• Australopithecus;
  • canine honing;
  • craniofacial;
  • hominid;
  • masticatory;
  • megadontia;
  • Miocene;
  • Pliocene


  1. Top of page
  2. Abstract
  3. Introduction
  4. Definitive facial morphology of panins and hominins
  5. The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?
  6. The major modifications to the ancestral morphology during hominin evolution
  7. Conclusions
  8. Acknowledgements
  9. References

This review uses the current morphological evidence to evaluate the facial morphology of the hypothetical last common ancestor (LCA) of the chimpanzee/bonobo (panin) and human (hominin) lineages. Some of the problems involved in reconstructing ancestral morphologies so close to the formation of a lineage are discussed. These include the prevalence of homoplasy and poor phylogenetic resolution due to a lack of defining derived features. Consequently the list of hypothetical features expected in the face of the LCA is very limited beyond its hypothesized similarity to extant Pan. It is not possible to determine with any confidence whether the facial morphology of any of the current candidate LCA taxa (Ardipithecus kadabba, Ardipithecus ramidus, Orrorin tugenensis and Sahelanthropus tchadensis) is representative of the LCA, or a stem hominin, or a stem panin or, in some cases, a hominid predating the emergence of the hominin lineage. The major evolutionary trends in the hominin lineage subsequent to the LCA are discussed in relation to the dental arcade and dentition, subnasal morphology and the size, position and prognathism of the facial skeleton.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Definitive facial morphology of panins and hominins
  5. The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?
  6. The major modifications to the ancestral morphology during hominin evolution
  7. Conclusions
  8. Acknowledgements
  9. References

The facial skeletons of Homo sapiens and its closest living relatives in the genus Pan differ considerably. The extant species of Pan, like other extant apes, are distinguished by their pronounced supraorbital tori and long projecting lower faces that house relatively larger anterior teeth (i.e. incisors and canines) than posterior teeth (i.e. premolars and molars). The incisors are large and broad, the canines are sexually dimorphic and occlude in such a way as to maintain their sharp tips, while the premolars and molars are relatively small. Modern humans possess similarly sized molars but have considerably reduced anterior dentitions and a concomitantly smaller palate that does not protrude anteriorly. The canines are more weakly sexually dimorphic, look more like incisors and, rather than being pointed, are worn down at the tip because of a different mode of occlusion. The modern human face and orbits are tucked under a brain that is substantially larger than that in Pan (or any other extant ape) and so the supraorbital morphology is reduced and modified. These profound differences between the extant representatives of the hominin and panin lineages have evolved over the past 5–8 million years. The differences between the earliest representatives of either of these two lineages, their last common ancestor (LCA) or the hominids that predate the divergence, are likely to be much more subtle. This paper reviews the possible facial morphologies of the LCA and major trends in facial evolution within the hominin lineage.

Any attempt to reconstruct the morphology of a last common ancestor requires a working phylogeny and a knowledge of the primitive and derived morphologies of the extant representatives of the sister taxa in question. As is often the case when reconstructing the hypothetical last common ancestor of two lineages, one of the lineages is of particular interest and the other lineage is determined by default as that of the most closely related extant sister taxon. Here, for anthropocentric reasons, the lineage of particular interest is our own hominin lineage and, through a combination of phylogeny and the serendipity of extinction, the other is that of the panins (Fig. 1a). Whilst this molecularly derived phylogeny is now widely, but not wholly, accepted by the anthropological community, it is only partially supported by morphological data (Miyamoto & Goodman, 1990; Ruvolo, 1997).


Figure 1. (a) Phylogeny of the hominin, panin and gorilline lineages. (b) Hypothetical phylogenetic relationships of stem hominin and panin-hominin LCA candidate taxa. The numbers refer to millions of years before the present. The broken lines approximate the limits of the consenus range for the age of the LCA of the panin and hominin lineages.

Download figure to PowerPoint

There are two main obstacles to identifying a fossil taxon as a last common ancestor. Firstly, there is the issue of homoplasy, which is believed to have been common in the facial skeleton during hominin (McHenry, 1996; Lockwood & Fleagle, 1999) and hominoid evolution (Pilbeam & Young, 2001, 2004; Andrews & Harrison, 2005; Wolpoff et al. 2006). Homoplasy could reasonably be hypothesized to have been at least as common, and harder to identify, in closely related taxa such as those close to the LCA. This likelihood is based upon the assumption that closely related and recently divergent taxa are likely to have similar morphologies and, importantly, similar patterns of integration, such that an evolutionary change in one morphological component (e.g. change in dental proportion) is likely to generate similar more global morphological responses (e.g. facial skeleton and perhaps cranium as a whole). Recent findings that hominids have distinct facial ontogenies (Bastir & Rosas, 2004; Cobb & O’Higgins, 2004; Mitteroecker et al. 2004; McNulty et al. 2006) and that differences and similarities in ontogenetic facial integration of the extant hominids during ontogeny do not follow a pattern that could be predicted from phylogeny (Ackermann, 2005) would seem to rule out this scenario, but the degree of phylogenetic separation among the extant hominids is substantially greater than the differences under discussion here. The second obstacle is related to the level of resolution one can realistically expect to achieve when identifying, or at least hypothesizing, the morphology of a LCA. Lineages are most easily identified and defined using derived morphological features. However, taxa close to the formation of the two lineages will be expected to have relatively fewer derived features. In fact there is no reason to assume that the earliest members of the hominin lineage would necessarily have possessed any facial autapomorphies. It is this relative lack of derived morphological features in the stem taxa that reduces the phylogenetic resolution around lineage divergences (Wood, 2002; Andrews & Harrison, 2005). For these reasons, even if sampling of the fossil record were not a problem, it is probably unrealistic to hope to distinguish the morphology of the LCA from that of closely related taxa (Fig. 1b).

Andrews & Harrison (2005) suggest that a bottom-up approach to studying fossil hominins, in which attention is paid to the ancestral Miocene morphologies, may go some way to highlighting potential homoplasies and reducing the problems associated with poor phylogenetic resolution. This approach is intended to augment the top-down approach of using extant sister taxa, Pan in this case, as the reference sample against which members of the hominin lineage can be identified. In this way one can gain additional information regarding both primitive hominoid features that may be present in taxa close to the LCA and the tendency of earlier Miocene hominids independently to acquire morphological features that are considered to define hominins as autapomorphies. This review will use both top-down and bottom-up approaches to reconstruct the facial morphology of the last common ancestor of chimpanzees, bonobos and modern humans.

Definitive facial morphology of panins and hominins

  1. Top of page
  2. Abstract
  3. Introduction
  4. Definitive facial morphology of panins and hominins
  5. The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?
  6. The major modifications to the ancestral morphology during hominin evolution
  7. Conclusions
  8. Acknowledgements
  9. References

As discussed above, a fossil hominin close to the split with the LCA is likely to lack most of the defining morphological characteristics of the lineage. In fact, it is not even straightforward to identify the definitive morphological features of later and less disputed hominins because the lineage includes such a diverse array of morphologies (Fig. 2, Tables 1, 2). However, based upon the modifications to facial morphology observed among later hominins, a series of potential autapomorphies can be drawn up that generally relate to brain size, mastication and canine sexual dimorphism.


Figure 2. Morphological diversity in the hominin facial skeleton. Top row (left to right): P. troglodytes, Au. afarensis, Au. africanus, Au. boisei. Bottom row (left to right): Homo erectus, Homo heidelbergensis, Homo neanderthalensis, Homo sapiens. Not to scale.

Download figure to PowerPoint

Table 1.  Summary of facial measurements of P. troglodytes and hominins
 Palatal length (mm)Palatal breadth at M2 (mm)Facial height (nasion-prosthion) (mm)Bizygomatic breadth (mm)Biorbital breadth (mm)
Pan troglodytes
Sahelanthropus tchadensis
 TM 266-01-60-1§§ 766075n/a91
Australopithecus afarensis††Mean66688713989
Range57–7656–82[74], 100136, 14283–95
Australopithecus africanus††Mean67687811582
Range64–6864–7471–90110, 12079, 85
Australopithecus robustus††Mean667076§13392
Range61–6968–72128, 13884, 100
Australopithecus boisei††Mean7680101166103
Range73, 7979, 8191, 110162–172100–105
Australopithecus aethiopicus
 KNM-WT 17000†† 78808814994
Early ‘Homo
 KNM-ER 1813‡‡ 606566*[117][86]
 KNM-ER 1470‡‡ n/a[> 80]91*n/a101
Homo erectus
 KNM-ER 3733‡‡ n/a6683*[138][104]
Homo sapiens††Mean52666911395
Table 2.  Trends in facial size and endocranial volume in the hominin lineage
 Inferior facial length (prosthion-porion) (mm)Superior facial length (nasion-porion) (mm)Facial height (nasion-prosthion) (mm)Endocranial volume (cm3)
Australopithecus africanusMean123*84*78**452a
Range112, 13375, 9371–90428–500
Australopithecus robustusMean127*83*71§530b
Range121, 133
Australopithecus boiseiMean140*98*101515c
Range136, 14393, 10291, 110500–530
Early ‘Homo
 KNM-ER 1813 111*97*67509d
 KNM-ER 1470 123*91*91752e
Homo erectus
 KNM-ER 3733 120*86*83850f
Middle Pleistocene non-NeanderthalsMean127.81290g
Upper Palaeolithic
Homo sapiensMean108.9†1450i
Homo sapiensMean104.293*69*1452j
*Data from Bilsborough & Wood (1988).
**Data from Kimbel et al. (2004).
†Data from Trinkaus (2003).
§Mean of SK48 from Bilsborough & Wood (1988), Wood (1991) and Kimbel et al. (2004).
‡Specimen means for OH5 and SK 48 calculated from Wood (1991) and Kimbel et al. (2004).
¶Mean of Bilsborough & Wood (1988) and Wood (1991).
a Mean calculated from: MLD 1, MLD 37/38, Sts 5, Sts 19/58, Sts 60, Sts 71 (Holloway, 1973).
b Mean calculated from: SK 1585 (Holloway, 1973).
c Mean calculated from: KNM-ER 406, KNM-ER 732, OH5 (Holloway, 1973); KNM-ER 407 (Flak & Kasinga, 1983); KSA 10-525 (Holloway & Yuan, in Kimbel et al. 2004).
d From Holloway (1978).
e From Tobias (1987).
f From Holloway, in Stringer (1984).
g Mean calculated from: Atapuerca 4, Atapuerca 5 (Arsuaga et al. 1993); Bodo (Conroy et al. 2000); Broken Hill/Kabwe (Holloway, 1981a); Djebel Irhoud I (Holloway, 1981b); Djebel Irhoud II (Holloway, 1985); Petralona (Protsch, in Stringer, 1984).
h Mean calculated from: Gibraltar 1 (Keith, 1915); Krapina B (Holloway, 1985); La Chapelle , La Ferrassie 1, La Quina 5 (Holloway, 1981b); Saccopastore I, Saccopastore II (Holloway, 1985).
i Mean calculated from: Cro-Magnon 1 (Day, 1986); Herto (White et al. 2003); Singa (Gröning, 2003).
j From Holloway & Yuan, in Kimbel et al. (2004).

