Jeffrey H. Schwartz The chin revisited: what is it and Department of Anthropology, who has it? University of Pittsburgh, Pittsburgh, PA 15260, Although the presence of a ‘‘chin’’ has long been recognized as U.S.A. E-mail: unique to Homo sapiens among mammals, both the ontogeny and the [email protected] morphological details of this structure have been largely overlooked. Here we point out the essential features of symphyseal morphology in Ian Tattersall H. sapiens, which are present and -defined in the fetus at least as Department of Anthropology, early as the fifth gestational month. Differences among adults in American Museum of Natural expression of these structures, particularly in the prominence of the History, Central Park West at mental tuberosity, are developmental epiphenomena and serve to 79th Street, New York, emphasize the importance of studying this region in juveniles when- NY 10024, U.S.A. ever possible. A survey of various middle to late Pleistocene E-mail: [email protected] hominids for which juveniles are known reveals that these features are present in some late Pleistocene specimens assigned to H. sapiens, but Received 7 May 1998 not in all of the presumed anatomically modern H. sapiens (i.e., Revision received 28 April Qafzeh 8, 9, and 11). The adult specimens from Skhu¯l, as well as the 1999 and accepted 26 May adult Qafzeh 7 specimen, are similarly distinctive in symphyseal 1999 morphology. are quite variable in their own right, and they as well as other middle to late Pleistocene lack the Keywords: chin, Homo sapiens, symphyseal features of H. sapiens. Some of the latter are, however, Neanderthals, Skhu¯l, seen in the Tighenif (Ternifine) . Qafzeh.  2000 Academic Press

Journal of (2000) 38, 367–409 doi: 10.1006/jhev.1999.0339 Available online at http://www.idealibrary.com on

Introduction characteristics became incorporated into scenarios of evolution. With regard to the More than two centuries ago, Blumenbach chin, which essentially became synonymous (1775) identified the ‘‘chin’’ as being among with ‘‘a prominence at the front of the the most fundamental of the features that he ,’’ the general concern among considered uniquely human. Of course, paleoanthropologists was with how an ape- when Blumenbach presented his discussion like could be converted to that of Homo of ‘‘what distinguished ,’’ it was in sapiens (Hrdlicˇka, 1911; Robinson, 1913; the context of ‘‘compared to other extant Wallis, 1917; Gregory, 1922; Howells, .’’ But with the acceptance in the 1959; Enlow, 1982; Stringer et al., 1984; nineteenth century of hominid fossils as real Daegling, 1993). entities, and the subsequent documentation Discussion of the human chin has been of a well-represented and diverse human approached from several different perspec- fossil record, features such as the chin, tives. Anatomical descriptions have focused which had a specific kind of significance on the protrusiveness of the symphyseal when the primary goal of the taxonomist region, with the mounded structure being was to distinguish between humans and identified in extant humans as the ‘‘mental other extant primates, became cast in a protuberance’’ (Johnston & Willis, 1954), different light. Rather than being static enti- the ‘‘mental eminence’’ (Lieberman, 1995), ties, as would be expected of organisms that the ‘‘mental osseum’’ (Bra¨uer, 1984; Rosas, had been specially created, morphological 1995), or the ‘‘tuber symphyseos’’ (Hublin

0047–2484/00/030367+43$35.00/0  2000 Academic Press 368 . .   . 

