The Gracile and Robust Australians of the Pleistocene and Holocene

There is no morphological evidence to support a Chinese origin for the Pleistocene Australians (e.g., Mungo and Keilor specimens) or Indonesians (e.g., Wajak cranium). And let us remember that the very Keilor specimen that is said by Thorne (1976) to be an important part of the "gracile" population is actually shown by Thorne and Wilson (1977) to be within the modern comparative Murray Valley population in shape, but larger! Indeed, the only truly "gracile" Australian specimen is LM 1. A single specimen falling outside the general range is certainly noteworthy, but must be inconclusive. More significantly, however, LM 1 could be a juvenile. When cremated, her skull came apart at the sutures, and because of the skull's fragility matrix has been left and still covers the crucial basicranial region, which would tell us whether she was mature or not (there is a key suture, the basilar suture — strictly speaking, the spheno-occipital synchondrosis — that fuses at between 18 and 25 years of age). Should she, then, be called "Mungo girl" and not "Mungo lady"? Maybe, maybe not. But it is very, very dangerous to base a whole supposed physical (racial, geographical) type on such an equivocal specimen.

The origin of the earliest Australians remains problematic. The overall derived morphology of the early Australians is clearly modern, and it best fits in with the arrival of a modern human population, immediately from Indonesia and ultimately "Out of Africa", that had left Africa some 40,000 years earlier. The physical and genetic migration of this modern human African population throughout Eurasia reached Australia by 60,000 or perhaps 45,000 years ago, earlier than any modern human occupation of Europe.

So what of the proposed phylogenetic relationship between the "robust" Australians and the more relict archaic H. erectus populations from

Indonesia? It is after all the later "robust" Australian material that is most frequently cited as supporting a multiregional paradigm. The features said to help define an Indonesian and Australian Pleistocene clade can be broken into two distinct matrices. The upper face, including the frontal squama, supraorbital, and facial prognathism, can be explained in terms of developmental processes associated with either (1) neuro-orbital disjunction in H. erectus or (2) degrees of cranial flexure and sphenoid lengthening in the "robust" Australians (and other Holocene "robusts" worldwide). The remaining features, including a developed nuchal torus, large postcanine dental complex, and a large robust mandible (all of which characterize all Aboriginal Australians anyway, especially the Pleistocene ones), can be explained in functional terms associated with the masticatory complex, which evolved within the context of adapting to prevailing conditions within Australia.

The upper face of the "robust" late Pleistocene/early Holocene Australians and the Indonesian middle Pleistocene populations has most frequently been used as a derived "complex" uniting these two groups, to the exclusion of all others. This is particularly true of the long, flat frontal and developed supraorbital region (Thorne, 1976; Thorne & Wolpoff, 1981, 1992; Wolpoff et al., 1984; Wolpoff, 1999). These two features, however, are intricately linked in developmental terms, for both cranial and facial developmental processes will readily influence the form of the frontal bone.

The supraorbital region of the Kow Swamp individuals is superficially similar to those observed in the Sangiran specimens, with developed and continuous brow ridges. They are, however, even more similar to Middle Pleistocene African fossils such as Kabwe (Zambia) and European ones such as Petralona (Greece), relative to H. erectus (Figure 10.4). Indeed they are well within the range of early and extant H. sapiens. The upper facial morphology observed in the "robust" Australians is not a primitive retention but a derived modern human feature, and any similarity is illusory. Given that much of the discussion by the multiregionalists has been based on an examination of the Kow Swamp and Sangiran specimens, especially Sangiran 17 (Thorne & Wolpoff, 1981; Wolpoff et al., 1984), let us review the upper facial morphology of these two groups (Figure 10.5).


The supraorbital is strongly developed and there is a slight inferior depression at its midregion at glabella (Rightmire, 1990). The supraorbital projects anteriorly over the face and is a dominant feature of the face, associated with

Figure 10.4 ► (Top left) Homo heidelbergensis specimen from Petralona, Greece. (Top right) Homo heidelbergensis specimen from Kabwe, Africa. (Bottom left) Modern human from Kow Swamp, Australia. (Bottom right) Homo erectus specimen Sangiran 17 from Indonesia. (See text for details.)