Arguably the most defining feature of hominin evolution is an increase in brain size, but because the majority of brain size increases occurred relatively late in the lineage (Table 2), it is not likely to prove a defining characteristic of taxa close to the LCA. The hominin masticatory apparatus is initially characterized by an adaptive trend towards harder and or tougher diets (Teaford & Ungar, 2000; Macho et al. 2005), with australopiths demonstrating changes in dental proportions, with an increase in the size of the posterior dentition and a decrease in the size of the anterior dentition. There is also a repositioning of masticatory muscles in relation to the palate, buttressing of the facial skeleton (presumably associated with resisting more powerful masticatory forces), and an increase in enamel thickness. Many of these masticatory adaptations to harder and or tougher diets are not seen in all hominins, especially the more recent hominins, but, most appropriately for the identification of Mio-Pliocene hominins, they are found in the majority of Pliocene taxa. The third main feature of the hominin lineage is canine reduction and the associated changes in the C/P3 honing complex. Apes tend to have a diastema mesial to the upper canine and relatively large and dimorphic canines (Kelley, 1995) with pointed tips that are maintained through wear along the mesial and distal surfaces. This ape pattern of canine wear is facilitated by an interlocking occlusion of the upper canine between the lower canine and elongated sectorial P3; it is this occlusal configuration that is collectively referred to as the C/P3 honing complex (Fig. 3a). Modern humans do not have a diastema, and there is a reduction in both canine sexual dimorphism and relative canine size, linked with more incisiform canines. Modern human canines, because they are short and lack both a sectorial premolar and a diastema, occlude at the tips and therefore wear apically. It is claimed that the hominin fossil record adequately samples and clearly documents progressive changes in this suite of features (e.g. Asfaw et al. 1999; White et al. 2006), and the possession of this complex of features has become an important factor in determining hominin status (Fig. 3) (White et al. 1994; Senut et al. 2001; Brunet et al. 2002; Haile-Selassie et al. 2004). It is therefore a working hypothesis that canine reduction and associated changes in the C/P3 honing complex, together with a suite of masticatory features associated with a diet requiring high occlusal loads, are the best way to define the hominin facial skeleton.


Figure 3. Early hominin canine morphology. (Adapted from Haile-Selassie et al. 2004).

Download figure to PowerPoint

When a bottom-up approach is adopted, these morphological features attract a number of caveats. First, as expanded upon below, some of these putative hominin features are also found in fossil non-hominin hominid taxa. In addition, once homoplasy of a morphological feature has been demonstrated between undisputed hominins and non-hominin hominids, the possibility still remains that there is also homoplasy at, and near, the base of the hominin lineage.

For example, some of the putative hominin synapomorphies related to an increasingly powerful masticatory apparatus (e.g. postcanine megadontia, thickened enamel) are not exclusive to the hominin lineage. Postcanine megadontia is apparently not unique to hominins since fossil hominids such as Ankarapithecus, Sivapithecus, Gigantopithecus and Ouranopithecus also demonstrate varying degrees of megadontia (Wood, 1981; Pilbeam et al. 1990; Andrews & Martin, 1991; Andrews & Alpagut, 2001; Kelley, 2001). Likewise, thick molar enamel is found in some megadont fossil non-hominin hominids (Andrews & Martin, 1991), most markedly so in Graecopithecus and Gigantopithecus. The non-dental features of the hominin facial skeleton associated with an increase in occlusal loads during mastication, with the exception of relatively non-specific increases in robusticity and buttressing, are not prevalent in the Miocene hominids. Oreopithecus has a relatively anteriorly placed zygoma, but whether this is related to a tougher diet, possibly more fibrous (Harrison & Rook, 1997), or to a different selective pressure is not clear (Moyà-Solà & Köhler, 1997).

Similar caveats also apply to canine reduction, a feature which has been reported in Ouranopithecus from the late Miocene of Greece (c. 9 mya) and controversially proposed as a synapomorphy linking that taxon with the modern human lineage (Bonis et al. 1990; Bonis & Koufos, 1993, 1994, 2001). In this case canine reduction was assessed by relating cervical crown dimensions of the canine to overall size of the molar dentition. However, Kelley (2001) showed the canines are relatively small not because they are absolutely smaller than in later taxa, but because the molars are absolutely larger. Canine crown height in Ouranopithecus far exceeds that expected for Australopithecus afarensis of a similar body size, but it is within the range of extant apes, excluding Pan paniscus (Kelley, 2001). Canine reduction and some associated modifications to the C/P3 complex are also known in at least two other fossil hominid taxa. A reduction in canine crown height and variable molarization of the P3 crown are characteristic of both Gigantopithecus species, with the Pleistocene Gigantopithecus blacki possessing markedly bicuspid P3 (Kelley, 2002; Andrews & Harrison, 2005; Wolpoff et al. 2006). The wear patterns on the canines of Gigantopithecus and Ouranopithecus are unlike those of other non-hominin hominids. Both taxa combine apical and distal wear (Bonis & Koufos, 2001; Wolpoff et al. 2006), so that the tips of the canines are not pointed, as might be expected for teeth used in aggressive display, but are instead blunted to the extent that the canines of older Gigantopithecus individuals are worn down to the occlusal plane. There is evidence of canine crown reduction in Oreopithecus and P. paniscus but these taxa have retained enough of the honing mechanism that the tips of the canines retain their pointed shape (Plavcan et al. 1995; Alba et al. 2001a; Andrews & Harrison, 2005).

A number of both dietary related and non-dietary related hypotheses have been proposed to explain canine reduction and are discussed below (e.g. Darwin, 1871; Kelley, 1989; Greenfield, 1990, 1992; Plavcan & Kelley, 1996; Plavcan & van Schaik, 1997; Alba et al. 2001a). Whether all the taxa that display this suite of features evolved in response to the same selective pressures is far from clear, and there is little consensus about the factors responsible for these evolutionary changes among the hominins (e.g. Darwin, 1871; Holloway, 1967; Kay, 1981; Lovejoy, 1981; Greenfield, 1990, 1992; Plavcan & Kelley, 1996; Plavcan & van Schaik, 1997).

Reductions in relative canine size and in canine size sexual dimorphism during hominin evolution have been attributed to reduced aggressive behaviours, non-biological weapon replacements or a combination of the two (reviewed in Greenfield, 1992). Canine reduction has been linked with a reduction in male–male competition due to the decreased selective pressure for aggressive behaviours involving canine display (e.g. Holloway, 1967; Kay, 1981; Lovejoy, 1981), a relationship that has now been demonstrated in extant primates (Plavcan & van Schaik, 1992; Plavcan et al. 1995). The combination of marked body size sexual dimorphism and reduced canine sexual dimorphism seen in Pliocene hominins (McHenry, 1991; Lockwood et al. 1996; Plavcan & van Schaik, 1997) casts doubt on there being a single cause of reduced male–male competition. Others have suggested that the canines were reduced in height and size so as not to interfere with a novel hominin chewing cycle pattern associated with a tougher diet (Brues, 1966; Jolly, 1970). Greenfield (1990) argued that the existence of such a novel pattern of chewing in hominins has been disproved (Hiiemae & Kay, 1972; Kay & Hiiemae, 1974) but the scenario still attracts support (e.g. Andrews & Alpagut, 2001).

Other researchers have suggested that canine reduction in hominins is a response to spatial constraints within the jaw (Jungers, 1978; Simpson et al. 1991; Alba et al. 2001a,b). Jungers (1978) proposed that canine reduction in hominins was the result of dental crowding due to the molarization and enlargement of the premolars. An objection to this model is that, if space within the dental row is the limiting factor, then the part of the canine with the greatest selection pressure to reduce in size would be the mesiodistal dimension, whereas it seems that selection focuses on canine height (Greenfield, 1992). Although it is true that the mesiodistal dimension of a tooth is related to the space available (or required) in the dental arcade, if only the mesiodistal dimension of a lower canine were reduced then the projecting tip of the tooth would still require a diastema between I2 and C in the maxilla. So, whilst the model could still apply to a reduction in the height of the lower canine, alone it may prove to be inconsequential for the upper canine.

