricting the H. habilis femur sample to OH 62,
priate correction for ascertainment bias (coding variable) was which comprises part of an individual assigned to H. habilis (see
implemented. Across the four runs, topological convergence was Johanson et al., 1987). There is no clear or unequivocal evidence
assessed using the standard deviation of split (clade) fre- that any femora belong in H. rudolfensis and we do not place any
quencies; parameter convergence was assessed using the potential femora in this species in our analyses (e.g., see McHenry, 1991;
scale reduction factor. In each analysis, the four runs converged Wood and Richmond, 2000; Haeusler and McHenry, 2004; Anto n
(s.d. split frequencies ~0.01; PSRF ~1.0) well before the burn-in of et al., 2014; Grabowski et al., 2015).
10 million generations. Post-burn-in tree samples for the four runs This reinterpretation of H. habilis resulted in changes to four
(8000 4) were then combined and the majority-rule consensus characters in this taxon for both datasets A (characters 110: 2 / 1,
obtained. 111: 1/?, 112: 2/?, 115: 4 / 3) and B (110: 1 and 2 / 1, 111: 1/?,
Table 2
List of characters and character states used in the study.a
Character Character and character states Character and character states Reference for use of character in phylogenetic
number Intraspecic variation coded Intraspecic variation coded analyses or its importance for species
as intermediate as 'polymorphic' differentiation
Dataset A Dataset B C used in previous cladistic analyses
1 Maximum cranial breadth Maximum cranial breadth Weidenreich, 1943:191

1. on parietals 1. on parietals Manzi et al., 2001:10014 (C)
2. at supramastoid region 2. at supramastoid region Kaifu et al., 2011:674
2 Facial prognathism Facial prognathism Tobias, 1991:775
1. prognathic 1. prognathic Lieberman et al., 1996:101e104 (C)
2. intermediate: prognathic and mesognathic 2. orthognathic Brown et al., 2004:1057
[3. orthognathicdnot observed] Smith and Grine, 2008:695 (C)
Kaifu et al., 2011:674
3* Inion relative to glabella Inion relative to glabella Begun et al., 1997: 398 (C)
(Ordered) (Ordered) Finarelli and Clyde, 2004:643 (C)
1. above 1. above Kimbel et al., 2004:229 (C) (low position of
2. on the same plane 2. on the same plane inion in the midsagittal arc in A. afarensis)
3. slightly below 3. slightly below
4. intermediate: slightly below and signicantly 4. signicantly below
below
5. signicantly below
4* Foramen spinosum Foramen spinosum Weidenreich, 1943:47
(Ordered) 1. contained on the suture Tobias, 1991:765
1. contained on the suture 2. contained within the greater wing of the Braga et al., 1996:387e399
2. intermediate: on the sphenosquamosal sphenoid
suture and contained within the greater wing of
the sphenoid
3. contained within the greater wing of the
sphenoid
5 Bregmatic eminence Bregmatic eminence Weidenreich, 1943:194
(Ordered) 1. present
1. present 2. absent
2. intermediate: present and absent
3. absent
6* Basicranial exion Basicranial exion Weidenreich, 1943:194
(Ordered) (Ordered) Tobias, 1991:807
1. not exed or minimal exion 1. not exed or minimal exion Skelton and McHenry, 1992:319 (C)
2. exed (137e138 ) 2. exed (137e138 ) Lieberman et al., 1996:101e104 (C)
3. strongly exed (c. 130 ) 3. strongly exed (c. 130 ) McHenry, 1996:90 (C)
Brown et al., 2004:1057
Strait and Grine, 2004:410 (C)
Smith and Grine, 2008:696 (C)
7 Superior-inferior height of nuchal region longer Superior-inferior height of nuchal region longer Weidenreich, 1943:193
than height of occipital than height of occipital Johanson et al., 1982:383
(Ordered) (Ordered) Groves, 1989:231 (C)
1. yes 1. yes Tobias, 1991:770
2. nuchal and occipital of equal length 2. nuchal and occipital of equal length Argue et al., 2006:Appendix 1 (C)
3. no 3. no Brown et al., 2004:1057
Kimbel et al., 2004:230 (C)
State 2 original researchb:
8 Posterior-anterior inclination of nuchal plane Posterior-anterior inclination of nuchal plane Lieberman et al., 1996:101e104 (C)
(Ordered) (Ordered) McHenry, 1996:90 (C)
1. steeply inclined 1. steeply inclined Kimbel et al., 2004:228 (C)
2. intermediate: steeply inclined and weakly 2. weakly inclined Strait and Grine, 2004:412 (C)
inclined 3. horizontal
3. weakly inclined
4. intermediate: weakly inclined and horizontal
9* Muscle attachments on nuchal region Muscle attachments on nuchal region Weidenreich, 1943:39
(Ordered) 1. marked Tobias, 1991:804
1. marked 2. not marked
2. intermediate: marked and not marked
3. not marked
10 Occipital torus Occipital torus Weidenreich, 1943:194
(Ordered) 1. present Groves, 1989:231 (C)
1. present 2. absent Manzi et al., 2001:10014 (C)
2. intermediate: present and absent Brown et al., 2004:1057
3. absent Argue et al., 2006:Appendix 1 (C)
Curnoe, 2010:160
11 Occipital sulcus Occipital sulcus Manzi et al., 2001:10014 (C)
(Ordered) 1. present Brown et al., 2004:1057
3. absent
12 External occipital protuberance External occipital protuberance Weidenreich, 1943:39e40
(Ordered) 1. prominent Begun et al., 1997:398 (C)
1. prominent 2. not prominent or absent Brown et al., 2004:1057
2. intermediate: prominent and not prominent Finarelli and Clyde, 2004:643 (C)
3. not prominent or absent
13* Foramen magnum Foramen magnum Tobias, 1991:762

(Ordered) 1. round Skelton and McHenry, 1992:320 (C)
1. round 2. oval McHenry, 1996:90
2. intermediate: round and oval Brown et al., 2004:1057
3. oval Strait and Grine, 2004:412 (C)
14 Mastoid process inferior projection Mastoid process inferior projection Kaifu et al., 2011:674
(Ordered) 1. projects below cranial base
1. projects below cranial base 2. does not project
2. intermediate: projects and does not project
3. does not project
15* Fissure between mastoid process and petrous Fissure between mastoid process and petrous Andrews, 1984:174 (C)
crest of tympanic crest of tympanic Tobias, 1991:766
3. absent
16* Recess between the tympanic plate and Recess between the tympanic plate and Weidenreich, 1943:201
entoglenoid pyramid entoglenoid pyramid Andrews, 1984:170 (C)
1. present 1. present Brown et al., 2004:1057
2. absent 2. absent Argue et al., 2006:Appendix 1 (C)
17 Vaginal process and crest Vaginal process and crest Groves, 1989:231 (C)
1. absent 1. absent Brown et al., 2004:1057
2. small 2. small Strait and Grine, 2004:412 (C)
3. intermediate: small and extends along 3. extends along inferior surface of tube Smith and Grine, 2008:696 (C)
inferior surface of tube Kaifu et al., 2011:674
4. extends along inferior surface of tube
18 Medio-lateral angulation of articular eminence Medio-lateral angulation of articular eminence Brown et al., 2004:1057
in basal view in basal view Argue et al., 2006:Appendix 1 (C)
(Ordered) (Ordered)
1. angled medio-anteriorly 1. angled medio-anteriorly
2. intermediate: angled medio-anteriorly and 2. parallel to coronal plane
parallel to coronal plane 3. angled medio-posteriorly
3. parallel to coronal plane
4. intermediate: angled medio-anteriorly and
angled medio posteriorly
5. angled medio-posteriorly
19 Superior-inferior angulation of anterior wall of Superior-inferior angulation of anterior wall of Kimbel and White, 1988:186
mandibular fossa mandibular fossa Aiello and Dean, 1990:103e104
(Ordered) (Ordered)
1. anterior wall horizontal 1. anterior wall horizontal
2. intermediate: horizontal and oblique 2. oblique
3. oblique 3. almost vertical
4. intermediate: oblique and almost vertical
5. almost vertical
20* Robustness of tympanic (thickness of tympanic Robustness of tympanic (thickness of tympanic Argue et al., 2006:Appendix 1 (C)
in norma lateralis, anterior edge of tympanic) in norma lateralis, anterior edge of tympanic) Weidenreich, 1943:53e54 (evolutionary
(Ordered) 1. thin (<2 mm) signicance of tympanic plate thickness)c
1. thin (<2 mm) 2. thick
2. intermediate: thin and thick
3. thick
21 Postglenoid process Postglenoid process Weidenreich, 1943:48, 201
1. postglenoid process makes up much of the 1. postglenoid process makes up much of the McHenry, 1996:90 (C)
wall of the mandibular fossa. wall of the mandibular fossa. Strait and Grine, 2004:410.
2. intermediate: postglenoid process makes up 2. postglenoid and tympanic equally form the (position and size) (C)
much of the wall of the mandibular fossa and wall Argue et al., 2006:Appendix 1 (C)
postglenoid and tympanic equally form the wall 3. the tympanic makes up most of the wall Smith and Grine, 2008:694
and the tympanic makes up most of the wall (position and size) (C)
3. postglenoid and tympanic equally form the Kaifu et al., 2011:674 (presence/absence of
wall postglenoid)
4. the tympanic makes up most of the wall
22 Auditory meatus shape Auditory meatus shape Weidenreich, 1943:55,203
(Ordered) 1. circular Lieberman et al., 1996:101e104 (C)
1. circular 2. anteroposteriorly compressed on vertical axis Martinez and Arsuaga, 1997:308
2. intermediate: circular and anteroposteriorly Kaifu et al., 2011:674
compressed on vertical axis
3. anteroposteriorly compressed on vertical axis
23 Greatest axis of auditory meatus from inferior Greatest axis of auditory meatus from inferior Weidenreich, 1943:55e56
margin to superior margin margin to superior margin Johanson et al., 1982:383
(Ordered) (Ordered)
(continued on next page)
1. slopes anteriorly 1. slopes anteriorly

2. intermediate: slopes anteriorly and vertical 2. vertical
3. vertical 3. slopes posteriorly
4. intermediate: vertical and slopes posteriorly
5. slopes posteriorly
24 Tympanic trough (a coronally oriented long Tympanic trough (a coronally oriented long Original research.d
narrow trough along the tympanic tube in basal narrow trough along the tympanic tube in basal Argue et al., 2006:Appendix 1 (C)
view) view)
3. absent
25 Supraorbital sulcus Supraorbital sulcus Rak, 1983:15
(Ordered) (Ordered) Groves, 1989:220 (C)
1. present 1. present Begun et al., 1997:397 (C)
2. intermediate: present and at plane 2. at plane Brown et al., 2004:1057
3. at plane 3. absent Finarelli and Clyde, 2004:643 (C)
4. absent Kaifu et al., 2011:674
26 Supraorbital torus Supraorbital torus Weidenreich, 1943:194
1. present 1. present Rak, 1983:15e16
2. absent 2. absent Tobias, 1991:759
Lieberman et al., 1996:101e104 (C)
Begun et al., 1997:397 (C)
Brown et al., 2004:1057
Finarelli and Clyde, 2004:643 (C)
27 Zygomaticoalveolar crest Zygomaticoalveolar crest Weidenreich, 1943:208
(Ordered) (Ordered) Rak, 1983:74
1. relatively straight 1. relatively straight Skelton and McHenry, 1992:315 (C)
2. forms an arc 2. forms an arc Kimbel et al., 2004:228 (C)
3. intermediate: forms an arc and forms an arch 3. forms an arch Argue et al., 2006:Appendix 1 (C)
4. forms an arch Harrison, 2011:165
28 Medial incursion of the superior temporal lines Medial incursion of the superior temporal lines Aiello and Dean, 1990:72e73
at coronal suture. at coronal suture.
(Ordered) (Ordered)
1. crest or nearly so 1. crest or nearly so
2. no inection 2. no inection
3. intermediate: no inection and inection 3. inection
4. inection
29 Metopic prominence Metopic prominence Weidenreich, 1943:224e225
1. present 2. absent Manzi et al., 2001:10014 (C)
2. intermediate: present and absent Argue et al., 2006:Appendix 1 (C)
3. absent Curnoe, 2010:160
30 Form of glabella in superior view Form of glabella in superior view Weidenreich, 1943:194
(Ordered) (Ordered) Rak, 1983:16
1. depressed 1. depressed Groves, 1989:230 (C)
2. intermediate: depressed and indistinct 2. indistinct Lieberman et al., 1996:101e104 (C)
3. indistinct 3. inated Argue et al., 2006:Appendix 1
4. intermediate: indistinct and inated Curnoe, 2010:160
5. inated
31 Orientation of Orientation of Rak, 1983:12e13
orbital region orbital region Brown et al., 2004:1057
(Ordered) (Ordered)
1. orientated posteriorly 1. oriented posteriorly
2. intermediate: orientated posteriorly and 2. oriented vertically
vertically 3. oriented anteriorly
3. oriented vertically
4. intermediate: oriented vertically and
anteriorly
5. oriented anteriorly
32* Size of infraorbital foramina Size of infraorbital foramina Original observatione
(Ordered) 1. large
1. large 2. small
2. intermediate: large and small
3. small
33 Location of infraorbital foramina Location of infraorbital foramina Strait and Grine, 2004:419 (C)
(Ordered) 1. medially placed (closer to nasal aperture)
1. medially placed (closer to nasal aperture) 2. mediolaterally placed under central lower
2. intermediate: medially and mediolaterally orbit margin (or nearly so)
placed
3. mediolaterally placed under central lower
orbit margin (or nearly so)
34 Anterior projection of zygomatic bone relative Anterior projection of zygomatic bone relative Skelton and McHenry, 1992:314 (C)
to piriform aperture (dishing). to piriform aperture (dishing). Lieberman et al., 1996:101e104 (C)
(Ordered) 1. posterior Strait and Grine, 2004:408 (C)
1. posterior 2. anterior Smith and Grine, 2008:694 (C)
2. intermediate: posterior and anterior Harrison, 2011:165
3. anterior
35 Canine juga Canine juga
(Ordered) 1. present Skelton and McHenry, 1992:315 (C)
1. present 2. absent Lieberman et al., 1996:101e104 (C)
2. intermediate: present and absent Brown et al., 2004:1057
3. absent
36 Anterior pillars Anterior pillars Rak, 1983:11e12
1. present 1. present Groves, 1989:219 (C)
2. intermediate: present and absent 2. absent Tobias, 1991:776
3. absent Skelton and McHenry, 1992:314 (C)
Brown et al., 2004:1057
37 Condition of the margo limitans Condition of the margo limitans Weidenreich, 1943:208
(Ordered) 1. forms a sill Tobias, 1991:776
1. forms a sill 2. smooth curve Strait and Grine, 2004:408 (C)
2. intermediate: presence of sill and smooth Smith and Grine, 2008:694 (C)
curve Argue et al., 2009:Appendix 1 (C)
3. smooth curve
38 Shape of clivus nas-oaveolaris medio-laterally Shape of clivus nas-oaveolaris medio-laterally Weidenreich, 1943:208
at midsection of the naso-alveolar clivus at midsection of the naso-alveolar clivus Rak, 1983:17e18; 74
1. convex and protrudes, or is rounded 1. convex and protrudes, or is rounded Groves, 1989:219 (C)
2. at or slightly concave 2. at or slightly concave Skelton and McHenry, 1992:314 (C)
3. has a central spine 3. has a central spine Kimbel et al., 2004:228 (C)
4. bi-convex with central hollow 4. bi-convex with central hollow Strait and Grine, 2004:408 (C)
39 Inferolateral nasal aperture margin Inferolateral nasal aperture margin Rak, 1983:16
(Ordered) 1. sharp Groves, 1989:219 (C)
1. sharp 2. blunt Harrison, 2011:178
2. intermediate: sharp and blunt
3. blunt
40 Antero-inferior slope of malar in lateral view Antero-inferior slope of malar in lateral view Weidenreich, 1943:209
(Ordered) (Ordered) Groves, 1989:219. (C)
1. slopes anteriorly 1. slopes anteriorly Lieberman et al., 1996:101e104 (C)
2. intermediate: slopes anteriorly and vertically 2. slopes vertically
3. slopes vertically 3. slopes posteriorly
4. slopes posteriorly
41 Flaring of zygomatic arch in superior view Flaring of zygomatic arch in superior view Weidenreich, 1943:45
(Ordered) 1. diverges widely from lateral edge Rak, 1983:30
1. diverges widely from lateral edge of cranium
of cranium 2. narrow divergence from lateral edge
2. intermediate: diverges widely from of cranium
lateral edge of cranium and narrow
divergence
3. narrow divergence from lateral edge of
cranium
42* Origin of masseter muscle Origin of masseter muscle Rosas, 2001:81, Table 8
(Ordered) 1. broad fossa Mounier et al., 2009:228, 230 (C)
1. broad fossa 2. narrow fossa
2. intermediate: broad and narrow
3. narrow fossa
43 Parietal keeling Parietal keeling Andrews, 1984:170 (C)
1. present 2. absent Brown et al., 2004:1057
2. intermediate: present and absent Kaifu et al., 2011:674
3. absent
44 Temporal line on the parietal Temporal line on the parietal Tobias, 1991:806
(Ordered) (Ordered) Argue et al., 2009:Appendix 1 (C)
1. no direct link with supramastoid crest 1. no direct link with supramastoid crest Kaifu et al., 2011:674
2. intermediate: no direct link with 2. continuity with supramastoid crest
supramastoid crest and continuity with 3. fuses with nuchal line
supramastoid crest
3. continuity with supramastoid crest
4. intermediate: continuity with supramastoid
crest and fuses with nuchal line
5. fuses with nuchal line
45 Obelionic depression Obelionic depression Weidenreich, 1943:199