Adapted from Rightmire (1990).

Figure 10.5 ► (Top) A modern human from Kow Swamp (KS1) and (bottom) the early H. erectus Sangiran 17 specimen from Indonesia.

Adapted from P. Brown (1981).

a developed frontal sinus complex (Thorne & Wolpoff, 1981). The torus deepens toward the lateral part of the orbit. This is also true of the Indonesian Middle Pleistocene (perhaps Upper Pleistocene?) Ngandong and Sambungmacan hominins (Santa Luca, 1980; Delson et al., 2001; Marquez et al., 2001). The frontal has a developed but small supraorbital sulcus (Thorne & Wolpoff, 1981). The supraorbital is only weakly arched. Glabella would have been more developed, though preservation has generally deflated this region (Thorne & Wolpoff, 1981; Rightmire, 1990). The frontal bone is broad and flat, with moderate postorbital constriction, though there is no midsagittal keel (Thorne & Wolpoff, 1981; Rightmire, 1990). The midface is moderately prognathic, relative to the projecting supraorbital torus (Thorne & Wolpoff, 1981; Wolpoff et al., 1984).

Kow Swamp

The supraorbitals, compared to most living humans, are developed, though relative to Indonesian H. erectus, they are reduced. Unlike H. erectus they tend to be separate arched tori above each orbit, not thicker (deeper) laterally, and the supraorbital region does not project beyond the upper face (see P. Brown, 1981; D.E. Lieberman, 1995). Most Kow Swamp specimens do not have a supraorbital sulcus (Wolpoff et al., 1984; Habgood, 1989), the only exception being Kow Swamp 7 (Wolpoff et al., 1984). Like all modern and more archaic humans, Kow Swamp specimens have a developed frontal sinus complex (primitive hominid feature). Glabella is undeveloped (P. Brown, 1981). As in the Indonesian hominins, the frontal bones are broad and flat, with some degree of postorbital constriction (P. Brown, 1981). The midface is prognathic (Thorne, 1976; Thorne & Wolpoff, 1981), but not to the degree observed in the middle Pleistocene hominins. Indeed, their range of facial prognathism is well within the range of many other terminal Pleistocene non-Australasian "robust" groups (P. Brown, 1992; partially Groves & Thorne, 1999).

The upper face of these hominins is distinct. Over the considerable time that separates these populations (at least 1 million years) there is a marked reduction of robusticity, while still retaining the developed supraorbital region and flat frontal, so defining a general evolutionary trend between the "robust" Australians and the Indonesian H. erectus populations, to the exclusion of all others according to Thorne & Wolpoff (1981). (See also Wolpoff et al., 1984; Thorne, 1984; Wolpoff, 1989).

Supraorbital development has been shown not to be under overt strain during mastication; its development is not related to masticatory demands. The original model of supraorbital formation proposed by Weidenreich (1941) has been reinvoked by a number of researchers (Moss & Young, 1960; Biegert, 1963; Shea, 1985, 1993; Ravosa, 1988; Hylander et al., 1991; Hylander & Rovosa, 1992; C. Wood, 1994; D.E. Lieberman, 1998). It is now commonly argued that the development of the supraorbital in most nonhuman primates, including early hominins, is associated with neuro-orbital disjunction. This is when the face is pushed out from the braincase, as in H. erectus; the brain-case, including the anterior cranial fossae that house the frontal lobes of the brain, is positioned posterior to the face, so the frontal lobes are long and relatively low (Broadfield et al, 2001). This is associated with the moderate degree of postorbital constriction and often a frontal sulcus. In other words, the robust supraorbital of H. erectus is the result of frontal bone drift, which is required to bridge the space between the upper face and the more posterior braincase. Thus, the supraorbital torus is a structural "supporting beam" connecting the large and posteriorly orientated braincase to the face (Hylander & Ravosa, 1992; D.E. Lieberman, 2000; Cameron, in press a).