It has also been proposed that the combination of orthognathism and a reduction in palatal length in hominins (Simpson et al. 1991), P. paniscus and Oreopithecus (Alba et al. 2001a,b) has imposed a spatial constraint on canine size. Alba et al. (2001b) suggest a number of selective scenarios for the evolution of orthognathism in Oreopithecus that may also be applicable to hominins, including: (1) the centre of gravity of the cranium becomes more posteriorly positioned, thus improving the biomechanical efficiency of supporting an upright head position in a bipedal posture; and (2) the zygomatic root becomes positioned more anteriorly along the dental row, increasing the mechanical advantage of the attached masticatory muscles as an adaptation to a more demanding diet. Under these scenarios, canine reduction is interpreted as a necessary consequence of size constraints imposed by an initial reduction in prognathism, presumably reducing palatal length. For this model to predict a reduction specifically in the canines, rather than any other tooth type, it must be assumed that there is also either a reduction in the selection pressure for aggressive displays from the canines or acquisition of a novel aggression display in the taxa.

All of the above mentioned spatial models (Jungers, 1978; Simpson et al. 1991; Alba et al. 2001a,b) propose a causal relationship between orthognathism and tooth size, specifically the canine, whereby the bone influences the dentition. However, as bone is known to be phenotypically more plastic than dental tissues, it is not unreasonable to suggest that the causal relationship could be the other way around, such that the dentition influences the development of the bone. This version of the spatial model would predict that a selection pressure to reduce the canine size and or postcanine size, as is the case in hominins, P. paniscus and Oreopithecus, would result in a correlated reduction in palatal length and or width. Whether a reduction in palatal, length results in reduced prognathism depends in turn on the position of the palate relative to the basicranium and/or the position of the zygomatic root and associated infraorbital region, as discussed in more detail below.

Canine reduction, modifications to the C/P3 honing complex and apical wear are therefore clearly not restricted to the hominins and may have been independently acquired a number of times in hominid evolution, possibly in response to different selective pressures. As suggested by Kelley (2001) and Andrews & Harrison (2005), the presence of all or part of this suite of features in Miocene hominids should therefore be viewed cautiously as a diagnostic feature of the hominin lineage. It is not just craniodental features that may have lost their diagnostic value, another classic diagnostic feature of the hominins, bipedalism, has recently started to come into question. An increasing number of researchers are also highlighting the possibility that bipedalism arose either independently a number of times early in the hominin lineage or even prior to the hominin lineage itself, and as such are raising doubts about its utility as a diagnostic feature of hominins (Oxnard, 1975; Andrews, 1995; Wood, 2002; Cameron & Groves, 2004; Andrews & Harrison, 2005; O’Higgins & Elton, 2007; Thorpe et al. 2007; Crompton, 2008, this volume). If this is the case, it may therefore be that there are no longer any features that can be regarded as clear hominin synapomorphies and as such, rather pessimistically, none of the traditionally accepted diagnostic hominin features associated with canine modification and bipedalism still stands. With the potential of such a severe handicap in identifying putative early hominins, there are two main ways by which this process may be improved. One is to adopt the bottom-up method discussed above (Andrews, 1995; Andrews & Harrison, 2005) and the latter is to hope that the Mio-Pliocene hominid fossil record will continue to improve.

The difficulty with which diagnostic autapomorphies are identified in the facial skeleton is not restricted to the hominins, it is also encountered with the panins. The relatively underived facial skeleton of extant Pan taxa and the absence of fossil evidence of the panin face, with the exception of just a small sample of relatively recent fossil teeth (c. 0.5 Ma; McBrearty & Jablonski, 2005), make it difficult to establish any model for the evolution of the panin face. Based on what little is known of the middle to late Miocene hominid facial and dental morphology and that of the extant Gorilla lineage, it would seem likely that the morphology of the earliest panins was little different from that of the extant Pan taxa (Wood, 2002; Pilbeam & Young, 2004), with the possible exception of having smaller incisors. This conclusion, in conjunction with those reached above, suggests that the LCA of hominins and panins had a facial morphology similar to that seen in the extant Pan.

The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?

  1. Top of page
  2. Abstract
  3. Introduction
  4. Definitive facial morphology of panins and hominins
  5. The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?
  6. The major modifications to the ancestral morphology during hominin evolution
  7. Conclusions
  8. Acknowledgements
  9. References

To qualify as a putative LCA, a taxon not only needs to satisfy morphological criteria, as discussed above, it also needs to be dated to the period of time during which the panin and hominin lineages diverged. The most recent estimates of the divergence date (reviewed in Bradley, this volume) currently give a broad and approximate consensus range of 4–8 Ma (e.g. Eizirik et al. 2004; Kumar et al. 2005; Steiper & Young, 2006). At present there are four hominid taxa from this time range, each of which has been proposed as either a candidate taxon for the LCA or as an early stem hominin.

Prior to 1994, the hominid fossil record during the critical period around the late Miocene and early Pliocene was represented by a handful of fragmentary and unassigned fossils (Hill, 1994). Ardipithecus ramidus, dated to around 4.4 Ma, was discovered in 1994 in the Aramis locality, Ethiopia, and reported to display a number of distinguishing hominin features (White et al. 1994, 1995). Ar. ramidus was originally assigned to the genus Australopithecus (White et al. 1994) but shortly afterwards was allocated to a new genus, Ardipithecus (White et al. 1995). The majority of the published material is dental and includes canines that are more incisiform and marginally smaller than those of extant panins and non-honing P3. Ar. ramidus also exhibits small incisors and molars with intermediate enamel thickness that are similar in size to Pan, but which are small relative to later hominins. The hominin status of this species (Cameron & Groves, 2004; Andrews & Harrison, 2005) has been questioned. More specifically, Andrews & Harrison (2005) express concern that, based on the sample available, the canine and premolar morphology is not dissimilar to that expected for a middle to late Miocene hominid.

More recently, material dated to 5.2–5.8 Ma and attributed to Ardipithecus kadabba, have been found at Asa Koma and a number of other localities in the Middle Awash, Ethiopia (Haile-Selassie, 2001; Haile-Selassie et al. 2004). Ar. kadabba possesses a functional honing complex and relatively long and ape-like canines (Fig. 3); the morphology was considered distinct enough to elevate it from a subspecies of Ar. ramidus to a separate species (Haile-Selassie et al. 2004). The differences between these two taxa may in fact be great enough to warrant a generic separation (Begun, 2004; Andrews & Harrison, 2005). Furthermore, Ar. kadabba is distinguished from extant and fossil apes on the basis of its more incisiform canines (Haile-Selassie, 2001); however the canines are primitive enough to have been interpreted by other researchers as indicative of an affinity with the panin lineage (Fig. 3) (Cameron & Groves, 2004). Senut et al. (2001) have also proposed that the genus Ardipithecus may be an ancestral panin, however, the reasons they cite, molar size and enamel thickness, are both known to be prone to homoplasy (Teaford & Ungar, 2000).

Another proposed early hominin is Orrorin tugenensis, dated to about 5.8–6.0 Ma from the Lukeino Formation, Kenya (Senut et al. 2001). As with the two species of Ardipithecus there are no facial remains of O. tugenensis but it is represented by some isolated teeth and mandibular fragments. Its molars are small, with relatively thick enamel, the large pointed canines are similar to those of female chimpanzee, and the relatively small and robust incisors resemble those of Ardipithecus (Senut et al. 2001). Molar size and enamel thickness have also been used to suggest a close relationship between O. tugenensis and later hominins to the exclusion of Ardipithecus (Senut et al. 2001).

The oldest and most recently discovered of the purported hominins, Sahelanthropus tchadensis, dates from ca. 7 Ma and was recovered from Toros-Menalla in Chad (Brunet et al. 2002, 2005; Guy et al. 2005). To date, material has been described from three localities of the same age: a well preserved but distorted cranium, a mandible and isolated teeth from TM 266 (Brunet et al. 2002, 2005), and mandibular fragments from TM 247 and TM 292 (Brunet et al. 2005). The face of the S. tchadensis cranium (TM 266-01-060-1) has been reconstructed (Zollikofer et al. 2005), and is notable for its small canines with apical and distal wear, small molars with moderately thick enamel, low degree of lower facial prognathism, anteroposteriorly short premaxilla, broad interobital region, large and continuous supraorbital torus and relatively square orbits. Brunet and colleagues (Brunet et al. 2002, 2005; Zollikofer et al. 2005) propose that S. tchadensis is the earliest known hominin on the basis that the small canines with apical wear (proposed as evidence of a non-honing C/P3 complex) and moderately thick/intermediate molar enamel are hominin autapomorphies, and that its large and continuous supraorbital torus, low degree of subnasal prognathism and absence of a canine diastema are features in common with later hominins such as Kenyanthropus and Homo.