(Ordered) 1. present Tobias, 1991:769
3. absent
46 Angular torus Angular torus Weidenreich, 1943:195
(Ordered) 1. present Andrews, 1984:170 (C)
1. present 2. absent Groves, 1989:231 (C)
2. intermediate: present and absent Tobias, 1991:769
3. absent Manzi et al., 2001:10014 (C)
Brown et al., 2004:1057
47 Divergence of tooth rows (lingual) Divergence of tooth rows (lingual) Johanson et al., 1978:6
(Ordered) (Ordered) Kimbel et al., 2004:230 (C)
1. converge posteriorly (and tooth rows are 1. converge posteriorly (and tooth rows are Harrison, 2011:179
convex laterally) convex laterally)
2. intermediate: converge posteriorly and 2. parallel
parallel 3. diverge posteriorly
3. parallel
4. intermediate: parallel and diverge posteriorly
5. diverge posteriorly
48 Palatine grooves and crests Palatine grooves and crests Original observation.f
(Ordered) 1. minimal or absent
1. minimal or absent 2. moderate or marked
2. intermediate: both states observed
3. moderate or marked
49 Depth of posterior palate Depth of posterior palate Johanson et al., 1978:6
(Ordered) (Ordered) Rak, 1983:28e29, 74
1. shallow posteriorly 1. shallow posteriorly Groves, 1989:231 (C)
2. intermediate: both states observed 2. intermediate: both states observed Skelton and McHenry, 1992:318 (C)
3. deep posteriorly 3. deep posteriorly Lieberman et al., 1996:101e104 (C)
Finarelli and Clyde, 2004:644 (posterior palate
depth) (C)
Kimbel et al., 2004:229 (C)
Harrison, 2011:168, 178
50* Incisive fossa Incisive fossa Original observation.g
(Ordered) 1. large
1. large 2. small
2. intermediate: large and small
3. small
51 Symphyseal region Symphyseal region Skelton and McHenry, 1992:319 (C)
(Ordered) (Ordered) Lieberman et al., 1996:104 (C)
1. no mental protuberance and retreats 1. no mental protuberance and retreats McHenry, 1996:90 (C)
2. no mental protuberance and vertical 2. no mental protuberance and vertical Schwartz and Tattersall, 2002:367e409
3. slight symphyseal ridge and retreats 3. slight symphyseal ridge and retreats Brown et al., 2004:1058
4. mental protuberance with keel (chin) present 4. mental protuberance with keel (chin) present Argue et al., 2009:Appendix 1 (C)
Brown and Maeda, 2009:576
Harrison, 2011:178
52 Ramus obscures M3 Ramus obscures M3 Brown and Maeda, 2009:578
(Ordered) 1. yes or partly
1. yes or partly 2. no
2. intermediate: yes or partly and no
3. no
53* Rami diverge inferiorly-superiorly (Ordered) Rami diverge inferiorly-superiorly Rosas, 2001:81
1. yes 1. yes
2. intermediate: yes and no 2. no
3. no
54 Position of mental foramina Position of mental foramina Brown et al., 2004:1058
(Ordered) (Ordered)
1. under P3 or anterior to P3 1. under P3 or anterior to P3
2. under septum P3-P4 2. under septum P3-P4
3. under P4 3. under P4
4. under P4-M1 septum 4. under P4-M1 septum
5. under M1 5. under M1
55 Number of mental foramina Number of mental foramina Johanson et al., 1982:411
(Ordered) 1. single Brown and Maeda, 2009:577
1. single 2. multiple
2. intermediate: single and multiple
3. multiple
56* Anterior notch on anterior surface of ramus in Anterior notch on anterior surface of ramus in Original researchh
lateral view lateral view
1. present 1. present
2. absent 2. absent
57* Internal ramus region form Internal ramus region form Johanson et al., 1982:412e413
1. smooth 1. smooth
2. smooth and pterygoid tubercles 2. smooth and pterygoid tubercles
3. rugose 3. rugose
4. rugose with pterygoid tubercles 4. rugose with pterygoid tubercles
58* Gonial region shape Gonial region shape
(Ordered) 1. no aring Johanson et al., 1982:412
1. no aring 2. aring (everted) Brown and Maeda, 2009:578
2. intermediate: no aring and aring
3. aring (everted)
59* Coronoid process Coronoid process Johanson et al., 1982:412
(Ordered) (Ordered) Brown et al., 2004:1058
1. higher than mandibular condyle 1. higher than mandibular condyle Brown and Maeda, 2009:578
2. intermediate: higher than and same height as 2. same height as mandibular condyle
mandibular condyle 3. lower than mandibular condyle
3. same height as mandibular condyle
4. lower than mandibular condyle
60* Deepest part of sigmoid notch Deepest part of sigmoid notch Rosas, 2001:79
(Ordered) (Ordered) Rak et al., 2002:191e200
1. towards coronoid 1. towards coronoid Mounier et al., 2009: 228, 230 (C)
2. intermediate: centrally and towards coronoid 2. centrally
3. centrally 3. towards condyle
4. towards condyle
61 Marginal torus Marginal torus Groves, 1989:232 (C)
3. absent
62 Extramolar sulcus (posterior to M3) Extramolar sulcus (posterior to M3)
(Ordered) (Ordered) Johanson et al., 1982:410
1. broad 1. broad Skelton and McHenry, 1992:319 (C)
2. intermediate: broad and medium 2. medium Kramer, 1994:10 (C)
3. medium 3. narrow McHenry, 1996:90 (C)
4. narrow 4. absent Brown et al., 2004:1058
5. intermediate: narrow and absent Strait and Grine, 2004:414
6. absent Kaifu et al., 2005:714
63 Muscle scarring on external surface of ramus Muscle scarring on external surface of ramus Original observationi
(Ordered) (Ordered)
1. absent 1. absent
2. intermediate: absent and minimal 2. minimal
3. minimal 3. marked
4. marked (not in any taxon scored)
64 Position of lateral prominence Position of lateral prominence Johanson et al., 1982:410
1. below M1 1. below M1 Bermdez de Castro et al., 2014:6
2. intermediate: below M1 and below M2 2. below M2
3. below M2 3. between M2/M3
4. intermediate below M2 and between M2/M3
5. between M2/M3
65* Mid-ramus waisting (Ordered) Mid-ramus waisting Brown and Maeda, 2009:578
1. present 1. present
2. intermediate: present and absent 2. absent
3. absent
66* Superior lateral torus Superior lateral torus
(Ordered) 1. absent Brown and Maeda, 2009:577
1. absent 2. narrow
2. intermediate: absent and narrow 3. very broad and bulbous
3. narrow
4. intermediate: narrow and very broad and
bulbous
5. very broad and bulbous
67* Mylohyoid line Mylohyoid line Brown and Maeda, 2009:578
(Ordered) (Ordered)
1. absent 1. absent
2. narrow 2. narrow
3. intermediate: narrow and broad 3. broad
4. broad
68 Superior transverse torus Superior transverse torus Johanson et al., 1982:383

(Ordered) (Ordered) Groves, 1989:219, 232 (C)
1. prominent to weak 1. prominent to weak Brown et al., 2004:1058
2. weak 2. weak Kaifu et al., 2005:714
3. very weak or absent 3. weak or absent Argue et al., 2009:Appendix 1 (C)
69 Inferior transverse torus Inferior transverse torus Johanson et al., 1982:383
(Ordered) (Ordered) Groves, 1989:219, 232 (C)
1. prominent 1. prominent Brown et al., 2004:1058
2. intermediate: prominent and weak 2. weak Argue et al., 2009:Appendix 1 (C)
3. weak 3. absent Brown and Maeda, 2009:588
4. intermediate: weak and absent
5. absent
70 Post-incisive planum Post-incisive planum Groves, 1989:232. (C)
1. strong 1. strong Kaifu et al., 2005:714
2. intermediate: strong and weak 2. weak Strait and Grine, 2004:414 (C)
3. weak 3. very weak, nearly vertical Berger et al., 2010a:199
4. very weak, nearly vertical
71 Digastric fossa Digastric fossa Andrews, 1984:175 (C)
1. absent 1. absent
2. shallow 2. shallow
3. deep 3. deep
72* Internal coronoid pillar Internal coronoid pillar Original research
3. absent
73 Submandibular fossa Submandibular fossa Mounier et al., 2009:223, 229 (C)
(Ordered) (Ordered)
1. absent 1. absent
2. short 2. short
3. intermediate: short and long 3. long
4. long
74 Genioglossal fossa Genioglossal fossa Brown and Maeda, 2009:588
(Ordered) 1. deep
1. deep 2. shallow
2. intermediate: deep and shallow 3. tubercles with no fossa
3. shallow 4. no fossa
4. intermediate: shallow and none
5. tubercles, no fossa.
6. no fossa
75* Curve of Spee Curve of Spee Groves, 1989:219 (C)
(Ordered) 1. present Brown et al., 2004:1058
3. absent
76 Basal view symphyseal region Basal view symphyseal region Johanson et al., 1982:412
1. thickest at molars/on lateral edges 1. thickest at molars/on lateral edges Brown and Maeda, 2009:577, 584
2. uniformly thick 2. uniformly thick
3. thickest at midline 3. thickest at midline
77* Curve of Monson Curve of Monson Original researchj
3. absent
78* P4 root morphology P4 root morphology Weidenreich, 1943:225
(Ordered) (Ordered) Wood et al., 1988:116e126
1. two roots 1. two roots Tobias, 1991:795
2. intermediate: Tomes and two roots 2. Tomes root Kramer, 1994:10 (C)
3. Tomes root 3. one root Brown et al., 2004:1058
4. intermediate: Tomes and one root Argue et al., 2009:Appendix 1 (C)
5. one root Brown and Maeda, 2009:578
Irish et al., 2013:1233062-2k
79* M1 ssure pattern M1 ssure pattern Johanson et al., 1978:7
(Ordered) 1. Y contact between metaconid and hypoconid Wood et al., 1983:294, 313
1. Y contact between metaconid and hypoconid 2. point contact between metaconid,
2. intermediate: Y contact between metaconid hypoconid, protoconid, entoconid
and hypoconid and point contact between
metaconid, hypoconid, protoconid, entoconid
3. point contact between metaconid,