& Tillier, 1981; Inke, 1967). Thus defined, It is the basal protrusion which has, of the may continue course, been the primary criterion for laterally on each side to some extent as a identifying the mental protuberance; and thickening of the inferior margin of the this, in turn, has become the primary ele- corpus. In some individuals, each lateral ment in the identification of a chin. From extremity of the mental protuberance may this perspective, it is perfectly reasonable to bear a blunt ‘‘corner,’’ which has been conclude that ‘‘chins are a notoriously vari- referred to as a ‘‘’’ (Johnston able character, as many modern humans & Willis, 1954; Rak, 1998)or‘‘tubercula lack them’’ (Lieberman, 1995:174). And lateralia’’ (Hublin & Tillier, 1981; Inke, when the focus of attention in this region 1967). Above each lateral extremity and to is the degree of symphyseal protrusion each side of the mental protuberance, lies a anteriorly—which is often judged against the depression, or ‘‘mental fossa’’ (Wolpoff, vertical plane of the or incisors (see 1996), which would appear to exist as a Enlow, 1982)—there might, indeed, appear default of the midline field of being to be ‘‘chinless’’ and ‘‘almost chinless’’ elevated or protrusive. Collectively, the human beings. mental protuberance and lateral extremities, The evolution of the human chin has been especially if they bear tubercles, have been discussed in various ways. Hrdlicˇka (1911) identified as the ‘‘mental trigon’’ (Rak, ventured the suggestion that this structure 1998), the ‘‘mental trigone’’ (Smith, 1984), evolved as a result not of the specific devel- or, if the structure is restricted quite opment of a protrusion, but of the reduction inferiorly along the corpus, the ‘‘trigonum of the dental arch (which essentially left the basale’’ (Bra¨uer, 1984). Just below the protrusion behind as the anterior margin alveolar margin, the bone may be curved retreated). This idea was elaborated upon by inward, presenting a shallow depression that Robinson (1913), who proposed that it was has been referred to as the ‘‘incurvatio the demands of the and laryngeal mandibularis’’ (Hublin & Tillier, 1981), the musculature that maintained mandible ‘‘incurvatio mandibulae’’ (Rosas, 1995), or length inferiorly as the region above it the ‘‘impressio subincisiva externa’’ (Inke, supposedly retreated posteriorly. Gregory 1967). Although there is obviously much (1922) added to these notions the possibility morphological detail to be recognized in the that the chin provides space at the front of symphyseal region of H. sapiens, the human the jaw that is necessary for the enlarging chin has frequently been identified solely crowns and roots of incisors and canines. on the basis of the general protrusiveness of Howells (1959) reiterated Robinson’s the symphyseal region (Lieberman, 1995; (1913) proposition and suggested further Smith, 1984; Wolpoff, 1980; Lam et al., that bone thickening, which ultimately pro- 1996). duced the chin, must have evolved in order The development of the human chin is to reinforce the front of the jaw against typically described as a postnatal phenom- masticatory stresses (which, in apes, were enon, a simple matter of absorbed by the internal buttress known as near the alveolar margin, with concomitant the ‘‘simian shelf’’). Reisenfeld (1969) deposition of bone more inferiorly (Enlow, thought that he had refuted this kind of 1982; see also Aiello & Dean, 1990; functional explanation by inducing in mice Daegling, 1993; Lieberman, 1995). Individ- what he believed to be a chin by extracting ual differences or variations in chin develop- the continually growing anterior teeth and ment result essentially from differential rates severing the muscles of mastication (which of resorption vs. deposition (Enlow, 1982). did, of course, have the effect of eliminating   ‘‘’’ 369 masticatory stress at the symphyseal region). a conspicuous incurvatio mandibulae’’ Wolpoff (1980) adopted a different stance (Rosas, 1995:550). on the subject by speculating that the The impetus for the present review came appearance of the chin was correlated with from two different and unexpected sources. expansion of the neurocranium and a con- First, in the course of conducting a twenty- comitant shift of the temporal muscles to a year analysis of the skeletal remains of peri- more medial angulation. Most recently, natal individuals from the Punic site of Daegling (1993) surveyed the various func- Carthage (n>650 individuals; Schwartz & tional interpretations put forth to explain the Houghton, in preparation), one of us (JHS, presence of the chin, and concluded that the unpublished data) discovered that it was best was ‘‘the hypothesis that the human possible to identify very small fragments chin represents a structural response to of the symphyseally unfused mandibles of resist vertical bending in the coronal plane.’’ these highly immature individuals on the Regardless of the proposed mechanisms basis of specific features of this region: behind the appearance of the chin as a namely, a continuous everted margin that structure, the language of the literature is proceeds along the edge of the symphysis invested with the notion of an evolutionary and laterally around along the inferior edge trend, in which a perceived increase in swell- and that, consequently, subtends a deep ing of the symphyseal region ultimately leads fossa. Second, as a result of realizing that the to the formation of the typical human chin. configuration of the brow in some of the Thus, it is common to read of the supposed most familiar of the supposed early ‘‘ana- existence of incipient chins in various tomically modern’’ H. sapiens (i.e., from Neanderthals (themselves considered Skhu¯l and Qafzeh) was not bipartite (cf. merely an archaic, primitive version of H. Stringer et al., 1984), as it is in late sapiens). These include adult specimens Pleistocene and extant representatives of from Guattari, S{ipka, and Vindija (Lam this species (Schwartz & Tattersall, 1999b), et al., 1996; Lieberman, 1995; Smith, 1984; we came to realize that these same fos- Wolpoff, 1980; Wolpoff et al., 1981) and the sils were also not H. sapiens-like in Archi child (Ascenzi & Sergi, mandibular symphyseal morphology. For 1971a,b; Mallegni & Trinkaus, 1997; the record, we must admit that when we first Stringer et al., 1984). Hence, because dis- studied the Skhu¯l specimens, we were not cussions of later human evolution are often initially struck by these disparities. But it informed by anticipation of discovering such was during our second round with these a transformation series, it is not surprising to specimens, after we had benefited from find statements such as: ‘‘[W]hile mental studying at first many more middle to eminences and trigones are markedly more late Pleistocene fossil hominids, that the developed in modern H. sapiens (even the differences in details of brow and mandibu- EMH [=, especially lar morphology became apparent between from Skhu¯l and Qafzeh] sample), a clear these specimens (as well as some specimens trend towards the EMH condition is present from Qafzeh) and extant H. sapiens. in the LN [=late Neanderthal] group’’ Here, we present the results of an (Smith, 1984:190). Even some of the larger extended developmental and morphological mandibles from the middle Pleistocene Sima overview of the symphyseal region in H. de los Huesos sample have been portrayed sapiens that focuses on distinguishing as having chinlike symphyseal regions, between those features that are invariably which have been described as displaying present from fetal to adult stages and ‘‘a well-developed trigonum mentale and those that represent individual variation. We 370 . .   .  suggest that only the constellation of mor- studied a sample of five hundred recent phological detail present from fetus to adult H. sapiens (Department of Anthropology, (in which H. sapiens appears to be strongly AMNH) and, as an outgroup, three hundred derived) should be referred to as a chin— nonhuman anthropoid primates (Depart- although we wonder, given the history of ment of Mammals, AMNH) (Table 1). The broad usage of this term, whether the reten- middle to late Pleistocene fossils hominids tion of this term as an anatomical descriptor that we have analyzed to date and that pre- is actually necessary or even constructive. serve enough of the mandibular symphyseal We also review the middle to late Pleis- region to be relevant to this discussion are tocene hominid fossils from Europe, the listed in Table 2. For the sake of simplicity, Levant, and northern Africa, with an and in order to facilitate reader recognition, emphasis not only on morphology, but on we present a rather traditional grouping of the ontogenetic context. We conclude that these specimens in Table 2, identifying them one can identify a chin, complete with conservatively either as ‘‘Neanderthal,’’ detailed morphology, in all late Pleistocene ‘‘Homo sapiens,’’ or ‘‘Other,’’ even though fossils traditionally assigned to modern H. we believe there are more satisfactory sapiens, but not in all early modern humans alternatives. from this time period; some of the latter As is usually the case with fossils, most of possess a configuration that differs from the available specimens are of adult individ- those of both modern H. sapiens and H. uals. Nonetheless, the later human fossil neanderthalensis. In addition, the Tighenif record does contain an important represen- mandibles provide further evidence that it is tation of younger individuals. The youngest the specific morphology of the symphyseal fossil individual in our sample is the region, rather than its simple protrusion or 10-month-old Neanderthal