The factors listed in support of S. tchadensis having a derived non-honing C/P3 complex of the type seen in later hominins, include the inferred absence of a lower diastema between C and P3, the observed absence of an upper diastema between the alveoli of I2 and C, evidence of the occlusion of the upper canine against the distal tubercle of the lower canine (inferred from a grooved wear strip that continues down the distal surface of the lower canine and terminates as an indentation on the distal tubercle) and an apically worn upper canine (Brunet et al. 2002, 2005). Andrews & Harrison (2005) have suggested that the morphology and inclination of the P3 roots (in the fragmentary mandible, TM-292-02-01) are indicative of an elongated and angled honing crown. However, even if the P3 shows no signs of honing wear it could be homoplastic rather than a hominin synapomorphy, for Ouranopithecus has sexually dimorphic canines that are large compared to any purported hominin but a relatively symmetrical P3 with a non-honing wear pattern very similar to Au. afarensis (Bonis & Koufos, 2001).

The hypothesis that the canine morphology of S. tchadensis is a hominin apomorphy rests on the assumptions that this morphology is not homoplastic with respect to the later hominins, and that the currently described canines are male. If the S. tchadensis cranium with a canine (TM 266-01-060-1), the mandibular fragment with a canine (TM-292-02-01) and the single isolated canine (TM-266-02-154-2) are female, then the synapomorphic diagnosis of this morphology, and in turn the hominin status of S. tchadensis, would be less convincing. The sex of the cranium was determined as male based upon the presence of the large, pronounced and continuous supraorbital torus (Brunet et al. 2002). There is, however, insufficient material attributed to this taxon at present with which to make such a judgment, in addition the hominin supraorbital torus is known to be an unreliable indicator of sex (McNulty & Baab, 2006), and so the sex assignment of this material remains to be more confidently assessed (Caspari, 2002; Andrews & Harrison, 2005; Wolpoff et al. 2006).

Some non-hominin hominids are known to have canine reduction with varying degrees of related premolar modification (see above) so homoplasy in canine reduction must be presumed to have occurred during hominid evolution. The absence of the canine diastema in hominids, as observed in S. tchadensis (Brunet et al. 2002, 2005), does not necessarily mean that there is no honing mechanism. Oreopithecus, for example, exhibits relatively small, but dimorphic and honed canines (especially in the cervical dimensions), a sectorial P3, but no diastema (Harrison & Rook, 1997; Alba et al. 2001a,b). Furthermore, the apical canine wear observed in S. tchadensis is not unique to hominins; for example, a similar pattern of wear is found in Gigantopithecus, associated with a high attrition diet, and on the large, sexually dimorphic canines of Ouranopithecus (Kelley, 2001; Wolpoff et al. 2006). As previously discussed, it is apparent that canine size reduction and the absence of honing complex features are not necessarily hominin apomorphies.

The supraorbital torus morphology of S. tchadensis has been compared to that in later hominins such as Kenyanthropus and Homo (Brunet et al. 2002; Wood, 2002). Similarities in the morphology of the supraorbital torus, however, are not restricted to hominins; together with the lateral orbital margin and the broad interorbital septum, S. tchadensis is very similar to both Gorilla and Ouranopithecus. In addition, the molars of S. tchadensis are similar to those of Ardipithecus, being relatively small and with an enamel thickness intermediate between that of Pan and later hominins.

S. tchadensis is of the right age and has a number of features that are consistent with it being a hominin; however, several of these features are prone to homoplasy in hominid evolution. It is therefore difficult to determine the taxonomic affiliation of this fossil. S. tchadensis is clearly a taxon that is close to the LCA of panins and hominins and currently it is the best candidate for being LCA. However, phylogenetic resolution close to the stem of any lineage is poor, and the fossil evidence presently available is insufficient to determine with any confidence whether it is most representative of the LCA, a stem hominin, a stem panin or a pre-divergence hominid (Fig. 1b). For the same reasons, it is not possible from facial evidence to determine which of the categories set out above Ar. kadabba, Ar. ramidus, and O. tugenensis belong to. The different morphologies exhibited by these potential earlier hominins has been interpreted by some (Wood, 2002; Begun, 2004) as evidence of an adaptive radiation early in the hominin lineage, whereas others (Haile-Selassie et al. 2004) argue that all four putative early hominin taxa (Ar. ramidus, Ar. kadabba, O. tugenensis and S. tchadensis) represent a single early hominin genus Ardipithecus.

The major modifications to the ancestral morphology during hominin evolution

  1. Top of page
  2. Abstract
  3. Introduction
  4. Definitive facial morphology of panins and hominins
  5. The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?
  6. The major modifications to the ancestral morphology during hominin evolution
  7. Conclusions
  8. Acknowledgements
  9. References

The discussion above of the morphology of the facial skeleton of the LCA and putative early hominins, whilst admittedly tentative and patchy in detail, provides the base from which to explore how the hypothetical ancestral morphology has been modified during the course of subsequent hominin evolution. Traditional interpretations of the evolution of the facial skeleton within the hominin lineage stress the influence of three main factors. In the early stages they are a reduction in canine sexual dimorphism, adaptations to a tougher and more abrasive diet, and later a combination of a dramatic increase in brain size, and adaptations to a less tough and obdurate diet. The major evolutionary trends are discussed below in relation to the dental arcade and dentition, subnasal morphology and the size, position and prognathism of the facial skeleton.

In association with major changes in the dentition, a number of modifications to the shape of the maxillary dental arcade occurred during hominin evolution. The earliest well preserved hominin palate is assigned to Australopithecus anamensis; unfortunately, it is only known from a small sample. The shape of the dental arcade and palate is pleisomorphic, closely resembling that of the African apes (Table 1). This morphology is exemplified by the typically great ape U shape of the maxillary dentition, with long parallel postcanine rows that converge slightly towards the Ms (Ward et al. 2001). The relative reduction of the canines and associated diastemata means that the incisors are not as anteriorly displaced as is typical of the African apes, making the anterior portion of arcade and the overall shape of the arcade more similar to that of female Pan troglodytes. This pattern is also seen in Au. afarensis, considered by some to be an direct descendant species of Au. anamensis (Kimbel et al. 2004; White et al. 2006); here the increased reduction of the canines furthers this trend such that, in the absence of large canines, the lateral incisors are more posteriorly positioned, thus giving the anterior arcade a more curved appearance than is typical of P. troglodytes (Fig. 4a,b) (Kimbel et al. 2004). The relative proportions of the tooth types in Australopithecus africanus differ from those of both Au. afarensis and Au. anamensis by displaying a relatively reduced incisor dentition and enlarged molar dentition (particularly M2 and M3; White et al. 1981). In spite of these differences in the dentition, the dental arcade in Au. africanus (Fig. 4c; Table 1) is generally consistent with this emerging archaic hominin pattern of an elongated and parabolic shape (Robinson, 1956; Kimbel et al. 2004). The dental arcade of Kenyanthropus platyops, while only represented by a single deformed specimen, is worthy of note for its distinct and apparently derived morphology for a hominin of its age (3.5–3.3 Ma) (Leakey et al. 2001). The dental arcade is relatively short and wide as seen independently both in various species of Homo and some megadont archaic hominins. The proportions of the individual tooth crowns in K. platyops are currently not well known, apart from the diminutive size of the M2, and so questions about the phylogenetic affinities of the taxon and the evolutionary modifications that contributed to this morphology remain unanswered.


Figure 4. Maxillary dental arcade morphology. The zygomatic processes of the maxillae are included to show their position relative to the dentition. Not exactly to scale.

Download figure to PowerPoint

The shape of the maxillary dental arcade of the megadont archaic hominins (Australopithecus aethiopicus, Australopithecus boisei, Australopithecus robustus and Australopithecus garhi) is very distinct from that of the above mentioned archaic hominins. The general trend among these taxa is a varying degree of enlargement of the postcanine dentition, thickened molar enamel, and molarization of the premolars (Robinson, 1956; Tobias, 1967; Grine, 1988; Wood, 1991; Asfaw et al. 1999; McCollum, 1999). In both Au. boisei (Fig. 4e) and Au. robustus, the postcanine enlargement is in conjunction with a considerable reduction in the anterior dentition. There is a concomitant modification to the shape of the dental arcade whereby the reduced anterior dentition (incisors and canines) are arranged coronally in a straight row, giving the palate a truncated appearance, and the elongated molar rows are straight and slightly divergent. Au. garhi, although only known from a single specimen (BOU-VP-12/130), displays a very different arrangement from that of Au. boisei and Au. robustus; all tooth types are large (a pattern similar to that of the Stw 252 cranium) with incisors and canines as large as those of the largest known australopiths (Asfaw et al. 1999) (Fig. 4d). The anterior portion of the dental arcade is therefore curved and more similar to the archaic hominins, whereas the postcanine portion is straight and divergent, as is typical of these megadont taxa. In the similarly dated Au. aethiopicus (c. 2.5 Ma), known from a single adult cranial specimen (KNM-WT 17000) and an edentulous hemimaxilla (EP 1500/01), the crowns are not well preserved and so while the size of the anterior dentition is not clearly known, the roots of the postcanine teeth indicate that the crowns were large (Walker et al. 1986; Leakey & Walker, 1988; Harrison, 2002). The dental arcade is large but quite distinct from the pattern seen in the other megadont australopith taxa, the postcanine tooth rows are not straight but converge posteriorly in a similar manner to Au. afarensis (Table 1). The overall shape of the archaic megadont hominin dental arcades, with the exception of Au. aethiopicus, is therefore somewhat trapezoid rather than the elongated parabolic shape of the archaic hominins with smaller postcanine dentitions.