hypoconid, protoconid, entoconid
80 M2 ssure pattern M2 ssure pattern Johanson et al., 1978:7
(Ordered) (Ordered) Wood et al., 1983:294, 313
1. Y contact between metaconid and hypoconid 1. Y contact between metaconid and hypoconid Irish et al., 2013:1233062-2 (C)
2. intermediate: Y contact between metaconid 2. point contact between metaconid,
and hypoconid and point contact between hypoconid, protoconid, entoconid
metaconid, hypoconid, protoconid, entoconid 3. X contact between protoconid and entoconid
4. X contact between protoconid and entoconid.
81* M3 ssure pattern M3 ssure pattern Wood et al., 1983:294, 313
(Ordered) 1. Y contact between metaconid and hypoconid
1. Y contact between metaconid and hypoconid 2. point contact between metaconid,
2. intermediate: Y contact between metaconid hypoconid, protoconid, entoconid
and hypoconid and point contact between
metaconid, hypoconid, protoconid, entoconid
82 Molars: tuberculum sextum Molars: tuberculum sextum Wood et al., 1983:307
1. present 2. absent Kramer, 1994:10 (C)
3. absent
83 C7 C7 Wood et al., 1983:290, 307
(Ordered) (Ordered) Kramer, 1994:10 (C)
1. C7 absent 1. C7 absent Irish et al., 2013:1233062-2 (C)
2. a short ssure is present branching off the 2. a short ssure is present branching off the
lingual ssure into the metaconid, but no cusp lingual ssure into the metaconid, but no cusp
can be identied can be identied
3. a longer and more denite ssure runs off the 3. a longer and more denite ssure runs off the
lingual ssure, with a small cusp-like formation lingual ssure, with a small cusp-like formation
identiable between them identiable between them
4. ssures clearly demarcate a well-developed 4. ssures clearly demarcate a well-developed
accessory cusp accessory cusp
(NB: Too many polymorphisms of non-adjacent
states to code them as separate states)
84 Molars: protostylid Molars: protostylid Wood et al., 1983:296
(Ordered) (Ordered) Wood, 1991:71e72
1. absent 1. absent Kramer, 1994:10 (C)
2. shape and branching of mesiobuccal groove 2. shape and branching of mesiobuccal groove Harrison, 2011:179
are suggestive of protostylid formation are suggestive of protostylid formation Irish et al., 2013:1233062-2 (C)
3. small protostylid present: a distinct branch 3. small protostylid present: a distinct branch
off the mesiobuccal groove associated with a off the mesiobuccal groove associated with a
small shelf of enamel below it small shelf of enamel below it
4. well-developed protostylid present: a distinct 4. well-developed protostylid present: a distinct
groove across buccal surface of protoconid assoc groove across buccal surface of protoconid assoc
with a long and well developed enamel shelf with a long and well developed enamel shelf
(NB: Too many polymorphisms of non-adjacent
states to code polymorphisms as separate
states)
85 P4 cusp number P4 cusp number Wood and Uytterschaut, 1987:125e128, 131,
(Ordered) (Ordered) 154
1. two cusps 1. two cusps Kramer, 1994:10 (C)
2. intermediate: two cusps and three cusps 2. three cusps Irish et al., 2013:1233062-2 (C)
3. three cusps 3. four cusps
4. intermediate: three cusps and four cusps
5. four cusps
86 Premolars: relative position of main lingual Premolars: relative position of main lingual Wood and Uytterschaut, 1987:125e128, 131,
cusp cusp 154
(Ordered) (Ordered) Kramer, 1994:10 (C)
1. bulk of deuteroconid is mesial to long axis of 1. bulk of deuteroconid is mesial to long axis of
central protoconid ridge central protoconid ridge
2. deuteroconid is centered on the long axis of 2. deuteroconid is centered on the long axis of
the ridge the ridge
3. bulk of deuteroconid is distal to long axis of 3. bulk of deuteroconid is distal to long axis of
the ridge the ridge
87* Premolars: median longitudinal ssure Premolars: median longitudinal ssure Wood and Uytterschaut, 1987:125e128, 131,
(Ordered) (Ordered) 154
1. deep and uninterrupted 1. deep and uninterrupted Kramer, 1994:10 (C)
2. intermediate: deep and uninterrupted and 2. ssure is evident but interrupted by enamel
evident but interrupted ridges leading from main cusps
3. ssure is evident but interrupted by enamel 3. ssure not evident
ridges leading from main cusps
4. intermediate: evident but interrupted and
not evident
5. ssure not evident
88 Premolars: mesiobuccal grooves Premolars: mesiobuccal grooves Wood and Uytterschaut, 1987:125e128, 132,
(Ordered) 1. present 154
1. present 2. absent Groves, 1989:218 (C)
2. intermediate: present and absent Berger et al., 2010a (C)
3. absent
89 Premolars: distobuccal grooves Premolars: distobuccal grooves Wood and Uytterschaut, 1987:125e128, 132,
(Ordered) 1. present 147, 150, 154
3. absent
90* Premolars: mesiolingual grooves Premolars: mesiolingual grooves Wood and Uytterschaut, 1987:125e128, 132,
(Ordered) 1. present 147, 154
3. absent
91* Premolars: distolingual grooves Premolars: distolingual grooves Wood and Uytterschaut, 1987:125e128, 132,
(Ordered) 1. present 147, 154
3. absent
92 Mesial occlusal edge shape of lower canine Mesial occlusal edge shape of lower canine Skelton and McHenry, 1992:316 (C)
1. prominent 1. prominent Berger et al., 2010a (C)
2. not prominent 2. not prominent
93* Prominence of lingual ridge Prominence of lingual ridge Wood and Uytterschaut, 1987:126
(Ordered) 1. prominent Berger et al., 2010a:199
1. prominent 2. not prominent
2. intermediate: prominent and not prominent
3. not prominent
94 P4 metaconid development P4 metaconid development Skelton and McHenry, 1992:317 (C)
1. absent 1. absent Harrison, 2011:150, 152, 179
2. weak 2. weak
3. intermediate: weak and strong 3. strong
4. strong
95* P3 talonid height in relation to protoconid P3 talonid height in relation to protoconid Wood and Uytterschaut, 1987:153e154
1. very low 1. very low Brown et al., 2004:1059
2. intermediate: very low and low 2. low
3. low 3. moderate
4. intermediate: low and moderate
4. moderate
96 P3 occlusal crown outline P3 occlusal crown outline Wood and Uytterschaut, 1987:142e143
(Ordered) (Ordered) Skelton and McHenry, 1992:317 (C)
1. long/narrow/oval 1. long/narrow/oval McHenry, 1996:90 (C)
2. intermediate: long/narrow/oval and round 2. intermediate: long/narrow/oval and round
3. round 3. round
97 P3 mesiolingual beak (Ordered) P3 mesiolingual beak Begun et al., 1997:396 (C)
1. present 1. present Finarelli and Clyde, 2004:642 (C)
2. intermediate: present and absent 2. absent Harrison, 2011:152
3. absent
98 P3 occlusal crown outline P3 occlusal crown outline Wood and Uytterschaut, 1987:129e130
(Ordered) 1. along axis of tooth row
1. along axis of tooth row 2. at angle to tooth row
2. intermediate: along axis of tooth row and at
angle to tooth row
3. at angle to tooth row
99 P3 metaconid (Ordered) P3 metaconid Begun et al., 1997:396 (C)
1. present 1. present Finarelli and Clyde, 2004:639 (C)
2. intermediate: present and absent 2. absent Strait and Grine, 2004:414 (C)
3. absent
100 M3/M2 size ratio (Ordered) M3/M2 size ratio Weidenreich, 1943:225
1. M3 > M2 1. M3 > M2 Johanson et al., 1978:7
2. intermediate: M3 > M2 and M3 M2 2. M3 M2 Lieberman et al., 1996:90 (C)
3. M3 M2 Begun et al., 1997:400 (C)
Brown et al., 2004:1058
Finarelli and Clyde, 2004:642 (C)

101* Infraglenoid tubercle (scapular) Infraglenoid tubercle (scapular) Begun et al., 1997:392 (C)
1. broad 1. broad Finarelli and Clyde, 2004:639 (C)
2. narrow 2. narrow Larson et al., 2009:566
102* Axillary border (scapular) Axillary border (scapular) Larson et al., 2009:566
1. pronounced and rounded 1. pronounced; rounded
2. attened 2. attened
3. sharp ridge 3. sharp ridge
103* Spinous process root (scapular) Spinous process root Begun et al., 1997:392 (C)
(Ordered) (scapular) Finarelli and Clyde, 2004:639 (C)
1. robust 1. robust
2. intermediate: robust and gracile 2. gracile
3. gracile
104* Orientation of scapular spine Orientation of scapular spine Larson et al., 2007:723
1. 20 e40 1. 20 e40 Larson et al., 2009:566
2. intermediate: 20 e40 and 41 e60 2. 41 e60 Data source for A. afarensis, A. africanus,
3. 41 e60 3. >60 H. ergaster, H. oresiensis, H. sapiens, Pan, Gorilla,
4. intermediate 41e60 and >60 Pongo: Larson et al., 2007:725, Figure A
5. >60
105* Bar-glenoid angle (scapular) Bar-glenoid angle (scapular) Larson et al., 2007:725
1. 135 (superiorly directed glenoid fossa) 1. 135 (superiorly directed glenoid fossa) Data source for A. afarensis, A. africanus,
2. 136 e150 2. 136 e150 H. ergaster, H. oresiensis, H. sapiens, Pan, Gorilla,
3. >150 3. >150 Pongo: Larson et al., 2007:725, Figure B
106* Sub-scapula fossa Subscapula fossa Larson et al., 2009:566
1. at to convex 1. at to convex
2. at to concave 2. at to concave
107* Infra spinous fossa (scapular) Infra spinous fossa (scapular) Larson et al., 2009:566e567
1. convex with gentle curve to axillary border 1. convex with gentle curve to axillary border
2. at or mild concavity with abrupt step to 2. at or mild concavity with abrupt step to
axillary border axillary border
108* Humerus: anteroposterior shaft Humerus: anteroposterior shaft curvature Groves, 1989:221 (C)
curvature (Ordered) 1. retroexed Begun et al., 1997:392 (C)
1. retroexed 2. straight Finarelli and Clyde, 2004:639 (C)
2. intermediate: retroexed and straight
3. straight
109 Humeral torsion Humeral torsion Larson et al., 2007:722
1. low 1. low Lordkipanidze et al., 2007:307
2. high 2. high Argue et al., 2009:Appendix 1 (C)
110 Linea aspera (femur) (Ordered) Linea aspera (femur) Brown et al., 2004:1059
1. well developed 1. well developed Jungers et al., 2009a:542
2. intermediate: well developed and not well 2. not well developed
developed
3. not well developed
111* Intertrochanteric crest (Ordered) Intertrochanteric crest Groves, 1989:221 (C)
1. attened 1. attened Brown et al., 2004:1059
2. intermediate: attened and strongly 2. strongly developed Jungers et al., 2009a:542
developed
3. strongly developed
112* Greater trochanter Greater trochanter Jungers et al., 2009a:543
1. prominent lateral apophysis 1. prominent lateral apophysis
2. intermediate: prominent lateral apophysis 2. lateral apophysis at
and at lateral apophysis
3. lateral apophysis at
113 Biomechanical neck length (cf femoral head Biomechanical neck length (cf femoral head Brown et al., 2004:1059
diameter) diameter)
1. relatively long 1. relatively long
2. short 2. short
114 Femur shaft in cross-section Femur shaft in cross-section Groves, 1989:221 (C)
1. circular 1. circular Jungers et al., 2009a:543
2. intermediate: circular and oval 2. oval
3. oval 3. tear-shaped
4. tear-shaped
115 Pilaster on femur shaft Pilaster on femur shaft Brown et al., 2004:1059
(Ordered) (Ordered) Jungers et al., 2009a:543
1. well-dened 1. well-dened
2. intermediate: well-dened and weakly
dened. 2. present but weakly dened

3. present but weakly dened 3. absent
4. intermediate: weakly dened and absent.
5. absent
116* Bicondylar angle Bicondylar angle Brown et al., 2004:1059
1. low 1. low Data source:
2. high 2. high Jungers et al., 2009a:544
117* Robustness of bula Robustness of bula Aiello and Dean, 1990:498
(Ordered) (Ordered)
1. high 1. high
2. medium 2. medium
3. low 3. low
118* Robustness of tibia Robustness of tibia Jungers et al., 2009a:545
1. high 1. high
2. low 2. low
119* Ilium Ilium Brown et al., 2004:1058
(Ordered) 1. ares strongly beyond margin of acetabulum Jungers et al., 2009a:541
1. ares strongly beyond margin of acetabulum 2. moderate aring
2. intermediate: strong aring and moderate
aring
3. moderate aring
120* Acetabulum notch Acetabulum notch Jungers et al., 2009a:541
1. broad 1. broad
2. narrow 2. narrow
121* Iliac pillar Iliac pillar Jungers et al., 2009a:540
1. virtually no pillar 1. virtually no pillar
2. intermediate: virtually no pillar and pillar 2. pillar anterior and weak
anterior and weak 3. more posterior and marked
3. pillar anterior and weak
4. more posterior and marked
122* Iliac crest Iliac crest Jungers et al., 2009a:540e541
(Ordered) 1. anterior medial twist
1. anterior medial twist 2. no twist
2. intermediate: anterior medial twist and no
twist
3. no twist
123* Postcranial proportions Postcranial/proportions Leakey et al., 1964l
(Ordered) (Ordered) Groves 1989:221 (C)
1. humerofemoral index high (>80) 1. humerofemoral index high (>80) Tobias, 1991:807l
2. humerofemoral index medium (61e80) 2. humerofemoral index medium (61e80) Morwood et al., 2005:1016
3. humerofemoral index low (<60) 3. humerofemoral index low (<60) Lordkipanidze et al., 2007:308
Data sources:
Lordkipanidze et al., 2007:308 (H. georgicus)
Jungers, 2009a:95, Figure 9.2
124* Tibia torsion Tibia torsion Aiello and Dean, 1990:497
(Ordered) (Ordered) Jungers et al., 2009a:546
1. negative torsion 1. negative torsion Pontzer et al., 2010:494
2. near zero 2. near zero
3. positive torsion 3. positive torsion
125* Tarsus: talar head torsion Tarsus: talar head torsion Day and Wood, 1968
1. low torsion 1. low torsion Lordkipanidze et al., 2007:308
2. high torsion 2. high torsion Jungers et al., 2009a:546
Data sources:
Lordkipanidze et al., 2007:308 (H. georgicus)
Jungers et al., 2009a:546 (H. oresiensis)
Berger et al., 2010b:20.
126* Navicular tuberosity Navicular tuberosity Jungers et al., 2009a:547
(Ordered) (Ordered) Data source:
1. large (tuberosities project beyond facet for 1. large (tuberosities project beyond facet for Jungers et al., 2009a:547
entocuneiform projection) entocuneiform projection)
2. medium projection of tuberosities 2. medium projection of tuberosities
3. reduced (tuberosities do not reach facet for 3. reduced (tuberosities do not reach facet for
entocuneiform) entocuneiform)
127* Femoral neck index (ratio of AP/SI) Femoral neck index (ratio of AP/SI) Groves, 1989 (C)
1. high index 1. high index Data source:
2. low index 2. low index Jungers et al., 2009a:543
128* Relative clavicle length (ratio of clavicle to Relative clavicle length (ratio of clavicle to Larson et al., 2007:723e274
humerus) humerus) Data source: Larson et al., 2007:724, Figure 7
(Ordered) (Ordered)
1. short 1. short
2. medium 2. medium
3. long 3. long
129* Cortical index Cortical index Jungers et al., 2009a:552