from Amud lack thereof, that is potentially phylogeneti- (Rak et al., 1994). The probable age cally and systematically significant. range for the Skhu¯l child and the typical Neanderthal children [from (Devil’s Tower), Peche de l’Aze´, and Roc de Materials Marsal] is 3–4 years (see Dean et al., 1986; Our developmental series of extant H. Zollikofer et al., 1995), although an older sapiens from second trimester through birth age of 5–6 years has also been suggested are in the permanent study and teaching (Tillier, 1983). collections of the anatomy department of the We shall first discuss the morphology and University of Pittsburgh School of Dental development of the symphyseal region in Medicine (UPD; n=10) and the American extant H. sapiens and of those specimens that Museum of Natural History (AMNH; n=8). are thought of as closely resembling them. These series are comparable in detail and We then turn to Neanderthals, and, finally, essentially identical to collections in medical to the remaining middle to late Pleistocene schools (e.g., see the series at Washington fossil hominids. We refer to the central and University) and are consistent with the lateral incisors as I1 and I2, and the anterior descriptions and illustrations in standard and posterior premolars as P1 and P2. textbooks (e.g., Johnston & Willis, 1954). Since we are not presenting these data as Results and discussion novel discoveries, but are rather using them as the basis for our descriptons and compari- Extant Homo sapiens son, corroboration of our descriptions is In adults (Figure 1), the symphyseal region readily available. One of us (JHS) also bears a variably long and raised vertical keel   ‘‘’’ 371

Table 1 Human and other anthropoid specimens that lies along the midline. This keel, which at the American Museum of Natural originates at or just below the alveolar History used in this study* margin, becomes more prominent inferiorly Nonhuman anthropoid primates as it fans out laterally along the somewhat New and Old World Monkeys distended inferior margin of the mandible. Ateles geoffroyi (n=19) Cebus apella (n=20) In some individuals it terminates in variably Cercopithecus cephus (n=19) thickened, but not discretely separate, C. mona (n=1) ‘‘corners’’ (the lateral tubercles). When C. neglectus (n=2) present, each blunt terminus of the dis- C. pogonias (n=7) Colobus polykomos (n=6) tended inferior margin lies well anterior to Macaca fascicularis (n=20) the , which, in turn, lies Mandrillus sphinx (n=12) variably anterior to the second premolar (see Papio anubis (n=1) P. cynocephalus (n=7) also Aiello & Dean, 1990; Daegling, 1993; Presbytis entellus (n=3) Lieberman, 1995). Together, the symphy- Hominoids seal keel and the distended inferior margin Hylobates agilis (n=39) form an inverted, roughly ‘‘T’’-shaped H. hoolock (n=14) structure; it is at the junction of the keel and Gorilla gorilla (n=21) Pan troglodytes (n=35) distended inferior margin that a mental pro- Pongo pygmaeus (n=14) tuberance is identified. A variably shallow, scallop-shaped depression (mental fossa) Recent Homo sapiens Africa (sub-Saharan; n=71) lies to each side of the central keel. These East Asia (n=80) mental fossae emphasize the keel as well as China (n=25) the distended inferior mandibular margin Korea (n=6) Japan (n=13) below. As can be observed even among the Ainu (n=4) handful of specimens illustrated in Figure 1, Siberia (Indian Point) (n=32) variation exists among individuals in the Australia Aborigine (n=26) degree to which these various features are Europe (n=70) expressed. However, the essential aspects of Hungary (n=9) shape, location, and morphological detail of Czech—Bohemia (n=10) the keel, protuberance, distended inferior Austria (n=10) (n=13) margin, and mental fossae, are maintained Italy (n=8) throughout. ‘‘Yugoslavia’’ (n=10) The constellation of consistent features Greece (n=10) India (n=26) just described for the adult is present (and North America (Arctic; n=139) its details crisply delineated) by at least the Thule (n=33) fifth fetal month (Figure 2). Although the Near Ipiutak (Point Hope, Alaska) (n=12) Ipiutak (Point Hope, Alaska) (n=46) halves of the mandible are still unfused Tigara (Point Hope, Alaska) (n=48) across the midline at this gestational age, North America (Sub-Arctic, Southwest; n=51) one can clearly observe that the vertical Native American, Utah, New Mexico (n=30) symphyseal margin of each half is raised or Arizona (n=21) South America everted, and that this vertical ridge is con- Bolivia, Chile, Peru (n=41) fluent with a similarly everted inferior margin that diminishes in prominence *A complete list of specimens is available upon posteriorly. Each everted marginal cresting request. system defines a deep fossa. During subsequent intrauterine growth, the two mandibular halves come together at the 372 . .   . 