Early Homo specimens display a more modern suite of dental proportions and dental arcade morphology as early as about 2.3 Ma (e.g. AL 666-1; Kimbel et al. 1997). The dental arcades of AL 666-1 and Homo habilis (Fig. 4f) are shorter than those of archaic hominins and have divergent postcanine dental rows (Table 1). This pattern appears to be consistent within the genus Homo with modifications in some taxa related to factors such as incisor enlargement in Homo erectus, and molar enlargement in Homo rudolfensis. With the possible exception of K. platyops, as described above, there is, however, no clear indication in the hominin fossil record of a transition in the shape of the dental arcade between the long and parabolic morphology of the archaic hominins and the generally broader and shorter parabolic dental arcades seen in early Homo and subsequent members of the genus, including anatomically modern Homo sapiens (Fig. 4g).

The general configuration of the premaxilla and hard palate in Au. anamensis, Au. afarensis and Au. africanus is pleisiomorphic, closely resembling that of P. troglodytes (McCollum et al. 1993; McCollum, 2000; Ward et al. 2001; Kimbel et al. 2004). This morphotype is typified by a varying degree of vertical relief in the transition between the superior (nasal) surfaces of the premaxilla and the hard palate, and a relatively thin hard palate (Fig. 5). The morphology of this region was previously referred to as either ‘stepped’ or ‘smooth’ (e.g. McCollum et al. 1993); however, recent studies have shown that hominoids in general display an array of variations of a ‘continuous’ morphology (McCollum, 2000). In essence the morphology of these archaic hominins essentially does not display the smooth and flat nasal floor seen in H. sapiens, but rather has some relief between the premaxilla and hard palate. The morphology and, most importantly, the point of insertion of the anterior tip of the vomer has been shown to vary between taxa (McCollum et al. 1993; McCollum & Ward, 1997; McCollum, 1997, 1999, 2000), it has been argued that where the vomer does not insert onto the premaxilla (e.g. Pan, see Fig. 5), the premaxilla and hard palate can develop and function with some degree of independence; however, insertion of the vomer onto the premaxilla acts as a constraint and so gives rise to a flat and smooth nasal floor (e.g. Au. boisei, see Fig. 5). Accordingly, the anterior tip of the vomer does not insert across the superior surface of the premaxilla in Au. afarensis and Au. africanus (this morphology is not preserved in Au. anamensis), resulting in some degree of relief in the nasal floor. In Au. africanus and, to a greater degree, Au. afarensis the vomer inserts on the posterior surface of the premaxilla, more anteriorly than is typical for P. troglodytes where the vomer tends to insert at the anterior of the hard palate (Fig. 5). Typically in all three of these species, as in P. troglodytes, the incisors are procumbent, and the premaxilla is long and prognathic, being orientated relatively horizontally to the hard palate and resulting in a shallow palate. In lateral profile, the premaxilla in these taxa protrudes anteriorly beyond the crowns and roots of the canines, thus pronouncing this subnasal prognathism (Fig. 6a). The subnasal region in K. platyops can only be observed in a single poorly preserved specimen, the palate is thin and the premaxilla is more vertically orientated than in nearly all australopiths (including the megadont taxa) and does not project beyond the anterior margins of the nasal aperture, thus adding to the derived appearance of this early specimen (Leakey et al. 2001; Spoor et al. 2005).


Figure 5. Subnasal morphology. Sagittal sections of the subnasal region taken lateral to the midline to preserve the nasal septum (McCollum et al. 1993; Fig. 5d has been modified in accordance with Fig. 5a,b is adapted from McCollum & Ward, 1997; Fig. 5c–e is adapted from McCollum, 2000).

Download figure to PowerPoint


Figure 6. Subnasal prognathism. (a) Schematic representation of a nasoalveolar clivus protruding anterior to anterior profile of the face, and (below) the arrangement of the dentition in the dental arcade that results in subnasal prognathism. (b) Schematic representation of a nasoalveolar clivus obscured by the anterior profile of the face, and (below) the arrangement of the dentition in the dental arcade.

Download figure to PowerPoint

The subnasal morphology of the megadont archaic hominins, with the exception of Au. garhi, is distinct from the above mentioned archaic hominins and all later hominins. The transition between the premaxilla and the nasal surface of the hard palate is smooth in Au. boisei, Au. robustus and Au. aethiopicus (McCollum, 2000) whereas in Au. garhi there is a discrete transition, more typical of the non-megadont archaic hominins (Asfaw et al. 1999). As these morphologies suggest, the vomer inserts on top of the premaxilla in Au. boisei, Au. robustus and Au. aethiopicus (McCollum, 2000). The description of Au. garhi does not specify the details of the vomeral insertion (Asfaw et al. 1999); however, the discrete transition in the nasal floor suggests that the vomeral insertion is also as found in the non-megadont archaic hominins. Au. aethiopicus and Au. garhi have the pleisiomorphic arrangement of procumbent incisors within a horizontally orientated premaxilla; Au. boisei and Au. robustus, on the other hand, have vertically orientated incisors, housed in a similarly orientated premaxilla. The palates of Au. aethiopicus and Au. garhi, as might therefore be expected, are shallow but in Au. boisei and Au. robustus the hard palate arches posteriorly and deepens considerably, particularly in Au. boisei. Subnasal prognathism is only obvious in the lateral profile of Au. aethiopicus and Au. garhi due to the projection of the anterior dental arch beyond the lateral margins of nasal aperture, as also seen in Au. anamensis, Au. afarensis and Au. africanus (Fig. 6a). In Au. boisei and Au. robustus a similar degree of subnasal prognathism is obscured by the relatively anterior position of the margins of the nasal cavity (Fig. 6b). The vertical height of the palate in Au. boisei, Au. robustus and Au. aethiopicus (McCollum et al. 1993) is greatly thickened in comparison with all other hominins; the palate of Au. garhi is much thinner than is typical of the megadont hominins (Asfaw et al. 1999).

The thickened hard palate found in these megadont australopiths has been variably considered a mechanical modification to withstand high masticatory forces in these taxa, in particular to reinforce the midpalatal suture (e.g. Robinson, 1954; Rak, 1983). Alternatively, McCollum (1997, 1999) proposed a developmental model which provides a non-mechanical explanation for the presence of a thickened hard palate. McCollum argues that the growth of the tall mandibular ramus in robust australopiths, recognized as a means of increasing the mechanical advantage of the masticatory apparatus, would have necessitated growing a tall maxilla, in itself achieved by the increased vertical growth of the maxilla. Critically, McCollum proposes that the relative developmental independence of nasal and oral modules means that, with an increase in maxillary height, the oral surface would remodel and drift inferiorly, whereas the functionally and developmentally independent nasal surface would not undergo any major remodelling. The result of such a scenario therefore is a dissociation of the nasal and oral surfaces of the hard palate and as such an increase in palatal thickness. Finite-element analyses have demonstrated that stress in the palate can be reduced by thickening the bone; however, the corollary is that such a rigid palate increases the stress in other areas (Strait et al. 2007). Robust australopiths with thickened palates do indeed have modifications to the rest of the facial skeletons to resist such stress and so the large suite of modifications, characteristic of the robust australopiths, may be a highly integrated response to a relatively small number of adaptations associated with increasing masticatory power (McCollum, 1999).

It is interesting to note that P. troglodytes and Au. afarensis have variably thick palates; in some cases the thickness in P. troglodytes approaches that of the robust australopiths. In P. troglodytes, and to a lesser extent in Au. afarensis (e.g. Taung) and Au. boisei (e.g. OH 5), the palate is pneumatized as an extension of the maxillary air sinus (Cave & Haines, 1940; Tobias, 1967; Conroy & Vannier, 1987; McCollum & Ward, 1997); however, at present it is not clear whether the development of thickened palates in these different taxa is homologous. If the developmental basis of this morphology is homologous, it is possible that in both cases the palate is thickened primarily as a result of non-mechanical factors during development and, in the case of robust australopiths, provides an exaptation to resist high masticatory forces, whereas in P. troglodytes relatively low masticatory strains have allowed the bone to be pneumatized by the maxillary sinus.

The limited well preserved sample of early Homo (AL 666-, OH 24, OH 62) and H. erectus (KNM-ER 15000 and Sangiran 17) indicates that an anterior nasal spine that protrudes from the nasal aperture is characteristic for all but the earliest Homo specimens (McCollum et al. 1993; Kimbel et al. 1997; Rightmire, 1998; McCollum, 2000) and a generally modern human pattern of a smooth nasal floor is characteristic of Homo, with the exception of the Dmanisi specimens (D2282, D2700), which display a more primitive discontinuous morphology (Rightmire et al. 2006). Moderate subnasal prognathism is found in some early Homo specimens (e.g. KNM-ER 1813, AL 666-, OH 62) but a more vertically orientated premaxilla is found in other early Homo specimens (e.g. KNM-ER 1470) and most later members of the Homo genus. This premaxilla morphology contributes to the increasingly orthognathic appearance of the facial skeleton of Homo. The depth of the palate shows considerable variation within the early Homo taxa, ranging from shallow (e.g. OH 13), to moderate in the Dmanisi and some earlier African Homo specimens (e.g. KNM-ER 1813, KNM-ER 42703), to deep in H. erectus (e.g. KNM-ER 3733, Sangiran 4) and some earlier members of the Homo genus (e.g. AL 666-1) (Kimbel et al. 2004; Rightmire et al. 2006; Spoor et al. 2007). The increase in palatal depth in Homo is not accompanied by a thickening of the palate as is seen in robust australopiths and so, according to McCollum's model (McCollum, 1997, 1999), would seem to develop via different developmental mechanisms.