(Ordered) (Ordered)
1. cortical bone thin 1. cortical bone thin
2. intermediate and thick and thin 2. intermediate: thick and relatively thin
3. thick 3. thick
130 Mastoid process inection Mastoid process inection Weidenreich, 1943:62, 205
(Ordered) (Ordered) Skelton and McHenry, 1992:320 (C)
1. inected under cranial base 1. inected under cranial base McHenry, 1996:90 (C)
2. intermediate: inected and not inected 2. intermediate: inected and not inected Martinez and Arsuaga, 1997:292
3. not inected 3. not inected Argue et al., 2009:Appendix 1 (C)
131 Supraorbital torus Supraorbital torus Groves, 1989:230 (C)
1. bar-like structure 1. bar-like structure Lieberman et al., 1996:101e104 (C)
2. arched over orbits 2. arched over orbits Brown et al., 2004:1057
132 Supraorbital torus aring of trigones Supraorbital torus aring of trigones Rak, 1983:74
1. present 1. present Groves, 1989:230 (C)
2. absent 2. absent Brown et al., 2004:1057
133 Sagittal crest Sagittal crest Weidenreich, 1943:194
(Ordered) 1. present Ashton and Zuckerman, 1956:Summary, dot
1. present 2. absent point 3. (Data source for Gorilla, Pan, Pongo)
2. intermediate: present and absent Tobias, 1991:749
3. absent Skelton and McHenry, 1992:321 (C)
*Denotes character unknown in 4 taxa, and excluded in one of the analyses (see text).
a
The second column lists characters and character states used in the study, as coded in dataset A (intraspecic variability coded as separate, intermediate state); the third
column comprises an alternate way of expressing the character states as employed in dataset B (intraspecic variability coded as polymorphism, i.e., multiple states). The
fourth column provides selected publications in which each character has been used in determining hominin phylogenies or its importance discussed, and the data sources we
used when character states for particular specimens were not available for original observation.
b
We found that the character was not represented solely by states 1 and 3. Data obtained from our measurements of inionopisthion and lambda-inion chords on originals
using digital calipers.
c
Assessed by measuring the thickness of the anterior perimeter of tympanic, using digital calipers.
d
Observation made on LB1 following information provided by Dr. David Bulbeck, The Australian National University, to DA and CG.
e
The infraorbital foramina on LB1 appeared large. We, therefore, added this character to assess any inter-species differences in size, where observable. Also see Swartz and
Tattersall, 2002:488 (large infraorbital foramen on H. erectus Sangiran 17).
f
We observed a number of anterior-posterior grooves and peaked crests on the LB1 palate. We, therefore, included this character to assess if it occurred in other species.
g
The incisive fossa on LB1 appeared large (disregarding the pathology evident in this region). We, therefore, added this character to assess any inter-species differences in
size.
h
Indicator of masticatory power and insertion of temporalis.
i
Noting that (lack of) muscular relief in this region is mentioned in relation to D2735 (Rightmire et al., 2006:129). Indicator of masticatory power.
j
Contributes to occlusal pattern.
k
Data is from our observation on mandibles where P4 roots visible; otherwise from Wood et al., 1988:116, Table 5.
l
importance of body proportions.
112: 1 and 2/?, 115: 2 and 3 / 2). We repeated the parsimony and or (2) the sister to a core Homo branch that includes H. habilis, H.
Bayesian analyses with these modied datasets. erectus, H. ergaster, and H. sapiens. In the latter arrangement, both
The PAUP and MrBayes executable analysis les are available as H. oresiensis and H. habilis typically appear as successive outgroups
SOM on the journal website. to advanced Homo. Thus, the analyses place H. oresiensis as either a
close relative to H. habilis, or at a similar evolutionary grade.
3. Results Parsimony analysis of dataset A with H. rudolfensis excluded
yields the shortest tree (Length 584; Fig. 1B) in which
Both the parsimony and Bayesian analyses gave highly concor- H. oresiensis groups with H. habilis. This clade has a Bremer sup-
dant results, whether analyses excluded H. rudolfensis (see Figs. 1 port of 4 and bootstrap of 68%. The next best position of
and 2, Table 3A) or included all taxa (SOM Fig. S1, Table 3B). Sup- H. oresiensis is outside core Homo (four steps longer, bootstrap
port was slightly higher when polymorphisms were treated as new 21.3%); this hypothesis thus cannot be robustly rejected (Table 3A;
states (dataset A: Fig. 1), compared to when polymorphisms were analyses with H. rudolfensis excluded). Relationships between
scored as multiple states (dataset B: Fig. 2), suggesting the former H. oresiensis and any single taxon other than H. habilis were even
coding method extracted a little more phylogenetic signal. In more unparsimonious (at least 11 extra steps cf best tree) and
virtually all analyses, the two best hypotheses for the afnities of occurred in 3.7% of bootstraps; these hypotheses can therefore be
H. oresiensis were that it was either (1) the sister to H. habilis alone condently rejected. Notably, afnities with H. erectus were
Figure 1. Afnities of Homo oresiensis based on phylogenetic analyses of dataset A (intraspecic variability coded as new state; see main text). Homo rudolfensis is excluded. Note
that while support for many nodes is not strong, most alternative hypotheses for the position of H. oresiensis can be rejected (see Table 3A). (A) Parsimony: majority-rule consensus
of bootstrap trees. Numbers at nodes refer to bootstrap frequency. (B) Parsimony: shortest tree, length 584. Numbers at nodes refer to branch (Bremer) support. (C) Bayesian:
majority-rule consensus of sampled post-burnin trees. Numbers at nodes refer to posterior probability.
Figure 2. Afnities of Homo oresiensis based on phylogenetic analyses of Dataset B (intraspecic variability coded as polymorphism, i.e., multiple states; see main text).
Homo rudolfensis is excluded. Note that while support for many nodes is not strong, most alternative hypotheses for the position of H. oresiensis can be rejected (see Table 3B). (A)
Parsimony: majority-rule consensus of bootstrap trees. Numbers at nodes refer to bootstrap frequency. (B) Parsimony: strict consensus of three shortest trees, length 274.
Numbers at nodes refer to branch (Bremer) support. (C) Bayesian: majority-rule consensus of sampled post-burnin trees. Numbers at nodes refer to posterior probability.
Table 3
Support for alternative hypotheses for Homo oresiensis.
(A) Polymorphism as intermediate state
Analyses with H. rudolfensis excluded Tree length MPTs Bootstrap% Bayesian pp
Sister to H. oresiensis
H. habilis 584 (best) 1 68.0 0.32
Outside Core Homoa 588 (4) 3 21.3 0.42
Dembo et al. topology (looseb) 592 (8) 3 5.6 0.24
A. sediba 595 (11) 1 3.7 0
A. afarensis 596 (12) 1 1.6 0.077
Dembo et al. topology (strictc) 598 (14) 1 0.4 0.005
H. erectus 598 (14) 1 1.9 0.005
A. africanus 598 (14) 1 0.7 0.011
H. naledi 599 (15) 1 0.2 0
H. georgicus 600 (16) 1 0 0
H. sapiens 608 (24)* 4 0 0
H. ergaster 609 (25)* 1 0 0
Analyses with H. rudolfensis included Tree length MPTs Bootstrap% Bayesian pp
H. habilis 596 (best) 1 66.5 0.338
Outside Core Homoa 599 (3) 1 22.4 0.435
H. rudolfensis 605 (9) 4 2.8 0.013
A. sediba 608 (12) 1 0.2 0
A. afarensis 609 (13) 1 1.1 0.078
H. erectus 610 (14) 3 1.3 0
A. africanus 611 (15) 2 0.9 0.008
Dembo et al. topology (strictc) 613 (17) 3 0.3 0.001
H. georgicus 613 (17) 4 0 0
H. naledi 613 (17) 1 0.1 0
H. sapiens 620 (24)* 1 0.1 0
H. ergaster 622 (28)* 2 0 0
(B) Polymorphism as variable
Analyses with H. rudolfensis excluded Tree length MPTs Bootstrap% Bayesian pp
Outside Core Homoa 274 (best) 3 31.4 0.329
H. habilis 275 (1) 11 54.0 0.506
H. erectus 278 (4) 1 1.2 0.007
A. sediba 279 (5) 1 1.9 0.005
A. africanus 279 (5) 1 4.6 0.019
A. afarensis 281 (7) 1 1.5 0.049
Dembo et al. topology (strictc) 283 (9) 2 0 0.001
H. naledi 283 (9) 2 0 0
H. georgicus 283 (9) 3 0.2 0
H. ergaster 286 (12) 7 0 0.005
H. sapiens 286 (12) 5 0.1 0.007
Analyses with H. rudolfensis included Tree length MPTs Bootstrap% Bayesian pp
H. habilis 280 (best) 1 50.6 0.484
Outside Core Homoa 280 (best) 9 37.9 0.376
H. rudolfensis 284 (4) 3 1.9 0.017
A. sediba 284 (4) 2 1.2 0.005
H. erectus 284 (4) 2 0.7 0.006
A. afarensis 286 (6) 1 1.5 0.047
A. africanus 286 (6) 5 1.9 0.012
H. georgicus 288 (8) 3 0.1 0
H. naledi 288 (8) 4 0.3 0
Dembo et al. topology (strictc) 290 (10) 2 0 0
H. ergaster 291 (11) 1 0 0
H. sapiens 291 (11)* 2 0.1 0.005
* Signicant at p < 0.05 under Templeton Test.