Table 2 Middle to late Pleistocene hominid mandibular specimens used in this study

Age Specimen Country (ka B.P.)

Homo sapiens Abri Pataud 22·0 Brno 2, 3 Czech Republic 26·0 Chancelade France 17·0? Cro-Magnon 1, 2, 3 France ca. 30? Dolní Ve˘stonice III, XII, XIV, XV, XVI Czech Republic 26–29 Engis 1 (adult) Belgium ? Grimaldi (Grotte des Enfants, Balzi Rossi) Italy ca. 36 Hahnofersand Germany ca. 36 Isturitz (includes children) France 18–30 Oberkassel (male, female) Germany 15? Ohalo 1, 2 ca. 19 Pr˘edmost´ı 3, 4 (casts) Czech Republic ca. 26 Qafzeh 1, 3, 6, 7, 8, 9, 11 (subadult) Israel ca. 92–95 Skhu¯l I (child), II, IV, V, VI, VII Israel 80–100+ Svitavka Czech Republic 25–30? Vogelherd 1, 2 Germany ca. 32 Neanderthal Amud (child and adult) Israel 40–50 Archi (child) Italy ? Ehringsdorf 1009/69, 1010/69 Germany ca. 320? Engis (child) Belgium ? Feldhofer Grotto Germany ca. 40 (child) Gibraltar 50? Grotta Guattari 2 and 3 Italy 50–60+ Kebara (child and 2) Israel ca. 60 51, 53, 54, 55, 56, 57, 59 ca. 130 La Chapelle-aux-Saints France 50+ [adult (5) and child] France ca. 65 France 60–70? (subadult) Germany ca. 40 Montmaurin France 150–200? Ochoz Czech Republic ? Pech de l’Aze´ (child) France 45–55 Regourdou France ? Roc de Marsal (child) France 50+ Saccopastore Italy 120–130 Saint-Ce´saire France 36 S{ipka (juvenile, cast) Czech Republic ? Spy 1 and 2 Belgium ? Subalyuk (adult) Hungary 60–70? Tabu¯n C1 Israel 34 Tabu¯n II Israel ? Vindija 75/206, 76/231, 76/226 Croatia ca. 40+ Zafarraya ca. 30 Other Arago 2 France 400? Biache 2 (cast) France 150–175? Bilzingsleben Germany 280–400 3 (cast, child) Morocco 90–125+ + Mauer Germany ca. 500 Rabat (Thomas Quarry) Morocco ca. 400 Tighenif (Ternifine) 1, 2, 3 Algeria ca. 700?   ‘‘’’ 373 symphysis, with the inverted ‘‘V’’-shaped fossa) and, at the same time, the potential gap between them closing in a supero- enlargement of other features (such as the inferior direction (i.e., from the region of the mental protuberance and lateral tubercles), presumptive alveolar crest down to the that results in the variation we see among inferior margin of the jaw) (see also illus- adults. Further, any increase in depth of the tration in Johnston & Willis, 1954:296). At mandible with concomitant expansion of this stage of growth, the mental protuber- the domain of alveolar bone (Enlow, 1982) ance is represented merely by the juncture of would account for the absence in some the inverted ‘‘T’’ that emerges with the individuals of any trace of the keel near the formation of a single midline symphyseal alveolar border—precisely because alveolar ridge and its intersection with the everted bone is a cellular derivative of tooth- inferior margin. In individuals in whom the producing stem cells and not of presumptive two mandibular halves have closed up but mandibular osteoblasts (e.g., see Ten Cate have not totally obliterated the symphyseal & Mills, 1972). gap, one can sometimes observe a shallow If we apply the term ‘‘chin’’ to the devel- inverted ‘‘V’’ indenting the inferior margin opment of a prominent mental protuberance at its midline (Figure 2). It is also important alone, then those (e.g., Lieberman, 1995) to note that the everted margins and the who describe some humans as lacking a mental fossae retain their integrity through- chin, as well as others (specifically, Enlow, out growth in spite of the bone alteration 1982) who define chin development as the that takes place around them in conjunction resorption of bone near the alveolar margin with tooth crown and crypt expansion. with concomitant deposition of bone more Postnatally, the features that are so well inferiorly, are, in general, correct. But, as we delineated in the fetus become softened and illustrate here, the size of the region of the less sharply defined (Figure 3). In two-year- mental protuberance, which can vary from olds, for instance, the symphyseal keel is one individual to the next, is independent of faint below the alveolar region. The keel is those features—the inverted ‘‘T’’ and its even less crisply defined in five-year-olds, in attendant mental fossa—that are established whom the mental fossae are also shallower fetally. In this context, Enlow’s observation as well as less expansive. The mental protu- (i.e., that differential bone resorption superi- berance of postnatal individuals is, however, orly in the symphyseal region, along with more prominent than in perinatal individ- bone deposition more inferiorly creates a