The overall proportions and relative position of the facial skeleton have undergone a number of major modifications during hominin evolution. The facial skeletons of Au. afarensis and Au. africanus are the most generalized and primitive in terms of facial prognathism preserved among the hominins, with more than half of the palate protruding anterior to the orbits (Kimbel et al. 2004). These early hominins are not as prognathic as P. troglodytes and have more anteriorly positioned zygomatic processes (Figs 2 and 4). Both prognathism and orthognathism are found in the hominin lineage and their relative magnitudes can derive from modifications in the length of the palate, the relative position of the palate or a combination of both (Rak, 1983). In comparison with earlier hominins, Au. robustus and, to a greater degree, Au. boisei are characterized by orthognathic faces (Robinson, 1972; Howell, 1978; Rak, 1983). The palates of both robust australopiths are longer than, or as long as, those of earlier hominins (Table 1); however, the reduction in the apparent degree of prognathism is achieved by a posterior retraction of the palate as a whole, such that the posterior surface of the maxilla is relocated closer to the glenoid region (Rak, 1983) (Fig. 7a). Au. aethiopicus, in contrast, is extremely prognathic with over 80% of the palate being anterior to the orbits (Kimbel et al. 2004). All three robust australopiths, however, display an anterior migration of the infraorbital region and zygomatic process (Figs 2, 4). Although the faces of Au. boisei and, to a lesser degree, Au. robustus appear to be orthognathic in lateral profile, there is some protrusion of the anterior palate relative to the upper face, just obscured by the anteriorly migrated infraorbital region (Bilsborough & Wood, 1988). This integrated suite of morphological modifications in the robust australopiths, together with the increase in maxillary height allow for an increased and more evenly distributed bite force along the elongated molariform postcanine tooth row (Ward & Molnar, 1980). The anterior migration of the infraorbital region and zygomatic process has two major consequences that are hypothesised to facilitate this modification. First, the masseter muscle migrates anteriorly with the zygomatic process with the effect of increasing the mechanical advantage of the masseter muscle over the postcanine dentition as a whole (Rak, 1983). Second, the more anterior position of the infraorbital region and zygomatic process provides support for the premolar region of the palate under these higher masticatory loads (Robinson, 1962; Rak, 1983). In the case of Au. robustus and Au. africanus, where the infraorbital region and zygomatic process are not as anterior positioned as in Au. boisei, it is hypothesised that the development of anterior buttressing, in the form of anterior pillars, has assisted in this role (Rak, 1983). The combination of enlarged premolars and a more plesiomorphic maxilla as observed in Au. garhi would, however, suggest that the acquisition of large premolars in australopiths was not necessarily associated with high anterior masticatory loads.


Figure 7. Schematic representation of modifications to the relative position of the lower face during hominin evolution. (a) Orthognathism due to retraction of the whole palate. (b) Orthognathism due to anterior palatal reduction. (c) Posterior palatal reduction. The broken lines represent the hypothetical ancestral morphology, solid lines represent the descendant morphology. Broken arrows indicate location of facial reduction, solid arrows indicate direction of shift in the relative position of the face.

Download figure to PowerPoint

As discussed above, many early Homo specimens are characterized by a relative reduction in palatal length, associated with a reduction in the anterior dentition (Fig. 4). The facial skeleton accommodates this palatal reduction by effectively retracting the lower face, resulting in an orthognathic profile (Fig. 7b); however, it is unclear whether this reduction in prognathism is simply achieved through a shortening of the anterior portion of the palate or through a combination of palatal reduction and repositioning. It is possible that a variety of combinations have produced the array of early Homo facial morphologies (e.g. KNM-ER 1470, KNM-ER 1813, KNM-ER 3733, SK 847). Although members of the genus Homo are characterized by a reduction in the degree to which the lower face projects beyond the upper face, there is also a trend of increasing overall facial size (Table 2) and increasing projection of the face as a whole relative to the neurocranium (Spoor et al. 1999; Lieberman et al. 2002; Trinkaus, 2003), which reaches a maximum in archaic humans such as Homo neanderthalensis and Homo heidelbergensis. Thus, even though brain size increases considerably in these taxa compared to earlier Homo (Table 2), the face is not retracted beneath the enlarged neurocranium; it is a modern human autapomorphy that the whole face is retracted relative to the neurocranium (Lieberman et al. 2002). There is no major increase in either cranial capacity (Table 2) or the length of the anterior cranial fossa (Spoor et al. 1999) in modern humans compared to the large-faced archaic humans, thus it is hypothesized (e.g. Spoor et al. 1999; Lieberman et al. 2002; Bastir et al. 2008) that modern human facial retraction is a product of facial reduction, rather than a product of an anterior projection of the anterior cranial fossa, or a combination of the two. The further reduction in palatal length in modern humans compared to archaic humans may account for some, if not most, of this anteroposterior facial reduction. In addition, recent work has shown that the middle cranial fossae, which abut the posterior surface of the maxillae, are relatively longer in modern humans than in archaic humans (Bastir et al. 2007, 2008). The anterior extension of the middle cranial fossa indicates that the posterior surface of the maxilla is anteriorly displaced relative to the neurocranium in modern compared to archaic humans (Fig. 7c).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Definitive facial morphology of panins and hominins
  5. The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?
  6. The major modifications to the ancestral morphology during hominin evolution
  7. Conclusions
  8. Acknowledgements
  9. References

The reconstruction of facial morphology and the determination of its taxonomic and phylogenetic significance is fraught with difficulty when the fossil evidence is close to the formation of a lineage. This is because of the likelihood of both homoplasy and a paucity of defining derived features. It is therefore difficult to list the facial morphology that would be hypothesized to distinguish the LCA of chimp/bonobos and modern humans from stem members of either the hominin or panin lineages. The facial morphology of the current candidate LCA taxa (Ar. ramidus, Ar. kabadda, O. tugenensis and S. tchadensis) have been reviewed. In light of the problem summarized above and the paucity of the fossil evidence of the face in the hypodigms of these four taxa, it is not possible to determine with any confidence whether any of them is the LCA , or a stem taxon in either lineage, or a member of an extinct, and until now unrecognized, hominid lineage. Even if the fossil record were better for this period, the problems of homoplasy mean that it might be unrealistic to think that facial morphology will neatly resolve the systematic ambiguities presented by these putative early hominin taxa.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Definitive facial morphology of panins and hominins
  5. The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?
  6. The major modifications to the ancestral morphology during hominin evolution
  7. Conclusions
  8. Acknowledgements
  9. References