a
Core Homo H. habilis, H. erectus, H. ergaster, H. sapiens.
b
Dembo et al. (loose) H. oresiensis more closely related to H. erectus, H. ergaster, and H. sapiens than to H. habilis (all other hominid taxa unconstrained).
c
Dembo et al. (strict) H. oresiensis more closely related to H. erectus, H. ergaster, H. sapiens, and H. rudolfensis (when included) than to H. habilis; H. habilis sister to A. sediba
(all other hominid taxa unconstrained).
strongly contradicted (14 extra steps, 1.9% bootstrap) and afnities reject most of these alternative arrangements for H. oresiensis, but
with H. sapiens was one of the worst evaluated hypotheses (24 extra rejected afnities with H. sapiens and with H. ergaster at p < 0.05
steps, 0% bootstrap). Templeton's (1983) nonparametric test, which (Table 3A). Amongst other relationships, H. erectus and H. ergaster
is likely overly conservative (e.g., Lee, 2000), predictably could not form a clade (62% bootstrap, Bremer 3); H. georgicus and H. naledi
group weakly together with only 46% bootstrap and Bremer 2; and i) that it derives from a putative population of H. erectus on Flores
A. africanus and A. sediba are also weakly grouped with only 25% that dwarfed as an evolutionary response to particular condi-
bootstrap and Bremer 2. tions found on islands; or
Parsimony analysis of dataset A with H. rudolfensis included gave ii) it is a late-surviving species of early Homo.
similar results regarding the immediate relationships of
H. oresiensis (SOM Fig. S1A, B): it emerged as either sister to Additionally, we address some issues in relation to the pathol-
H. habilis or outside of core Homo. H. rudolfensis and H. naledi, ogy hypotheses for H. oresiensis.
however, emerged in a heterodox position at the base of the tree, As presented above, all the analyses (parsimony/Bayesian,
possibly due to large amounts of missing data for these taxa (>50% dataset A/B) were generally concordant regarding this key question,
for both taxa). so for brevity our discussion focuses on the parsimony analysis of
Parsimony analyses of dataset B yielded similar results (Fig. 2A, dataset A without H. rudolfensis, hereinafter termed the focal
B; Table 3B). In this analysis, the favored bootstrap result again analysis (Fig. 1A, B; SOM Appendix 1).
grouped H. oresiensis with H. habilis (54%). The hypothesis that Support for most clades in all analyses was rather low as measured
H. oresiensis was outside of core Homo was slightly more parsi- by traditional methods (bootstrap and Bremer support in parsimony,
monious at one step shorter, though it occurred in fewer boot- posterior probability in Bayesian inference), but are nevertheless
straps, i.e., 31.4% (Table 3B; analyses with H. rudolfensis excluded). broadly concordant with values from other similar studies (e.g.,
Other tested hypotheses are substantially less parsimonious (4 Dembo et al., 2015). Note that low support for a particular clade
steps cf shortest tree), very infrequently sampled during boot- merely means that one or more conicting clades might need to be
strapping (4.6%), and can be rejected. The hypothesis that considered; it does not mean that all other possibilities should be
H. oresiensis was sister to H. sapiens was the equal-worst hy- entertained. As we discuss below, the low support for H. oresiensis
pothesis evaluated (12 extra steps, 0% bootstrap). Templeton's with H. habilis is due to the existence of a plausible alternative, where
(highly conservative) nonparametric test, however, could not reject H. oresiensis moves slightly further down the tree and is sister to
most of these alternative arrangements for H. oresiensis; the sole what we term core Homo. While these two hypotheses cannot be
hypothesis rejected at p < 0.05 was afnities with H. sapiens when rejected, certain other positions of H. oresiensis (notably, sister to
H. rudolfensis was included (Table 3B). H. erectus or to H. sapiens) can be rejected with high condence. We
Bayesian analyses of datasets A and B gave concordant results are aware, however that our results for H. naledi and H. georgicus may
with the parsimony analyses and consistently retrieved the same be a consequence of the degree of missing data in these species.
two arrangements regarding H. oresiensis. For dataset A,
H. oresiensis was consistently sister to core Homo in the majority- 4.1. H. oresiensis and H. habilis analysis
rule consensus tree (Fig. 1C, SOM Fig. S1C), but a sister-group
relationship to H. habilis occurred at high frequency (>0.32; The branch leading to these two sister taxa is diagnosed by
Table 3A). For dataset B, H. oresiensis was consistently sister to number of unambiguous (i.e., optimization-independent) derived
H. habilis in the majority-rule consensus tree, but was sister to core character states. For some of these traits, the morphological con-
Homo at high frequency (>0.32; Fig. 2C, Table 3B). dition in H. oresiensis and H. habilis is unique, i.e., not found (either
Excluding characters with relatively poor sampling did not as a xed or polymorphic state) in all other sampled Homo; these
appreciably change the topological results grouping of unique conditions are identied with an asterisk (*).
H. oresiensis and H. habilis. Parsimony analysis of (reduced) dataset
A grouped H. oresiensis with H. habilis (bootstrap 80%; cf. Fig. 1A, External occipital protuberance prominent (character 12, state
B), as did Bayesian analysis (posterior probability 0.62; cf Fig. 1C). 1dall states listed refer to codings as per dataset A)
Parsimony analysis of (reduced) dataset B grouped H. oresiensis Fissure between mastoid process and petrous crest of tympanic
with H. habilis (bootstrap 58%; cf. Fig. 2A, B), as did Bayesian anal- present (character 15, state 1)
ysis (posterior probability 0.65; cf Fig. 2C). Some other relationships Greatest axis of auditory meatus from inferior margin to supe-
changed, however, most notably the two species that are the more rior margin slopes posteriorly (character 23, state 5)*
poorly sampled, H. georgicus and H. naledi, grouped with H. sapiens, Supraorbital sulcus present (character 25, state 1)
a heterodox grouping that is an artifact of this character deletion Metopic prominence at least variably present (character 29,
strategy and that thus supports the view that inclusion of all state 2.)
characters, including those with some arbitrary level of missing Temporal line not directly linked with supramastoid crest
data, is advisable (Wiens, 2000). (character 44, state 1)
Recoding H. habilis by excluding equivocally-attributed post- Inferior transverse torus prominent (character 69, state 1)*
cranial remains also did not appreciably change the grouping of Digastric fossa absent (character 71, state 1)
H. oresiensis and H. habilis topological results. Parsimony analysis Contact between metaconid and hypoconid on M2 is Y-shaped
of dataset A grouped H. oresiensis with H. habilis (bootstrap 72%; (character 80, state 1)
cf Fig. 1A, B), as did Bayesian analyses (posterior probability y Lower premolar mesiobuccal grooves present (character 88,
0.36; cf. Fig. 1C). Parsimony analysis of dataset B grouped state 1)
H. oresiensis with H. habilis in the bootstrap consensus (49%; cf Lower premolar distobuccal grooves present (character 89,
Fig. 2A), but placed it as sister to H. habilis plus core Homo in the state 1)
most-parsimonious trees (cf Fig. 2B); Bayesian analyses grouped Lower premolar mesiolingual grooves present (character 90,
H. oresiensis with H. habilis (posterior probability 0.48; cf Fig. 2C). state 1)
Relationships between the other taxa were also similar to the P3 mesiolingual beak present (character 97, state 1)
original analyses. Pilaster on femur shaft at least weakly dened and absent, or
absent entirely (character 115, state 4)
4. Discussion Humerofemoral index medium (character 123, state 2)*
Our objective was to test the two mainstream hypotheses for These characters span the entire skeleton. One recent iteration
H. oresiensis: of the pathology hypothesis (Henneberg and Thorne, 2004) argued
that the distinguishing features of H. oresiensis are due to Down clivus nasoaveolaris at or slightly concave medio-laterally at
syndrome. As noted by Dembo et al. (2015), this hypothesis midsection (character 38, state 2)
implicitly predicts that the characters separating H. oresiensis from angular torus at least variably present (character 46, state 2)
H. sapiens on any phylogenetic tree should be related to Down extramolar sulcus broad (character 62, state 1)
syndrome, but none of the traits listed above (except for 123 to an protostylid small (character 84, state 3)
extent; see Baab et al., 2016) are among the pathologies listed M3 > M2 (character 100, state 1)*.
(Henneberg and Thorne, 2004: Table 1) as correlated with Down
syndrome. Thus, the phylogenetic position of H. oresiensis ob- We therefore do not robustly resolve the phylogenetic re-
tained here cannot be ascribed to Down syndrome. lationships of H. naledi, or indeed H. georgicus, though it should be
Conversely, H. erectus always groups with H. ergaster and emphasized that neither of these Operational Taxonomic Units are
H. sapiens (and thus, is separated from H. oresiensis) across all ana- well-represented in the data matrix compared to other species
lyses, though with varying support levels (Figs. 1 and 2, SOM Fig. S1). (except H. rudolfensis) and this might have impacted upon our re-
The following unambiguous (optimization-independent) derived sults for these taxa. Further, our character and taxon sampling was
traits diagnose the erectus-ergaster-sapiens branch in the focal anal- not aimed at resolving this problem. Schroeder et al. (2017), having
ysis; some of these are fully absent in all other sampled Homo (*): performed a generalized Procrustes analysis of cranial and
mandibular shape in Homo, found that H. naledi is similar in aspects
Medial incursion of the superior temporal lines at coronal su- of cranial shape to H. georgicus (H. erectus in their analyses) crania
ture: inection at least variably present (character 28, state 3) D22801 and D27002, but that it also shares cranial shape similarities
Canine juga absent (character 35, state 3) with H. ergaster KNM-WT 150003 and KNM-ER 38834 (H. erectus in
Origin of masseter muscle is a narrow fossa (character 42, state 3) their analyses), H. rudolfensis KNM-ER 14705, H. habilis KNM-ER
*(H. erectus unknown) 18136, and A. sediba MH17. Schroeder et al. (2017) concluded they
Obelionic depression at least variably present (character 45, state 2) were unable to obtain any clear and consistent associations of
Genioglossal fossa shallow or absent (character 74, state 4) H. naledi with any other species in their analyses beyond its broader
P3 talonid height in relation to protoconid is moderate afliation with Homo.
(H. erectus n/a; character 95, state 4) Several studies have examined the phylogenetic position of
P3 occlusal crown outline along axis of tooth row at least var- A. sediba. Based on craniodental characteristics, Berger et al. (2010a)
iably present (character 98, state 2) propose that A. sediba is basal to H. habilis, H. rudolfensis, H. erectus,
Orientation of scapula spine is intermediate, between 41 and and SK 847, in which H. erectus is represented by Baringo, Chem-
60 , and >60 (character 104, state 4) (H. erectus unknown) eron, Dmanisi, East and West Lake Turkana, Konso, Olduvai Gorge,
Iliac pillar posterior and marked (character 121, state 4) Sangiran, Swartkrans, Tighenif, Trinil, and Zhoukoudian. Dembo
(H. erectus unknown)* et al. (2015) also nd that the best model for A. sediba is as the
Femoral neck index (ratio of AP/SI) high (character 127, state 1) sister taxon of a clade comprising all Homo species. Irish et al.
(H. erectus unknown)* (2013) present a cladogram derived from the analysis of 22
nonmetric dental traits in which A. africanus and A. sediba form a
The bootstrap of 68% for the H. oresiensis-H. habilis clade in our clade with a bootstrap value of 48%. They suggest that the species is
focal analysis, however, indicates that this clade is not present in close to A. africanus and that both share a number of apomorphies
the remaining 32% of bootstrap trees. Some other conicting ar- with Homo. Our results also show similarly low bootstrap values for
rangements presumably occur at appreciable frequency. One places an A. sediba/A. africanus clade. We, therefore, make no strong
H. oresiensis as a sister group to core Homo: H. habilis, H. erectus, conclusions as to A. sediba's phylogenetic relationship to
H. ergaster, and H. sapiens (Table 3A). This alternative position en- A. africanus.
tails four extra steps and occurs with a bootstrap frequency of 21.3% Dembo et al. (2015) analyzed the relationships of all hominins,
(Table 3A), and thus cannot be rejected. Tests for possible phylo- including H. oresiensis, using Bayesian methods. While they
genetic relationships of H. oresiensis with any of the other species recovered a slightly higher position for H. oresiensis, it is notable
in the analyses returned unparsimonious results (at least 11 extra that they too did not unite it with H. erectus (or H. sapiens for that
steps and 3.7% of bootstrap resamples). In particular, placing matter). We tested two versions of their phylogenetic hypothesis
H. oresiensis as sister to H. erectus entailed 14 extra steps, and this for H. oresiensis (their Fig. 1). In the loose version, H. oresiensis is
arrangement occurred in only 1.9% of bootstraps. Incidentally, constrained to be more closely related to H. erectus, H. ergaster, and
placing H. oresiensis as sister to H. sapiens fared particularly badly, H. sapiens than is H. habilis (all other hominins were left uncon-
entailing 24 extra steps and never occurring in bootstraps (0%). strained, including the wildcard H. rudolfensis). In the strict
The other parsimony and Bayesian analyses returned concor- version, H. oresiensis is again constrained to be more closely
dant results: the same hypotheses for H. oresiensis were consis- related to H. erectus, H. ergaster, H. sapiens, and (when included)
tently retrieved as the most plausible (H. oresiensis sister to either H. rudolfensis than is H. habilis, with H. habilis constrained to be
H. habilis alone or to core Homo including H. habilis). In contrast, the sister to A. sediba (again, all other hominins were left
hypotheses that H. oresiensis was related to H. erectus or H. sapiens unconstrained).
were both highly suboptimal, with the latter being consistently one The loose constraint for H. oresiensis is seven steps longer than
of the worst arrangements investigated (Table 3). our most parsimonious tree and has a bootstrap of 4.6%, and thus
Although H. naledi and H. georgicus form a clade in our focal
analysis, it is not very well-supported (46% bootstrap; Bremer of 2).
The branch leading to these two taxa is diagnosed by only seven
1
unambiguous (optimization-independent) derived characters, and Schroeder et al., 2017:Fig. 2, Analysis A, Principal component (PC) 1.
2
Schroeder et al., 2017:Fig. 2, Analysis B; PC2.
only one (*) does not occur in any of the other sampled Homo: 3
Schroeder et al., 2017:Fig. 2, Analysis A; PC1; Analysis D, PC 2.
4
Schroeder et al., 2017:Fig. 2, Analysis A, PC1; Analysis B, PC 1; Analysis C, PC 1.
mastoid process does not project inferiorly (character 14, state 3) 5
Schroeder et al., 2017:Fig. 2, Analysis B, PC 1.
medio-lateral angulation of the articular eminence is parallel to 6
Schroeder et al., 2017:Fig. 2, Analysis D, PC 1.
7
the coronal plane (character 18, state 3) Schroeder et al., 2017:Fig. 2, Analysis A, PC 2.
cannot be rejected by our dataset (Table 3A; analyses with cladistic) study of the hominin dentition, favoring the hypothesis
H. rudolfensis included). The strict constraint is 17 steps longer than that H. oresiensis evolved from an early Javanese H. erectus
our most parsimonious solution and has 0.3% bootstrap, and thus is population or a related form. This conclusion was based primarily
more strongly rejected (Table 3A). Bayesian analyses gave broadly on clustering of H. oresiensis and H. erectus in the crown contours
compatible results. Trees compatible with the loose constraint were of three teeth8, but this is not a uniquely derived character, the
sampled with high frequency (0.204; Bayes Factor <3), but trees crown contours of these three teeth being also within the range of
compatible with the strict constraint were almost never sampled H. sapiens (Kaifu et al., 2015). Moreover, Kaifu et al. (2015) showed
(0.001; BF 6). that the crown contour shapes of eight other molars and pre-
There are several possible reasons for the differences between molars in H. oresiensis fall outside the H. erectus range9, and one
our preferred phylogeny and that of Dembo et al. (2015), although it premolar (LB2 P3) was within the H. habilis range10, and they
should be noted that both analyses agree in placing H. oresiensis concluded that none showed similarities to H. erectus. While their
outside the H. erectuseH. sapiens clade. Dembo et al. (2015) used results remain to be explained, it does not appear that they are
exclusively craniodental data; we include postcranial characters so sufciently unambiguous to contradict the results reported in the
that these important aspects of morphology can contribute to present study.
phylogenetic signal. There are also differences in the hypodigms To support the hypothesis that H. oresiensis was an island
(species boundaries) used in our respective studies, in particular dwarfed form of Javanese H. erectus, we must demonstrate that it
the composition of H. erectus. To test the hypothesis that was closely related to that species; in this case, we would expect
H. oresiensis is a descendant of H. erectus from Sangiran, Java (Kaifu H. oresiensis and H. erectus to form sister taxa in optimal phylo-
et al., 2011, 2015), our hypodigm comprises just H. erectus sensu genetic trees. They do not, but we may further evaluate the plau-
strictodthe Sangiran/Trinil hominins from Java. H. erectus in sibility of this hypothesis by constraining H. oresiensis with
Dembo et al. (2015) comprises Hexian, Zhoukoudan/Sinanthropus, H. erectus. Our results (Table 3A; analyses with H. rudolfensis