uals, not only as a forward projection, but protuberance that projects farther forward also as a generally triangular eminence with than the alveolar region), is exactly what one its base seated along the inferior margin of would expect in the development of those the jaw (forming the so-called trigonum features that vary individually in adults, but basale). When viewed from below, the front not in the formation of the symphyseal of a young child’s mandible is quite thick details seen in the fetus. anteroposteriorly (=buccolingually) in the area subtended by the variably developed Late Pleistocene Homo sapiens lateral tubercles, although immediately The inverted ‘‘T’’ and associated mental lateral the bone is thinner (also see metric fossae are found in specimens from the late analysis by Zollikofer et al., 1995). In Pleistocene sites of Barma Grande, Brno, addition to an overall enlargement of the Chancelade, Cro-Magnon, Dolní Ve˘stonice, mandible, including in its superoinferior Grimaldi, Isturitz, Oberkassel, Ohalo, depth, it is the continued softening or blunt- Pr˘edmost´ı, and Svitavka (see Figure 4). In ing of some features (such as the keel and many of these individuals, 374 . .   .    ‘‘’’ 375 the mental trigon/protuberance is expansive, vations are interesting in light of the fact that as well as protrusive, and often (e.g., Brno, of this group only Qafzeh 9 and 11 may have Pr˘edmost´ı, Okerkassel) exceeds that seen in borne bipartite brows, whereas the supra- the most markedly endowed individuals in orbital margins of the adult specimens our sample of extant H. sapiens. The Isturitz from Skhu¯l, and of Qafzeh 6, are defi- sample, which contains a number of nitely smoothly continuous (Schwartz & juveniles of varying ages, as well as old Tattersall, 1996b). adults, is particularly interesting in terms of The intact symphyseal region of the what is apparently a populational develop- Qafzeh 11 subadult bears a long, thin ment variant: the ‘‘cleft chin.’’ As seen in central keel that begins its gentle rise a Figure 5, individuals from this site have an short distance below the alveolar margin inverted midline ‘‘V’’-shaped cleft that parts (Figure 6). Toward the inferior margin of the inferior margin with varying degrees of the mandible the keel fans out bilaterally into openness and upward extension. As inferred a short, broad, and somewhat protruding from our study of extant H. sapiens, the triangular mental protuberance, which be- simplest and developmentally most reason- comes most protrusive just prior to curving able interpretation is that in these individ- into the less distended inferior margin. Shal- uals from Isturitz, as well as in others from low mental fossae are also discernible on the site not illustrated here, coalescence of either side of the keel, and the mental fo- the mandibular halves did not proceed fully ramen lies just anterior to the root of P2.In down from the alveolar crest to the inferior inferior view, although the front of the jaw is margin. Symphyseal clefting among individ- not squared off as it can be in H. sapiens, the uals simply reflects the differing degrees to symphyseal region is more thickened antero- which the symphysis fused along its length. posteriorly than the short stretch of bone on Aside from this developmental peculiarity, each side of it, posterior to which the corpus, however, in every other detail of mor- in turn, becomes thicker. phology, the Isturitz mandibles are typically The symphyseal regions of the adults H. sapiens in possessing the inverted ‘‘T’’ Qafzeh 8 and 9 are, unfortunately, some- and mental fossae. what damaged (Figure 6). However, in Qafzeh 9 the bone on each side of the Middle to Late Pleistocene Homo sapiens damaged midline appears to have been from the Levant inflected outward, and the inferior margin of In this sample, only the adults Qafzeh 8 and the symphyseal region would have been 9 and the subadult 11 present any indication somewhat projecting. In the even more of having had an inverted ‘‘T’’ and mental poorly preserved Qafzeh 8 mandible there fossae (Figure 6). The adult Qafzeh 7 does is in concert definite evidence of a raised not possess these features, nor does any mental protuberance lying close to the individual from the broadly contempor- slightly distended inferior margin. The aneous site of Skhu¯l, including the child, latter two features together present the Skhu¯lI(Figures 7 and 8). These obser- configuration of an inverted ‘‘T.’’