I thank Sarah Elton and Bernard Wood for their invitation to contribute to this volume, their considerable patience and their helpful comments on this paper. Thanks also to ASGBI for supporting and hosting the symposium; and to Markus Bastir, Jay Kelley, Kiernan McNulty and Paul O’Higgins for their comments and help.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Definitive facial morphology of panins and hominins
  5. The candidates in the fossil record: last common ancestor, stem hominins, stem panins or pre-divergence hominids?
  6. The major modifications to the ancestral morphology during hominin evolution
  7. Conclusions
  8. Acknowledgements
  9. References
  • Ackermann R (2005) Ontogenetic integration of the hominoid face. J Hum Evol 48, 175197.
  • Alba DM, Moyà-Solà S, Köhler M (2001a) Canine reduction in the Miocene hominoid Oreopithecus bambolii: behavioural and evolutionary implications. J Hum Evol 40, 116.
  • Alba DM, Moyà-Solà S, Köhler M, Rook L (2001b) Heterochrony and the cranial anatomy of Oreopithecus: some cladistic fallacies and the significance of developmental constraints in phylogenetic analysis. In Hominoid Evolution and Climatic Change in Europe: Phylogeny of the Neogene Hominoid Primates of Eurasia (eds De BonisL, KoufosGD, AndrewsP), pp. 284315. Cambridge: Cambridge University Press.
  • Andrews P (1995) Ecological apes and ancestors. Nature 376, 555556.
  • Andrews P, Alpagut B (2001) Functional morphology of Ankarapithecus meteai. In Hominoid Evolution and Climatic Change in Europe: Phylogeny of the Neogene Hominoid Primates of Eurasia (eds De BonisL, KoufosGD, AndrewsP), pp. 213230. Cambridge: Cambridge University Press.
  • Andrews P, Harrison T (2005) The last common ancestor of apes and humans. In Interpreting the Past: Essays on Human, Primate, and Mammal Evolution (eds LiebermanDE, SmithRJ, KellyJ), pp. 103121. Boston: Brill Academic Publishers.
  • Andrews P, Martin L (1991) Hominoid dietary evolution. Philos Trans R Soc Lond B Biol Sci 334, 199209.
  • Arsuaga JL, Martinez I, Gracia A, Carretero JM, Carbonell E (1993) Three new human skulls from the Sima de los Huesos Middle Pleistocene site in Sierra de Atapuerca, Spain. Nature 362, 534537.
  • Asfaw B, White TD, Lovejoy CO, et al. (1999) Australopithecus garhi: a new species of early hominid from Ethiopia. Science 284, 629635.
  • Bastir M, Rosas A (2004) Facial heights: Evolutionary relevance of postnatal ontogeny for facial orientation and skull morphology in humans and chimpanzees. J Hum Evol 47, 359381.
  • Bastir M, Rosas A, O’Higgins P (2007) 3D architecture of the middle cranial fossa and modern human origins. Am J Phys Anthropol 132 (S44), 69.
  • Bastir M, Rosas A, Lieberman DE, O’Higgins P (2008) Middle cranial fossa anatomy and the origin of modern humans. Anat Rec Part 291, 130140.
  • Begun DR (2004) The earliest hominins – is less more? Science 303, 14781480.
  • Bilsborough A, Wood BA (1988) Cranial morphometry of early hominids: facial region. Am J Phys Anthropol 76, 6186.
  • De Bonis L, Koufos GD (1993) The face and the mandible of Ouranoithecus macedoniensis: description of new specimens and comparisons. J Hum Evol 24, 46991.
  • De Bonis L, Koufos GD (1994) Our ancestors’ ancestor: Ouranoithecus is a Greek link in human ancestry. Evol Anthropol 3, 7583.
  • De Bonis L, Koufos GD (2001) Phylogenetic relationships of Ouranoithecus macedoniensis (Mammalia, Primates, Hominoidea, Hominidae) of the late Miocene deposits of Central Macedonia (Greece). In Hominoid Evolution and Climatic Change in Europe: Phylogeny of the Neogene Hominoid Primates of Eurasia (eds De BonisL, KoufosGD, AndrewsP), pp. 254268. Cambridge: Cambridge University Press.
  • De Bonis L, Bouvrain G, Geraads D, Koufos GD (1990) New hominid skull material from the late Miocene of Macedonia in northern Greece. Nature 345, 712714.
  • Brues A (1966) Probable mutation effect and the evolution of hominid teeth and jaws. Am J Phys Anthropol 25, 169170.
  • Brunet M, Guy F, Pilbeam D, et al. (2002) A new hominid from the Upper Miocene of Chad, Central Africa. Nature 418, 145151.
  • Brunet M, Guy F, Pilbeam D, et al. (2005) New material of the earliest hominid from the Upper Miocene of Chad. Nature 434, 752755.
  • Cameron DW, Groves CP (2004) Bones, Stones and Molecules: ‘Out of Africa’ and Human Origins. London: Academic Press.
  • Caspari R (2002) It's gotta walk the walk: making sense of the newly reported Sahelanthropus tchadensis fossil remains from Chad. Gen Anthropol 9, 110.
  • Cave AJE, Haines WB (1940) The paranasal sinuses of the African apes. J Anat 72, 493523.
  • Cobb SN, O’Higgins P (2004) Hominins do not share a common postnatal facial ontogenetic shape trajectory. J Exp Zool (Mol Dev Evol) 302B, 302321.
  • Conroy GC, Vannier MW (1987) Dental development of the Taung skull from computerized tomography. Nature 329, 625627.
  • Conroy GC, Jolly CJ, Cramer D, Kalb JE (2000) Newly discovered fossil hominid skull from the Afar depression, Ethiopia. Nature 276, 6770.
  • Crompton RH, Vereecke EE, Thorpe SKS (2008) Locomotion and posture from the common hominoid ancestor to fully modern hominins, with special reference to the common Panin/Hominin ancestor. J Anat 212, 501543.
  • Darwin C (1871) The Descent of Man. London: Murray.
  • Day MH (1986) Guide to Fossil Man. London: Cassells.
  • Eizirik E, Murphy WJ, Springer MS, O’Brien SJ (2004) Molecular phylogeny and dating of early primate divergences. In Anthropoid Origins: New Visions (eds RossCF, KayRF), pp. 4564. New York: Kluwer Academic/Plenum Publishers.
  • Flak D, Kasinga S (1983) Cranial capacity of a female robust australopithecine (KNM ER 407) from Kenya. J Hum Evol 12, 515518.
  • Greenfield LO (1990) Canine reduction in early man: a critique of three mechanical models. J Hum Evol 5, 213226.
  • Greenfield LO (1992) Origin of human canine: a new solution to an old enigma. Yearb Phys Anthropol 35, 153185.
  • Grine FE (1988) Evolutionary History of the ‘Robust’ Australopithecines. New York: Aldine de Gruyter.
  • Gröning F (2003) Zur endo- und intracranialen Variabilität der Frontalregion beim archaischen und modernen Homo sapiens in Afrika und Europa: Eine Studie mit Hilfe CT-basierender Rekonstruktion. Masters thesis. University of Hamburg.
  • Guy F, Lieberman DE, Pilbeam D, et al. (2005) Morphological affinities of the Sahelanthropus tchadensis (Late Miocene hominid from Chad) cranium. Proc Natl Acad Sci USA102, 1883618841.
  • Haile-Selassie Y (2001) Late Miocene hominids from the Middle Awash, Ethiopia. Nature 412:178181.
  • Haile-Selassie Y, Suwa G, White TD (2004) Late Miocene teeth from Middle Awash, Ethiopia, and early hominid dental evolution. Science 303, 15031505.
  • Harrison T (2002) The first record of fossil hominins from the Ndolanya Beds, Laetoli, Tanzania. Am J Phys Anthropol S34, 83.
  • Harrison T, Rook L (1997) Enigmatic anthropoid or misunderstood ape? The phylogenetic status of Oreopithecus bambolii reconsidered. In Function, Phylogeny and Fossils: Miocene Hominoid Evolution and Adaptations (eds BegunDB, WardCV, Rose, MD), pp. 327362. New York: Plenum Press.
  • Hiiemae KM, Kay RF (1972) Trends in the evolution of primate mastication. Nature 240, 486487.
  • Hill A (1994) Late Miocene and early Pliocene Hominoids from Africa. In Paleoclimate and Evolution, with Emphasis on Human Origins (eds VrbaES, DentonGH, PartridgeTC, BurkleLH), pp. 178193. New Haven: Yale University Press.
  • Holloway R (1967) Tools and teeth: some speculations regarding canine reduction. Am Anthropol 69, 6367.
  • Holloway R (1973) Endocranial volumes of early African hominids and the role of the brain in human mosaic evolution. J Hum Evol 2, 449458.
  • Holloway R (1978) Problems of brain endocast interpretation ana African hominid evolution. In Early Hominids of Africa (ed. JollyCJ), pp. 379401. London: Duckworth.
  • Holloway R (1981a) Exploring the dorsal surface of hominoid brain endocasts by stereoplotter and discriminant analysis. Phil Trans R Soc Lond B Biol Sci 292, 155166.
  • Holloway R (1981b) Volumetric and asymmetry determinations on recent hominid endocasts: Spy I and II, Djebel Ihroud I, and the Salè Homo erectus specimen, with some notes on Neandertal brain size. Am J Phys Anthropol 55, 385394.
  • Holloway R (1985) The poor brain of Homo sapiens neanderthalensis: see what you please ... In Ancestors: The Hard Evidence (ed. DelsonE), pp. 319324. New York: A.R. Liss, Inc.
  • Howell FC (1978) Hominidae. In Evolution of African Mammals (eds MaglioVJ, CookeHBS), pp. 154248. Cambridge: Harvard University Press.
  • Jolly C (1970) The seed-eaters; a new model of hominid differentiation based on a baboon analogy. Man 5, 526.
  • Jungers WL (1978) On canine reduction in early hominids. Curr Anthropol 19, 155156.
  • Kay RF (1981) The nut-crackers – a new theory of the adaptations of Ramapithecinae. Am J Phys Anthropol 55, 141151.
  • Kay RF, Hiiemae KM (1974) Jaw movement and tooth use in recent and fossil primates. Am J Phys Anthropol 40, 227256.
  • Keith A (1915) The Antiquity of Man. London: Williams and Norgate.
  • Kelley J (1989) Behavioral implications of sexual dimorphism in Australopithecus afarensis. Am J Phys Anthropol 78, 251.
  • Kelley J (1995) Sexual dimorphism in canine shape among extant great apes. Am J Phys Anthropol 96, 365389.