Maba (China); Ngandong, Sambungmacan, Sangiran, Ngawi excluded) show that forcing a sister-group relationship of
(Indonesia); and KNM-ER 992, Tigenif, and OH 9 (Africa). Dembo H. oresiensis and H. erectus produces trees that are 14 steps longer
et al. (2015) therefore tested a different hypothesis for than the most parsimonious tree, and this arrangement is rarely
H. oresiensisdwhether H. oresiensis is a dwarfed morphology of a found in bootstraps (1.9%) and almost never in Bayesian inference
suite of hominins variously attributed to H. erectus from Indonesia, (0.005). This phylogenetic relationship is thus rejected by our data.
China, and Africa. Hominins from the latter two regions, however, Accordingly, there is strong evidence against the hypothesis that
have not been proposed as ancestral to H. oresiensis. Other hypo- H. oresiensis was an island dwarf form of H. erectus.
digms are also not equivalent across the two studies. Dembo et al. All parsimony analyses can also reject with high condence the
(2015) included a considerably wider geographic range in their following as sister taxa to H. oresiensis: H. sapiens, H. ergaster,
H. habilis sample than we included; and their sample for H. georgicus, A. afarensis, A. sediba, and A. africanus (Table 3A). All
H. rudolfensis was far broader than some consider should belong in these pairings are rarely sampled, found in 3.7% of bootstrap trees
this species. Such differences are inevitable given the debate about and 0.05 of Bayesian trees. It is nonetheless interesting that, while
species boundaries in hominins and a priority for future phyloge- all these values are low, the highest values often involve pairings
netic studies should be to agree on hypodigms (if necessary, split- between H. oresiensis and Australopithecus species (e.g., 3.7%
ting up contentious taxa into smaller groupings that are not bootstrap with A. sediba, 0.077 Bayesian prior probability with
disputed). Finally, while Dembo et al.'s (2015) matrix includes more A. afarensis). This is more consistent with H. oresiensis having the
taxa and more characters (and thus resolves total hominin phy- morphology characteristic of a relatively primitive (Australopithe-
logeny more completely than do our data), it has only 12% of cine-like) Homo, and less consistent with H. oresiensis being
characters scored for H. oresiensis. derived in the direction of H. sapiens and/or H. erectus.
Brown et al. (2004) initially hypothesized that the rst hominin Our analyses support two positions for H. oresiensis, both
immigrants on Flores had a similar morphology and body size to consistent with the broad hypothesis that it is descended from an
H. erectus but subsequently dwarfed. When new skeletal remains early Homo lineage (rather than H. erectus):
comprising a second mandible, a radius, a tibia, and the right hu-
merus and ulna of LB1 were described, however, Morwood et al. i) It is sister taxon to H. habilis: they share a common ancestor that
(2005) concluded from the new evidence that the ancestors of is not shared with any other species in the analysis. In all
H. oresiensis could not be attributed to H. erectus. Argue et al. parsimony and Bayesian analyses, the terminal branches leading
(2006) later independently supported this conclusion (and see to H. habilis and to H. oresiensis are substantial, suggesting that
Larson et al., 2007; Tocheri et al., 2007; Argue et al., 2009; Baab and the two species are reciprocally monophyletic, rather than
McNulty, 2009; Brown and Maeda, 2009; Brown, 2012; Jungers having any ancestor-descendent relationship. If so, both lineages
et al., 2009b). On the other hand, Kaifu et al. (2011) described and must be at least as old as the H. habilis lineage, the earliest ev-
compared the cranium of H. oresiensis with a large sample of Homo idence for which is represented by the KNM-ER 1501 mandible
taxa, explored shape and form differences between them, and dated to 1.75 Mya (Gathogo and Brown, 2006; see also; Fiebel
concluded that the cranio-facial morphology of LB1 was consistent et al., 1989).
with a hypothesis that H. oresiensis was derived from Javanese ii) It is basal taxon to a clade consisting of at least H. habilis,
H. erectus. Kubo et al. (2013) supported the island dwarng hy- H. erectus, H. ergaster, and H. sapiens. In this case, the
pothesis for H. oresiensis following their recalculation of the H. oresiensis lineage must be even older than the oldest of these
endocranial volume (ECV) of LB1 from 417 cc (Falk et al., 2005) to species, that is, older than the H. habilis lineage. The minimum
426 cc (Kubu et al., 2013) using micro-CT imaging, although they
noted that under this hypothesis some degree of brain size reduc-
tion still remained to be explained by factors other than body size
scaling. 8
Kaifu et al., 2015: Figs 4A, 4C, 4D, 5A, 5C, 6C.
Kaifu et al. (2015; see also Van den Bergh et al., 2016) rejected a 9
LB1 P3; LB15 P4; LB1 upper and lower rst molars; LB1 and LB6 M1; LB1 M1;
phylogenetic relationship between H. oresiensis and early hom- LB2 M2.
10
inins such as H. habilis from their comparative phenetic (not Kaifu et al., 2015:Figs. 4A, C, D, 5A, C, 6C.
age for the divergence of H. oresiensis under this hypothesis A. africanus. The remote phylogenetic position we obtain for
would be > 1.75 Ma. H. oresiensis (relative to H. sapiens) is not due to pathology in a
population of anatomically modern humans or traits correlated
Thus, H. oresiensis is either closely related to H. habilis or is at with Down syndrome, a conclusion supported by recent fossils
least at a similar grade. demonstrating the longevity of the H. oresiensis morphotype. Our
Both the phylogenetic positions imply an Early Pleistocene phylogenetic results imply a long (1.75 Ma) ghost lineage for
appearance for the H. oresiensis lineage. The only physical evi- H. oresiensis, which furthermore likely originates in Africa.
dence for the H. oresiensis lineage is from Liang Bua cave, Flores,
dated at ~86,000 years ago (Sutikna et al., 2016), and now, probably, Acknowledgements
Mata Menge, dated at ~700,000 years ago (Brumm et al., 2016; Van
den Bergh et al., 2016). Most early fossil hominins (including We thank the Australian Research Council for providing funding
H. habilis) from the Early Pleistocene are from Africa; one species, under the Discovery Project Scheme (DP1096870). We sincerely
H. georgicus, is known from Eurasia, but our analysis indicates that thank the following people and institutions for providing access to
there is no link between this taxon and H. oresiensis. We thus fossils under their curation: Dr. Berhane Asfaw, National Museum
anticipate that Africa was the origin of the H. oresiensis lineage. If of Ethiopia; Dr. Fachroel Aziz, Centre for Geological Survey, Ban-
so, this lineage must have migrated from Africa at least once at any dung, Indonesia; Professor Lee Berger, University of Witwatersrand,
time since the Early Pleistocene until ~700 ka (when it potentially South Africa; Dr. Tony Djubiantono, ARKENAS, Indonesia; Mr.
rst appears on Flores). Tomas Dnyango Mukhuyu and Yared, Curators, National Museum of
A small number of hominins attributed to Homo have been Ethiopia; Dr. Denise Donlon, Shellshear Museum, University of
identied in the fossil record that pre-date H. habilis. The rst of Sydney, Australia; Dr. Andrew Gallagher, School of Anatomical
these to be discovered was a fossil hominin temporal bone, KNM- Sciences, University of Witwatersrand; Dr. Sandy Ingleby, Austra-
BC1, found in 1965 by John Kimengich in the Chemeron Forma- lian Museum, Sydney, Australia; Ms. Bonita de Klerk, University of
tion in Kenya (Martyn, 1967) that is described by P.V. Tobias in the Witwatersrand, South Africa; Dr. Ottmar Kullmer, Research Insti-
succeeding communication in Martyn (Tobias, 1967). It is dated to tute Senckenberg, Frankfurt, Germany; Dr. A. Kwekason, National
2.4 mya (Hill et al., 1992) and has a number of features of Homo. Hill Museum and House of Culture, Dar es Salaam, Tanzania; Ms. Wilma
et al.'s (1992) comparative morphological analyses showed differ- Lawrence, University of Witwatersrand, South Africa; Dr. Emma
ences in functional anatomy of the temporomandibular joint from Mbua, Kenya National Museums, Nairobi, Kenya; the late Professor
Australopithecus, including indications of brain expansion. It could Mike Morwood, University of Wollongong, Australia; Dr. Paul
not be attributed to H. erectus or H. sapiens, and Hill et al. (1992) Msemwa, Director, National Museum and House of Culture, Dar es
attribute it to Homo sp. indet. The hominin maxilla and dentition Salaam, Tanzania; Dr. Christine Ngereza, National Museum and
A.L. 666-1 discovered in the Hadar region of Ethiopia is dated to House of Culture, Dar es Salaam, Tanzania; Ms. Stephany Potze,
2.33 0.07 mya (the Hadar A. afarensis are earlier, 3.4e3.0 mya; Ditsong Museum, Pretoria, South Africa; Dr. Friedemann Schrenk,
Kimbel et al., 1997). The maxilla shows afnities with Homo, and Research Institute Senckenberg, Frankfurt, Germany; Mr. Thomas
Kimbel et al. (1997) attribute A.L. 666-1 to Homo aff. H. habilis, Sutikna, ARKENAS, Indonesia; Dr. Jon de Vos, Naturalis, Leiden,
perhaps representing a male of the species. An innominate, ER 3228 Holland; Dr. Harry Widianto, Sangiran Museum, Indonesia; and Dr.
from Koobi Fora dated as 1.95 mya, is generally deemed erectine Bernard Zipfel, University of Witwatersrand, Johannesburg, South
grade, but in any case indicates that this lineage had already Africa. Several anonymous reviewers provided most constructive
separated; and the dates of some other Koobi Fora fossils may also comments on earlier versions of this ms, which we greatly appre-
be of this age. Recently, Villmoare et al. (2015) have reported the ciated. We thank Erik Seiffert, Stony Brook University, for his input
discovery in the Ledi-Geraru region of Ethiopia of a partial hominin to an earlier version of this ms and Ms. Christine Cave, School of
mandible dated to 2.80e2.75 mya. The mandible combines aus- Archaeology and Anthropology, Australian National University, for
tralopith traits with more derived characters observed in later technical support.
Homo, and Villmoare et al. (2015) conclude that it is a likely pre-
decessor to early Pleistocene Homo. If our second hypothesis is
Supplementary Online Material
correct, it is possible that taxa on the branch leading to
H. oresiensis and H. habilis may yet be identied among these >2
Supplementary online material related to this article can be
million year old Homo, or among future discoveries of such early
found at http://dx.doi.org/10.1016/j.jhevol.2017.02.006.
Homo.
References
5. Conclusion
Aberer, A.J., Krompass, D., Stamatakis, A., 2013. Pruning rogue taxa improves
We use a dataset comprising 50 cranial, 26 mandibular, 24 phylogenetic accuracy: an efcient algorithm and webservice. Syst. Biol. 62,
dental, and 33 postcranial characters to infer the relationships of 162e166. http://dx.doi.org/10.1093/sysbio/sys078.
Aiello, L.C., 2010. Five years of Homo oresiensis. Am. J. Phys. Anthropol. 142,
H. oresiensis and test two competing hypotheses: H. oresiensis is a 167e179.
late survivor of an early hominin lineage or is a descendant of Aiello, L.C., Dean, C., 1990. An Introduction to Human Evolutionary Anatomy. Aca-
H. erectus. We hypothesize that H. oresiensis either shared a demic Press, London.
Alcover, J.-A., 1976. L'evolucio de Myotragus Bate 1909 (Artiodactyla, Rupicaprini), un
common ancestor with H. habilis or represents a sister group to a Proce s biologic lligat al fenomen de la insularitat. (The evolution of Myotragus
clade consisting of at least H. habilis, H. erectus, H. ergaster, and Bate 1909 (Artiodactyla, Rupicaprini), a biological process linked to the insularity
H. sapiens. A close relationship between H. erectus and H. oresiensis phenomenon). Bull. Inst. Cat. Hist. Nat. 40, 59e94.
Andrews, P., 1984. On the characters that dene Homo erectus. In: Andrews, P.,
is rejected, which contradicts the proposal that island dwarng of
Franzen, J.L. (Eds.), The Early Evolution of Man with Special Emphasis on
Asian H. erectus led to H. oresiensis. The hypothesis that Southeast Asia and Africa. Cour. Forsch.-Inst. Senckenberg 69, Frankfurt,
H. oresiensis represents a sister clade to all the above taxa except 197e179.
Anto n, S.C., Potts, R., Aiello, L.C., 2014. Evolution of early Homo: An integrated
H. habilis cannot be rejected, however. Additionally, we can reject
biological perspective. Science 345, 45e59.
close phylogenetic relationships for H. oresiensis with H. sapiens, Argue, D., Donlon, D., Groves, C., Wright, R., 2006. Homo oresiensis: Microcephalic,
H. ergaster, H. georgicus, H. naledi, A. afarensis, A. sediba, and pygmoid, Australopithecus or Homo? J. Hum. Evol. 51, 360e374.
Argue, D., Morwood, M., Sutikna, T., Jatmiko, Saptomo, W.E., 2009. Homo or- Finarelli, J.A., Clyde, W.C., 2004. Reassessing hominoid phylogeny: evaluating
esiensis: a cladistic analysis. J. Hum. Evol. 57, 623e629. congruence in the morphological and temporal data. Paleobiology 30, 614e651.
Ashton, E.H., Zuckerman, S., 1956. Cranial crests in the Anthropoidea. Proc. Zool. Soc. Foster, J.B., 1964. Evolution of mammals on islands. Nature 202, 234e235.
Lond. 126, 581e634. Gabunia, L., Vekua, A., Lordkipanidze, D., Swisher, C., 2000. Earliest Pleistocene
Baab, K.L., McNulty, K.P., 2009. Size, shape and symmetry in fossil hominins: The hominid cranial remains from Dmanisi Republic of Georgia: taxonomy,
status of the LB1 cranium based on 3D morphometric analyses. J. Hum. Evol. 57, geological setting and age. Science 288, 1019e1025.
608e622. Gathogo, P.N., Brown, F.H., 2006. Revised stratigraphy of Area 123, Koobi Fora,
Baab, K.L., McNulty, K.P., Harvati, K., 2013. Homo oresiensis contextualized: a Kenya, and new age estimates of its fossil mammals, including hominins.
geometric morphometric comparative analysis of fossil and pathological hu- J. Hum. Evol. 51, 471e479.
man samples. PLoS One 8(7), 1e11. Godfrey, L.R., Sutherland, M., 1996. Paradox of peramorphic peramorphosis: het-
Baab, K.L., Brown, P., Falk, D., Richtsmeier, J.T., Hildebolt, C.F., Smith, K., Jungers, W., erochrony and human evolution. Am. J. Phys. Anthropol. 99, 17e42.
2016. A critical evaluation of the Down syndrome diagnosis for LB1, type Gordon, A.D., Nevell, L., Wood, B., 2008. The Homo oresiensis cranium (LB1): Size,
specimen of Homo oresiensis. PLoS One 11(6), e0155731. http://dx.doi.org/ scaling, and early Homo afnities. Proc. Natl. Acad. Sci. USA 105, 4650e4655.
10.1371/journal.pone.0155731. Grabowski, M., Hatala, K.G., Jungers, W.L., Richmond, B.G., 2015. Body mass esti-
Begun, D.R., Ward, C.V., Rose, M.D., 1997. Events in hominoid evolution. In: mates of hominin fossils and the evolution of human body size. J. Hum. Evol. 83,
Begun, D.R., Ward, C.V., Rose, M.D. (Eds.), Function, Phylogeny, and Fossils. 75e93.
Plenum Press, New York, pp. 389e417. Groves, C.P., 1989. A Theory of Human and Primate Evolution. Clarendon Press,
Berger, L.R., de Ruiter, D.J., Churchill, S.E., Schmidt, P., Carlson, K.J., Dirks, P.H.G.M., Oxford.
Kibii, J.M., 2010a. Australopithecus sediba: a new species of Homo-like austral- Groves, C.P., 2007. The Homo oresiensis controversy. Hayati Journal of Biosciences
opith from South Africa. Science 328, 195e205. 14, 123e126.
Berger, L.R., de Ruiter, D.J., Churchill, S.E., Schmid, P., Carlson, K.J., Dirks, P.H.G.M., Haeusler, M., McHenry, H.M., 2004. Body proportions of Homo habilis reviewed.
Kibii, J.M., 2010b. Supporting Online Material for Australopithecus sediba: A New J. Hum. Evol. 46, 433e465.
Species of Homo-like Australopith from South Africa. http://science.sciencemag. Harrison, T., 2011. Hominins from the Upper Laetoli and Upper Ndolanya Beds,
org/content/sci/suppl/2010/04/08/328.5975. 195.DC1/Berger.SOM.pdf. Laetoli. In: Harrison, T. (Ed.), Paleontology and Geology of Laetoli: Human
Berger, L.R., Hawks, J., de Ruiter, D.J., Churchill, S.E., Schmid, P., Delezene, L.K., Evolution in Context. Springer, Dordrecht, pp. 141e188.
Kivell, T.L., Garvin, H.M., Williams, S.A., DeSilva, J.M., Skinner, M.M., Heaney, L.R., 1978. Island area and body size of insular mammals: evidence from the
Musiba, C.M., Cameron, N., Holliday, T.W., Harcourt-Smith, W., Ackermann, R.R., Tri-colored Squirrel (Callosciurus prevosti) of South East Asia. Evolution 32,
Bastir, M., Bogin, B., Bolter, D., Brophy, J., Cofran, Z.D., Congdon, K.A., Deane, A.S., 29e44.
Dembo, M., Drapeau, M., Elliott, M., Feuerriegel, E.M., Garcia-Martinez, D., Henneberg, M., Thorne, A., 2004. Flores human may be a pathological Homo sapiens.
Green, D.J., Gurtov, A., Irish, J.D., Kruger, A., Laird, M.F., Marchi, D., Meyer, M.R., Before Farming 4, 2e4.
Nalla, S., Negash, E.W., Orr, C.M., Radov ci
c, D., Schroeder, L., Scott, J.E., Henneberg, M., Eckhardt, R.B., Chavanaves, S., Hsu, K.J., 2014. Evolved develop-
Throckmorton, Z., Tocheri, M.W., VanSickle, C., Walker, C.S., Wei, P., Zipfel, B., mental homeostasis disturbed in LB1 from Flores, Indonesia denotes Down
2015. Homo naledi, a new species of the genus Homo from the Dinaledi syndrome and not diagnostic traits of the invalid species Homo oresiensis. Proc.
Chamber, South Africa. eLife 2015 4, e09560. http://dx.doi.org/10.7554/ Natl. Acad. Sci. 111(33), 1967e11972.
eLife.09560, 1e35. Hershkovitz, I., Kornreich, L., Laron, Z., 2007. Comparative skeletal features between
Bermdez de Castro, J.M., Martino n-Torres, M., Sier, M.J., Martin-France s, L., 2014. Homo oresiensis and patients with primary growth hormone insensitivity
On the variability of the Dmanisi mandibles. PLoS One 9(2), 1e15. (Laron Syndrome). Am. J. Phys. Anthropol. 134, 198e208.
Braga, J., Crubezy, E., Elyaqtine, M., 1996. The posterior border of the sphenoid Hill, A., Ward, S., Curtis, G., Drake, R., 1992. Earliest Homo. Nature 355, 719e722.
greater wing and its phylogenetic usefulness in human evolution. Am. J. Phys. Hillis, D.M., Bull, J.J., 1993. An empirical test of bootstrapping as a method for
Anthropol. 107, 387e399. assessing condence in phylogenetic analysis. Syst. Biol. 42(2), 182e192.
uer, G., Schultz, M., 1996. The morphological afnities of the Plio-Pleistocene