The condition in H. sapiens over the last 170,000 years or so is that the anterior cranial fossa has moved anterosuperiorly, and so the frontal lobes are located directly above the orbits, as opposed to the more posterior position of H. erectus. This results in a vertically oriented frontal bone, with the face positioned directly beneath, reducing the supraorbital region and the frontal sulcus. This can be referred to as neuro-orbital "convergence." This is clearly the way it is in Kow Swamp individuals (Figure 10.6), even taking into account those individuals whose frontals have been affected by intentional cranial deformation; indeed, individuals that are not influenced much at all by intentional deformation show that the anterior cranial fossae are located directly above the orbits in the typical modern H. sapiens fashion, and not posterior to them as in H. erectus (see also radiographs published in D.E. Lieberman, 1995, Figure 2). As will be discussed presently, the formation of the supraorbital in these modern human populations is also associated with an increase in sphenoid length and cranial base extension (see D.E. Lieberman, 1998). Therefore, the suggested 'similarity' between these two populations is the result of two very different processes, one biological the other cultural; i.e., they clearly arose independently.

Large supraorbitals are known to occur in a number of recent H. sapiens, especially in large robust males with narrow skulls, including some northern Europeans and African Bushmen (D.E. Lieberman, 2000), and in a number

Figure 10.6 ► (Top) A modern human from Kow Swamp (KS1) showing the modern human condition of neuro-orbital convergence, as opposed to (bottom) the more archaic condition of neuro-orbital disjunction observed in the early H. erectus Sangiran 17 specimen. (See text for details.)

of fossil H. sapiens from Africa (Dar es-Soltan), Western Asia (Skhul V, Qafzeh), Europe (Madlec, Predmosti, Cro-Magnon), and East Asia (Zhoukoudian Upper Cave specimen 101) (see Wu, 1961; F.H. Smith & Raymond, 1980; Bräuer, 1984; Stringer et al., 1984; Kamminga & Wright, 1988; Corrucini, 1992; Etler & Li, 1994; Wu & Poirier, 1995; D.E. Lieberman, 2000). The robust supraorbital is a primitive hominin feature and is not evidence for regional continuity (see Groves, 1989b). Its development in these H. sapiens is a result of increased anterior cranial base angu-lation, and thus it is distinct from the neuro-orbital disjunction, which is its origin in H. erectus and other earlier hominins. It also explains the increase in midfacial prognathism, which is a by-product of cranial base angulation (see D.E. Lieberman, 1998, 2000).

The degree to which the cranial base is flexed will affect the development of the supraorbital and will directly influence the degree to which the mid-face is prognathic. A more flexed cranial base will position more of the face beneath the anterior cranial fossa; a more extended cranial base will position more of the face in front of the fossa (D.E. Lieberman, 2000), associated with an increased length of the sphenoid (D.E. Lieberman, 1998).

The sphenoid is the central bone of the cranial base, from which the face grows anteriorly during ontogeny (D.E. Lieberman, 1998); with increased sphenoid length, the upper face will move forward, which in turn results in increased supraorbital development (see Enlow and Hans, 1996). A similar pattern can be observed in some Neanderthal specimens, and it is due to similar processes (D.E. Lieberman, 1998).

Finally, Westaway and Cameron (submitted) have tested the suggested affiliations between the erectine populations of Asia and the "robust" populations of Australia using parsimony analysis (see Chapter 5 for discussion of this method). Because we were testing the proposed relationship of the Ngandong populations (Indonesia) and the Kow Swamp people (Australia), and because the Ngandong fossils are represented by cranial specimens only (i.e., no facial specimens), we used 27 cranial features preserved in the Ngandong fossils. The resulting analyses are shown in Figures 10.7 and 10.8. Figure 10.7 is based on a strict consensus from

Figure 10.7 ► Strict consensus tree of 27 cranial features. (See text for details.)
0 0

Post a comment