  • Kelley J (2001) Phylogeny and sexually dimorphic characters: Canine reduction in Ouranopithecus. In Hominoid Evolution and Climatic Change in Europe: Phylogeny of the Neogene Hominoid Primates of Eurasia (eds De BonisL, KoufosGD, AndrewsP), pp. 269283. Cambridge: Cambridge University Press.
  • Kelley J (2002) The hominoid radiation in Asia. In The Primate Fossil Record (ed. HartwigWC), pp. 369384. Cambridge: Cambridge University Press.
  • Kimbel WH, Johanson DC, Rak Y (1997) Systematic assessment of a maxilla of Homo from Hadar, Ethiopia. Am J Phys Anthropol 103, 235262.
  • Kimbel WH, Rak Y, Johanson DC (2004) The Skull of Australopithecus afarensis. New York: Oxford University Press.
  • Kumar S, Filipski A, Swama V, Walker A, Hedges SB (2005) Placing confidence limits on the molecular age of the human–chimpanzee divergence. Proc Natl Acad Sci USA 102, 1884218847.
  • Leakey MG, Spoor F, Brown FH, et al. (2001) New hominin genus from eastern Africa shows diverse middle Pliocene lineages. Nature 410, 433440.
  • Leakey R, Walker AC (1988) New Australopithecus boisei specimens from East and West Lake Turkana, Kenya. Am J Phys Anthropol 76, 124.
  • Lieberman DE, McBratney BM, Krovitz G (2002) The evolution and development of cranial form in Homo sapiens. Proc Natl Acad Sci USA 99, 11341139.
  • Lockwood CA, Fleagle JG (1999) The recognition and evaluation of homoplasy in primate and human evolution. Yearbook of Physical Anthropology 42, 189232.
  • Lockwood CA, Richmond BG, Jungers WL, Kimbel WH (1996) Randomization procedures and sexual dimorphism in Australopithecus afarensis. J Hum Evol 31, 537548.
  • Lovejoy CO (1981) The origin of man. Science 211, 341350.
  • Macho GA, Shimizu D, Jiang Y, Spears IR (2005) Australopithecus anamensis: a finite-element approach to studying the functional adaptations of extinct hominins. Anat Rec Part A 283, 310318.
  • McBrearty S, Jablonski N (2005) The first fossil chimpanzee. Nature 437, 105108.
  • McCollum MA (1997) Palatal thickening and facial form in Paranthropus: examination of alternative developmental models. Am J Phys Anthropol 103, 375392.
  • McCollum MA (1999) The robust australopithecine face: a morphogenetic perspective. Science 284, 301305.
  • McCollum MA (2000) Subnasal morphological variation in fossil hominids: a reassessment based on new observations and recent developmental findings. Am J Phys Anthropol 112, 275283.
  • McCollum MA, Ward SC (1997) Subnasoalveolar anatomy and hominoid phylogeny: evidence from comparative ontogeny. Am J Phys Anthropol 102, 377405.
  • McCollum MA, Grine FE, Ward SC, Kimbel WH (1993) Subnasal morphological variation in extant hominoids and fossil hominids. J Hum Evol 24:87111.
  • McHenry HM (1991) Sexual dimorphism in Australopithecus afarensis. J Hum Evol 20, 2132.
  • McHenry HM (1996) Homoplasy, clades and hominid phylogeny. In Contemporary Issues in Human Evolution (eds MeikleWF, HowellsFC, JablonskiNG), pp. 7792. San Francisco: California Academy of Sciences.
  • McNulty KP, Baab KL (2006) Sexual dimorphism in hominine supraorbital morphology. Am J Phys Anthropol 129 (S42), 128129.
  • McNulty KP, Frost SR, Strait DS (2006) Examining affinities of the Taung child by developmental simulation. J Hum Evol 51, 274296.
  • Mitteroecker P, Gunz P, Bernhard M, Schaefer K, Bookstein FL (2004) Comparison of cranial ontogenetic trajectories among great apes and humans. J Hum Evol 46, 679698.
  • Miyamoto MM, Goodman M (1990) DNA systematics and evolution of primates. Annual Review of Ecology and Systematics 2, 197220.
  • Moyà-Solà S, Köhler M (1997) The phylogenetic relationships of Oreopithecus bambolii Gervais, 1872. C R Acad Sci Ser IIA 324, 141148.
  • O’Higgins P, Elton S (2007) Walking on trees. Science 316, 12921294.
  • Oxnard CE (1975) Uniqueness and Diversity in Human Evolution: Morphometric studies of australopithecines. Chicago: the University of Chicago Press.
  • Pilbeam DR, Young NM (2001) Sivapithecus and hominoid evolution: some brief comments. In Hominoid Evolution and Climatic Change in Europe: Phylogeny of the Neogene Hominoid Primates of Eurasia (eds De BonisL, KoufosGD, AndrewsP), pp. 349364. Cambridge: Cambridge University Press.
  • Pilbeam DR, Young NM (2004) Hominoid evolution: synthesising disparate data. C R Palevol 3, 305321.
  • Pilbeam DR, Rose MD, Barry J, Shah I (1990) New Sivapithecus humeri in Pakistan and the relationship of Sivapithecus to Pongo. Nature 348, 237239.
  • Plavcan JM, Kelley J (1996) Evaluating the ‘dual selection’ hypothesis of canine reduction. Am J Phys Anthropol 99, 379387.
  • Plavcan JM, Van Schaik C (1992) Intrasexual competition and canine dimorphism in anthropoid primates. Am J Phys Anthropol 87, 461477.
  • Plavcan JM, Van Schaik CP (1997) Interpreting hominid behavior on the basis of sexual dimorphism. J Hum Evol 32, 345374.
  • Plavcan JM, Van Schaik CP, Kappeler PM (1995) Competition, coalitions and canine size in primates. J Hum Evol 28, 245276.
  • Rak Y (1983) The Australopithecine Face. London: Academic Press.
  • Rightmire GP (1998) Evidence from facial morphology for similarity of Asian and African representatives of Homo erectus. Am J Phys Anthropol 106, 6185.
  • Rightmire GP, Lordkipanidze D, Vekua A (2006) Anatomical descriptions, comparative studies and evolutionary significance of the hominin skulls from Dmanisi, Republic of Georgia. J Hum Evol 50, 115141.
  • Robinson JT (1954) The genera and species of the Australopithecinae. Am J Phys Anthropol 12, 181200.
  • Robinson JT (1956) The dentition of the Australopithecinae. Transvaal Mus Mem 9, 1179.
  • Robinson JT (1962) The origin and adaptive radiation of the australopithecines. In Evolution und Hominization (ed. KurthG), pp. 120140. Stuttgart: Gustav Fischer.
  • Robinson JT (1972) Early Hominid Posture and Locomotion. Chicago: University of Chicago Press.
  • Ruvolo M (1997) Genetic diversity in hominoid primates. Annual Review of Anthropology 26, 515540.
  • Senut B, Pickford M, Gommery D, Mein P, Cheboi K (2001) First hominid from the Miocene (Lukeino formation, Kenya). C R Acad Sci Ser II A 332, 137144.
  • Simpson SW, Lovejoy CO, Meindl RS (1991) Relative dental development in hominoids and its failure to predict somatic growth velocity. Am J Phys Anthropol 86, 113120.
  • Spoor F, O’Higgins P, Dean C, Lieberman DE (1999) Anterior sphenoid in modern humans. Nature 397, 572.
  • Spoor F, Leakey MG, Leakey LN (2005) Correlation of cranial and mandibular prognathism in extant and fossil hominids. Trans R Soc S Afr 60, 8589.
  • Spoor F, Leakey MG, Gathogo PN, et al. (2007) Implications of new early Homo fossils from Ileret, east of Lake Turkana, Kenya. Nature 448, 688691.
  • Steiper ME, Young NM (2006) Primate molecular divergence dates. Molecular Phylogenetics and Evolution 41, 384394.
  • Strait DS, Richmond BG, Spencer MA, et al. (2007) Masticatory biomechanics and its relevance to early hominid phylogeny: An examination of palatal thickness using finite-element analysis. J Hum Evol 52, 585599.
  • Stringer CB (1984) The definition of Homo erectus and the existence of the species in Africa and Europe. Cour Forschungsinst Senckenberg 69, 131143.
  • Teaford MS, Ungar PS (2000) Diet and the evolution of the earliest human ancestors. Proc Natl Acad Sci USA 97, 1350613511.
  • Thorpe, SKS, Holder RL, Crompton RH (2007) Origin of human bipedalism as an adaptation for locomotion on flexible branches. Science 316, 13281331.
  • Tobias PV (1967) Olduvai Gorge, Volume 2. The Cranium and Maxillary Dentition of Australopithecus (Zinjanthropus) boisei. Cambridge: Cambridge University Press.
  • Tobias PV (1987) The brain of Homo habilis: a new level of organization in cerebral evolution. J Hum Evol 16, 741761.
  • Trinkaus E (2003) Neandertal faces were not long; modern human faces are short. Proc Natl Acad Sci USA 100, 81428145.
  • Walker A, Leakey RE, Harris JM, Brown FH (1986) 2.5-Myr Australopithecus boisei from west of Lake Turkana, Kenya. Nature 322, 517522.
  • Ward CV, Leakey MG, Walker A (2001) Morphology of Australopithecus anamensis from Kanapoi and Allia Bay, Kenya. J Hum Evol 41, 255368.
  • Ward SC, Molnar S (1980) Experimental stress analysis of topographic diversity in early hominid gnathic morphology. Am J Phys Anthropol 53, 383395.
  • White TD, Johanson DC, Kimbel WH (1981) Australopithecus africanus: its phyletic position reconsidered. S Afr J Sci 77, 445470.
  • White TD, Suwa G, Asfaw B (1994) Australopithecus ramidus, a new species of early hominid from Aramis, Ethiopia. Nature 371, 306312.
  • White TD, Suwa G, Asfaw B (1995) Australopithecus ramidus, a new species of early hominid from Aramis, Ethiopia. Nature 375, 88.
  • White TD, Asfaw B, DeGusta D, et al. (2003) Pleistocene Homo sapiens from Middle Awash, Ethiopia. Nature 423, 742747.
  • White TD, WoldeGabriel G, Asfaw B, et al. (2006) Asa Issie, Aramis and the origin of Australopithecus. Nature 440, 883889.
  • Wolpoff MH, Hawks J, Senut B, Pickford M, Ahern J (2006) An ape or the ape: is the Toumaï cranium TM 266 a hominid? PaleoAnthropology 2006, 3650.
  • Wood BA (1981) Tooth size and shape and their relevance to studies of hominid evolution. Philos Trans R Soc Lond B Biol Sci 292, 6576.
  • Wood BA (1991) Koobi Fora Research Project IV: Hominid Cranial Remains from Koobi Fora. Oxford: Clarendon Press.
  • Wood BA (2002) Hominid revelations from Chad. Nature 418, 133135.
  • Zollikofer CPE, Ponce De León MS, Lieberman DE, et al. (2005) Virtual cranial reconstruction of Sahelanthropus tchadensis. Nature 434, 755759.