Bra Holliday, T.W., Franciscus, R.G., 2009. Body size and its consequences: allometry and
mandible from Dmanisi, Georgia. J. Hum. Evol 30, 445e481. the lower limb length of Liang Bua 1 (Homo oresiensis). J. Hum. Evol. 57,
Bremer, K., 1994. Branch support and tree stability. Cladistics 10, 295e304. 223e228. http://dx.doi.org/10.1016/j.jhevol.2009.04.007.
Brown, P., 2012. LB1 and LB6 Homo oresiensis are not modern human (Homo sa- Irish, J.D., Guatelli-Steinberg, D., Legge, S.S., De Ruiter, D.J., Berger, L.R., 2013. Dental
piens) cretins. J. Hum. Evol. 62, 201e224. morphology and the phylogenetic Place of Australopithecus sediba. Science 340,
Brown, P., Maeda, T., 2009. Liang Bua Homo oresiensis mandibles and mandibular 1233062-1e1233062-4.
teeth: a contribution to the comparative morphology of a new hominin species. Jacob, T., Indriati, E., Soejono, R.P., Hs, K., Frayer, D.W., Eckhardt, R.B.,
J. Hum. Evol. 57, 571e596. Kuperavage, A.J., Thorne, A., 2006. Pygmoid Austromelanesian Homo sapiens
Brown, P., Sutikna, T., Morwood, M.J., Soejono, R.P., Jatmiko, Saptomo, W.E., skeletal remains from Liang Bua, Flores: population afnities and pathological
Awe, R.D., 2004. A new small-bodied hominin from the Late Pleistocene of abnormalities. Proc. Natl. Acad. Sci. 103(36), 13421e13426.
Flores, Indonesia. Nature 431, 1055e1061. Johanson, D.C., Masaw, F.T., Eck, G.G., White, T.D., Walter, R.C., Kimbel, W.H.,
Brumm, A., van den Bergh, G.D., Storey, M., Kurniawan, I., Alloway, B.V., Setiawan, R., Asfar, B., Manega, P., Ndessokia, P., Sewa, G., 1987. New partial skeleton of Homo
Setiyabuti, E., Grn, R., Moore, M.W., Puspaningrum, M.R., Wibowo, U.P., habilis from Olduvai gorge, Tanzania. Nature 327, 205e209.
Insani, H., Sutisna, I., Westgate, J.A., Pearce, N.J.G., Duval, M., Meijer, H.J.M., Johanson, D.C., Taieb, M., Coppens, Y., 1982. Pliocene Hominids from the Hadar
Aziz, F., Sutikna, T., van der Kaars, S., Flude, S., Morwood, M.J., 2016. Age and Formation, Ethiopia (1973e1977): stratigraphic, chronologic, and paleoenvir-
context of the oldest known hominin fossils from Flores. Nature 534, 249e253. onmental contexts, with notes on Hominid morphology and systematics. Am. J.
Curnoe, D., 2010. A review of early Homo in southern Africa focusing on cranial, Phys. Anthropol. 57(4), 373e402.
mandibular and dental remains, with the description of a new species (Homo Johanson, D.C., White, T.D., Coppens, Y., 1978. A new species of the genus Austral-
gautengensis sp. nov.). Homo - Journal of Comparative Human Biology 61, opithecus (Primates:Hominidae) from the Pliocene of eastern Africa. Kirtlandia
151e177. 28, 1e14.
Day, M.H., Wood, B.A., 1968. Functional afnities of the Olduvai Hominid 8 talus. Jungers, W.L., 2009. Interlimb proportions in humans and fossil hominins: vari-
Man 3, 440e455. ability and scaling. In: Grine, F.E., Leakey, R.E., Fleagle, J.G. (Eds.), The First
Dayan, T., Simberloff, S., 1998. Size patterns among competitors: ecological char- Humans: Origins of the Genus Homo. Springer, Dordrecht, pp. 93e98.
acter displacement and character release in mammals, with special reference to Jungers, W.L., 2013. Homo oresiensis. In: Begun, D.R. (Ed.), Companion to Paleo-
island populations. Mammal Rev. 28, 99e124. anthropology. Wiley-Blackwell, Malden, pp. 582e598.
Dembo, M., Matzke, N.J., Mooers, A.., Collard, M., 2015. Bayesian analysis of a Jungers, W.L., Larson, S., Harcourt-Smith, G.W., Morwood, M.J., Sutikna, T., Awe, R.D.,
morphological supermatrix sheds light on controversial fossil hominin re- Djubiantono, T., 2009a. Descriptions of the lower limb skeleton of Homo or-
lationships. Proc. R. Soc. B 282, 20150943. http://dx.doi.org/10.1098/ esiensis. J. Hum. Evol. 57, 538e554.
rspb.2015.0943. Jungers, W.L., Harcourt-Smith, W.E.H., Wunderlich, R.E., Tocheri, M.W., Larson, S.G.,
Falk, D., Hildebolt, C., Smith, K., Morwood, M., Sutikna, T., Brown, P., Jatmiko, Sutikna, T., Awe, R.D., Morwood, M.J., 2009b. The foot of Homo oresiensis Na-
Saptomo, W.E., Brunsden, B., Prior, F., 2005. The brain of Homo oresiensis. ture, 458, 81e84.
Science 308, 242e245. Jungers, W.L., Grabowski, M., Hatala, K.G., Richmond, B.G., 2016. The evolution
Falk, D., Hilderbolt, C., Smith, K., Morwood, M.J., Sutikna, T., Jatmiko, W., of body size and shape in the human career. Phil. Trans. R. Soc. B 371,
Saptomo, W.E., Imhof, H., Seidler, H., Prior, F., 2007. Brain shape in human mi- 21050247.
crocephalics and Homo oresiensis. Proc. Natl. Acad. Sci. 104(6), 2513e2518. Kaifu, Y., Baba, H., Aziz, F., Indriati, E., Schrenk, F., Jacob, T., 2005. Taxonomic af-
Falk, D., Hildebolt, C., Smith, K., Morwood, M.J., Sutikna, T., Saptomo, W.E., Prior, F., nities and evolutionary history of the Early Pleistocene Hominins of Java:
2009. LB1's virtual endocast, microcephaly, and hominin brain evolution. dentognathic evidence. Am. J. Phys. Anthropol. 128, 709e726.
J. Hum. Evol. 57, 597e607. Kaifu, Y., Baba, H., Sutikna, T., Morwood, M.J., Kubo, D., Saptomo, W.E., Jatmiko, Awe, R.D.,
Felsenstein, J., 1985. Condence limits on phylogenies: an approach using the Djubiantono, T., 2011. Craniofacial morphology of Homo oresiensis: Description,
bootstrap. Evolution 39, 783e791. taxonomic afnities, and evolutionary implication. J. Hum. Evol. 61, 664e682.
Fiebel, C.S., Brown, F.H., McDougall, I., 1989. Stratigraphic context of fossil Hominids Kaifu, Y., Kono, R.T., Sutikna, T., Saptomo, W.E., Awe, R.D., 2015. Unique dental
from the Omo group deposits, Northern Turkana Basin, Kenya and Ethiopia. Am. morphology of Homo oresiensis and its evolutionary implications. PLoS One
J. Phys. Anthropol. 78, 595e622. 10(11), e0141614. http://dx.doi.org/10.1371/journal.pone.0141614.
Kimbel, W., White, T.D., 1988. Variation, sexual dimorphism and the taxonomy of Oxnard, C., Obendorf, P.J., Kefford, B.J., Dennison, J., 2012. More on the Liang Bua and
Australopithecus. In: Grine, F.E. (Ed.), Evolutionary History of the Robust Aus- modern human cretins. Homo - Journal of Comparative Human Biology 63,
tralopithecines. Aldine de Gruyter, New York, pp. 175e193. 407e412.
Kimbel, W., Rak, Y., Johanson, D.C., 2004. The Skull of Australopithecus afarensis. Poe, S., Wiens, J.J., 2000. Character selection and the methodology of morphological
Oxford University Press, Oxford. phylogenetics. In: Wiens, J.J. (Ed.), Phylogenetic Analysis of Morphological Data.
Kimbel, W.H., Johanson, D.C., Rak, Y., 1997. Systematic assessment of a maxilla of Smithsonian Institution Press, Washington, p. 29.
Homo from Hadar, Ethiopia. Am. J. Phys. Anthropol. 103, 235e262. Pontzer, H., Rolian, C., Rightmire, G.P., Jashashvili, T., Ponce de Leo n, M.S.,
Kohler, M., Moy a-Sola, S., 2004. Reduction in brain and sense organs in the fossil Lordkipanidze, D., Zollikofer, C.P.E., 2010. Locomotor anatomy and biomechanics
insular bovid Myotragus. Brain Behav. Evol. 63, 125e140. of the Dmanisi hominins. J. Hum. Evol. 58, 492e504.
Kramer, A., 1994. A critical analysis of claims for the existence of Southeast Asian Rak, Y., 1983. The Australopithecine Face. Academic Press, New York.
Australopithecines. J. Hum. Evol. 26, 3e21. Rak, Y., Ginsburg, A., Geffen, E., 2002. Does Homo neanderthalensis play a role in
Kubo, D., Kono, R.T., Kaifu, Y., 2013. Brain size of Homo oresiensis and its evolu- modern human ancestry? The mandibular evidence. Am. J. Phys. Anthropol. 119,
tionary implications. Proc. R. Soc. Lond. B 280, 1e8. 199e204.
Larson, S.G., Jungers, W.L., Morwood, M.J., Sutikna, T., Saptomo, W.E., Awe, R.D., Richards, G.D., 2006. Genetic, physiologic and ecogeographic factors contributing to
Djubiantono, T., 2007. Homo oresiensis and the evolution of the hominin variation in Homo sapiens: Homo oresiensis reconsidered. J. Evol. Biol. 19(6),
shoulder. J. Hum. Evol. 53, 718e731. 1744e1767.
Larson, S.G., Jungers, W.L., Tocheri, M.W., Orr, C.M., Morwood, M.J., Sutikna, T., Rightmire, G.P., Lordkipanidze, D., Vekua, A., 2006. Anatomical descriptions,
Saptomo, W.E., Awe, R.D., Djubiantono, T., 2009. Descriptions of the upper limb comparative studies and evolutionary signicance of the hominin skulls from
skeleton of Homo oresiensis. J. Hum. Evol. 57, 555e570. Dmanisi, Republic of Georgia. J. Hum. Evol. 50, 115e141.
Lawlor, T.L., 1982. The evolution of body size in mammals: evidence from insular Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Ho hna, S.,
populations in Mexico. Am. Nat. 119, 54e72. Larget, B., Liu, L., Suchard, M.A., Huelsenbeck, J.P., 2012. MrBayes 3.2: efcient
Leakey, L.S.B., Tobias, P.V., Napier, J.R., 1964. A new species of the genus Homo from Bayesian phylogenetic inference and model choice across a large model space.
Olduvai Gorge. Nature 202, 7e9. Syst. Biol. 61(3), 539e542.
Lee, M.S.Y., 2000. Tree robustness and clade signicance. Syst. Biol. 49, 829e836. Rosas, A., 2001. Occurrence of Neanderthal features in mandibles from the Ata-
Libois, R., Fons, R., Bordenave, D., 1993. Mediterranean small mammals and insular puerca SH Site. Am. J. Phys. Anthropol. 114, 74e91.
syndrome: biometric study of the long-tailed eld mouse (Apodemus sylvaticus) Rosas, A., de Castro, B.J., 1998. On the taxonomic afnities of the Dmanisi Mandible
(Rodentia-Muridae) of Corsica. Bonn. Zool. Beitr. 44, 147e163. (Georgia). Am. J. Phys. Anthropol. 107, 145e162.
Lieberman, D.E., Wood, B.A., Pilbeam, D.R., 1996. Homoplasy and early Homo: an Schroeder, L., Scott, J.E., Garvin, H.M.M., Laird, M.F., Dembo, M., Radov ci
c, D.,
analysis of the evolutionary relationships of H. habilis sensu stricto and Berger, L.R., de Ruiter, D.J., Ackermann, R.R., 2017. Skull diversity in the Homo
H. rudolfensis. J. Hum. Evol. 30, 97e120. lineage and the relative position of Homo naledi. J. Hum. Evol. 104, 124e135.
Lordkipanidze, D., Vekua, A., Ferring, R., Rightmire, G.P., Zollikofer, P.E., Ponce de http://dx.doi.org/10.1016/j.jhevol.2016.09.014.
Leo n, M., Agusti, J., Kiladze, G., Mouskhelishvili, A., Nioradze, M., Tappen, M., Schwartz, J.H., Tattersall, I., 2002. The Human Fossil Record, Volume Two. Cranio-
2006. A fourth hominin skull from Dmanisi, Georgia. Anat. Rec. Part A 288A, dental Morphology of the Genus Homo (Africa and Asia). Wiley-Liss, USA,
1146e1157. p. 488.
Lordkipanidze, D., Jashashvili, T., Vekua, A., Ponce de Leo n, M.S., Zollikoffer, C.P.E., Skelton, R.R., McHenry, H.M., 1992. Evolutionary relationships among early homi-
Rightmire, G.P., Pontzer, H., Ferring, R., Oms, O., Tappen, M., Bukhsianidze, M., nids. J. Hum. Evol. 23, 309e349.
Agusti, J., Kahike, R., Kiladze, G., Martinez-Navarro, B., Mouskhelishvili, A., Smith, H.F., Grine, F.E., 2008. Cladistic analysis of early Homo crania from Swartk-
Nioradze, M., Rook, L., 2007. Postcranial evidence from early Homo form rans and Sterkfontein, South Africa. J. Hum. Evol. 54, 684e704.
Dmanisi, Georgia. Nature 449, 305e310. Sondaar, P.Y., 1977. Patterns in evolutionary changes and their possible meaning. In:
Lordkipanidze, D., Ponce de Leo n, M.S., Margvelashvili, A., Rak, Y., Rightmire, G.P., Hecht, M.K., Goody, P.C., Hecht, B. (Eds.), Major Patterns in Vertebrate Evolution.
Vekua, A., Zollikofer, C.P.E., 2013. A complete skull from Dmanisi, Georgia, and Plenum Press, New York.
the evolutionary biology of early Homo. Science 342, 326e331. Sorenson, M.D., Franzosa, E.A., 2007. TreeRot, version 3. Boston University, Boston.
Lyras, G.A., Dermitzakas, M.A., Van de Geer, A.A.E., Van de Geer, S.B., De Vos, J., 2009. Strait, D.S., Grine, F.E., 2004. Inferring hominoid and early hominid phylogeny
The origin of Homo oresiensis and its relation to evolutionary processes under using craniodental characters: the role of fossil taxa. J. Hum. Evol. 47,
isolation. J. Anthropol. Sci. 117, 33e43. 399e452.
Manzi, G., Mallegni, F., Ascenzi, A., 2001. A cranium for the earliest Europeans: Sutikna, T., Tocheri, M.W., Morwood, M.J., Saptomo, W.E., Jatmiko, Awe, R.D., Sri
phylogenetic position of the hominid from Ceprano, Italy. Proc. Natl. Acad. Sci. Wasisto, Westaway, K.E., Aubert, M., Li, B., Zhao, J.-X., Storey, M., Alloway, B.V.,
USA 98, 10011e10016. Morley, M.W., Hanneke, J.M., Meijer, M., van den Bergh, G.D., Grn, R.,
Martin, R.D., MacLarnon, A.M., Phillips, J.L., Dobyns, W.B., 2006. Flores hominid: Dosseto, A., Brumm, A., Jungers, W.L., Roberts, R.G., 2016. Revised stratigraphy
new species or microcephalic dwarf? Anat. Rec. Part A 11, 1123e1145. and chronology for Homo oresiensis at Liang Bua in Indonesia. Nature
Martinez, I., Arsuaga, J.L., 1997. The temporal bones from Sima de los Huesos Middle 532(7599), 366e369.
Pleistocene site (Sierra de Atapuerca, Spain). A phylogenetic approach. J. Hum. Swofford, D.L., 2003. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other
Evol. 23, 283e318. Methods). Version 4. Sinauer Associates, Sunderland, Massachusetts.
Martino n-Torres, M., de Castro, B.J., Go mez-Robles, A., Margvelashvili, A., Prado, L., Templeton, A.R., 1983. Phylogenetic inference from restriction endonuclease
Lordkipanidze, D., Vekua, A., 2008. Dental remains from Dmanisi (Republic of cleavage site maps with particular reference to the evolution of humans and the
Georgia): morphological analysis and comparative study. J. Hum Evol. 55, apes. Evolution 37, 221e244.
249e273. Tobias, P.V., 1967. Pleistocene deposits and new fossil localities in Kenya. Nature
Martyn, J., 1967. Pleistocene deposits and new fossil localities in Kenya. Nature 215, 215, 479.
476e479. Tobias, P.V., 1991. The skulls, endocasts and teeth of Homo habilis. Summary of
McHenry, H.M., 1991. Femoral lengths and stature in Plio-Pleistocene hominids. Am. cranial, mandibular, encephalic and dental features of Homo habilis, in com-
J. Phys. Anthropol. 851, 149e158. parison with those of some other hominid taxa. In: Tobias, P.V. (Ed.), Olduvai
McHenry, H.M., 1996. Homoplasy, clades, and hominid phylogeny. In: Meikle, F., Gorge, Volume 4. Parts V-IX. Cambridge University Press, Cambridge.
Clark Howell, N.G. (Eds.), Contemporary Issues in Human Evolution. California Tocheri, M., Orr, C.M., Larson, S.G., Sutikna, T., Jatmiko, Saptomo, W.E., Awe, R.D.,
Academy of Sciences Memoir 21, San Francisco, pp. 77e92. Djubiantono, T., Morwood, M.J., Jungers, W.L., 2007. The primitive wrist bone of
Meiri, S., Cooper, N., Purvis, A., 2008. The island rule: made to be broken? Proc. R. Homo oresiensis and its implications for hominin evolution. Science 317,
Soc. Lond. B 275, 141e148. 1743e1745.
Melton, R.H., 1982. Body size and island Peromyscus: a pattern and a hypothesis. Van Arsdale, A., 2006. Mandibular Variation in early Homo from Dmanisi, Georgia.
Evol. Theory 6, 113e126. Ph.D. Dissertation, University of Michigan.
Millien, V., 2006. Morphological evolution is accelerated among island mammals. Van den Bergh, G.D., Kaifu, Y., Kurniawan, I., Kono, R.T., Brumm, A., Setiyabudi, E.,
PLoS Biol. 4 (10), e321. http://dx.doi.org/10.1371/journal.pbio.0040321. Aziz, F., Morwood, M., 2016. Homo oresiensis-like fossils from the early Middle
Morwood, M.J., Brown, P., Jatmiko, Sutikna, T., Saptomo, W.E., Westaway, K.E., Pleistocene of Flores. Nature 534, 245e248.
Awe, R.D., Roberts, R.G., Maeda, T., Wasisto, S., Djubiantono, T., 2005. Further Van Heteren, A.H., 2008. Homo oresiensis as an island form. PalArch's J. Vert.
evidence for small-bodied hominins from the late Pleistocene of Flores, Palaeontol. 5(2), 1e12.
Indonesia. Nature 437, 1012e1017. Van Heteren, A.H., 2012. Les hominins of Flores: insular adaptations of the lower
Mounier, A., Marchal, F., Condema, S., 2009. Is Homo heidelbergensis a distinct body. C.R. Pale vol 11, 169e179.
species? New insight on the Mauer mandible. J. Hum. Evol. 56, 219e246. Van Heteren, A.H., 2013. Is LB1 diseased or modern? A review of the proposed
Niven, J.E., 2007. Response to Ko hler et al.: Impossible arguments about possible pathologies. Gene 528(1), 12e20.
species? Trends in Ecol. Evol. 23, 8. Villmoare, B., Kimble, W.H., Seyoum, C., Campisano, C.J., DiMaggio, E., Rowan, J.,
Obendorf, P.J., Oxnard, C.E., Kefford, B.J., 2008. Are the small human-like fossils Braun, D.R., Arrowsmith, J.R., Reed, K.E., 2015. Early Homo at 2.8 Ma from Ledi-
found on Flores human endemic cretins? Proc. R. Soc. Series B 275(1640), Geraru, Afar, Ethiopia. Science 347, 1352e1355.
1287e1296. Wassersug, R.J., Yang, H., Sepkoski, J.J., Raup, D.M., 1979. The evolution of body size
Oxnard, C., Obendorf, P.J., Kefford, B.J., 2010. Post-cranial skeletons of hypothyroid on islands: a computer simulation. Am. Nat. 114, 287e295.
cretins show a similar anotomical mosaic as Homo oresiensis. PLoS One 5 (9), Weidenreich, F., 1943. The Skull of Sinanthropus pekinensis. A comparative study on a
e13018. http://dx.doi.org/10.1371/journal.pone.0013018. primitive hominid skull. Geological Survey of China, Pehpei, Chungking, China.
Wiens, J.J., 1995. Polymorphic characters in phylogenetic systematics. Syst. Biol. 44, Wood, B.A., Richmond, B.G., 2000. Human evolution: taxonomy and paleobiology.
482e500. J. Anat. 196, 19e60.
Wiens, J.J. (Ed.), 2000. Phylogenetic Analysis of Morphological Data. Smithsonian Wood, B.A., Abbott, S.A., Graham, S.H., 1983. Analysis of the dental morphology of
Institution Press, Washington, D.C. Plio-Pleistocene hominids. II. Mandibular molarsdstudy of cusp areas, ssure
White, T., 2010. Human evolution: how has Darwin done? In: Bell, M.A., pattern and cross sectional shape of the crown. J. Anat. 137(2), 287e314.
Futuyma, D.J., Eanes, W.F., Levinton, J.S. (Eds.), Evolution Since DarwindThe Wood, B.A., Abbott, A., Uytterschaut, H., 1988. Analysis of the dental morphology of
First 150 Years. Sinauer Associates, Sunderland, p. 553. Plio-Pleistocene hominids. IV. Mandibular postcanine root morphology. J. Anat.
Wood, B., 1991. Koobi Fora Research Project. Hominid Cranial Remains. In: 156, 107e139.
Leakey, R.E. (Ed.), Koobi Fora Researches into Geology, Palaeontology, and Hu- Xie, W., Lewis, P.O., Fan, Y., Kuo, L., Chen, M.H., 2011. Improving marginal likelihood
man Origins, Volume 4. Clarendon Press, Oxford, pp. 277e281. estimation for Bayesian phylogenetic model selection. Syst. Biol. 60(2),
Wood, B.A., Uytterschaut, H., 1987. Analysis of the dental morphology of Plio- 150e160. http://dx.doi.org/10.1093/sysbio/.
Pleistocene hominids. III. Mandibular premolar crowns. J. Anat. 154, 121e156.

1 s2.0 S0047248417300866 Main

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 s2.0 S0047248417300866 Main

Uploaded by

Copyright:

Available Formats

Journal of Human Evolution 107 (2017) 107e133

Contents lists available at ScienceDirect

Journal of Human Evolution

The afnities of Homo oresiensis based on phylogenetic analyses

1. Introduction three broad ways: 1) that H. oresiensis is a distinct species and

Species and institution Skeletal component Specimen

Species and institution Skeletal component Specimen

1 Maximum cranial breadth Maximum cranial breadth Weidenreich, 1943:191

13* Foramen magnum Foramen magnum Tobias, 1991:762

1. slopes anteriorly 1. slopes anteriorly

45 Obelionic depression Obelionic depression Weidenreich, 1943:199

68 Superior transverse torus Superior transverse torus Johanson et al., 1982:383

3. point contact between metaconid,

Finarelli and Clyde, 2004:642 (C)

dened. 2. present but weakly dened

129* Cortical index Cortical index Jungers et al., 2009a:552

(A) Polymorphism as intermediate state

Analyses with H. rudolfensis excluded Tree length MPTs Bootstrap% Bayesian pp

Analyses with H. rudolfensis included Tree length MPTs Bootstrap% Bayesian pp

(B) Polymorphism as variable

Analyses with H. rudolfensis excluded Tree length MPTs Bootstrap% Bayesian pp

Analyses with H. rudolfensis included Tree length MPTs Bootstrap% Bayesian pp

* Signicant at p < 0.05 under Templeton Test.

You might also like