Figure 1. Mandibles of recent adult Homo sapiens to illustrate the basic configuration of the mental protuberance and trigon. Top, UPD #95 (provenence unknown); middle, AMNH 99/108 (Thule, Smith Sound, Alaska); bottom, UPD #66 (Indo-Pakistan?). Although the degree to which these features are pronounced, all three possess paired mental fossae and a clearly defined central keel that thickens at the mental tuberosity and fans out along the margin of the symphyseal region, terminating bilaterally in variably thickened ‘‘corners.’’ Bar=1 cm. 376 . .   .    ‘‘’’ 377

In contrast to Qafzeh 8, 9, and 11, Qafzeh they are depressions, they leave between 7(Figure 7) presents a distinctive symphy- them what appears to be a small elevated seal configuration. Immediately below the region of bone low down in the midline. As alveolar crest lies a superoinferiorly short seen in profile, however, this midline struc- but moderately deep depression, which is ture does not protrude beyond the normal laterally confined to the region of the incisor surface of the bone itself. Posteriorly, as roots. The remainder of the symphyseal preserved especially well on the right side, region is adorned with a low, teardrop- the depression extends below a swollen shaped bulge. The apex of this bulge arises region beneath the dm2 that must corre- within the subalveolar depression and its spond to the enlarging P2 in its crypt. These body broadens toward the inferior margin of depressions are strongly reminiscent of the mandible. The sides of this bulge slope those seen in the Archi 1 Neanderthal child gently and smoothly into the contour of the (see discussion below and Figure 12). surrounding bone of the corpus. The inci- Superiorly, the bone around the root of the sors overhang the subalveolar depression not fully erupted right permanent central and the bulge presents a continuous and incisor is markedly indented or creased. By gently backwardly tilted profile. Conse- default, this creates the impression that the quently, when seen in profile, the subalveo- bone around this depression is elevated, lar depression represents the space between although in reality it is actually a smooth, the somewhat procumbent incisor region unembellished surface. Clearly, Qafzeh 4 above and the longer slope of the symphy- does not display the salient features of the seal bulge below. In inferior view, although symphyseal region typical of H. sapiens, the latter bulge is noticeable, it is also the which are especially well defined in young case that, buccolingually, the bone of the children. As such, Qafzeh 4 cannot be corpus is fairly uniformly thick throughout regarded as a juvenile precursor of the more its length. The mental foramen lies under mature Qafzeh 8, 9, and 11. Perhaps it the region of P2. reflects the juvenile state of the adult Tillier (1979) has described the mandible represented by Qafzeh 7. of the juvenile (6–8 years?) Qafzeh 4 as The adult specimens from Skhu¯l that pre- having ‘‘modern’’ symphyseal features, serve the anterior portion of the mandible including mental fossae, a mental trigon, (II, IV, and V) are not similar to extant H. and lateral tubercles. The illustration she sapiens and related fossils. Rather, they provides portrays a reconstructed symphy- resemble Qafzeh 7 (Figure 7). Although seal region that preserves a reasonable there is individual variation in the degree to amount of identifiable morphology. Of note which the symphyseal region is broad and is the presence of two well-developed deep (it is most accentuated in Skhu¯l IV, less depressions that lie virtually at the inferior so in Skhu¯l V, and least in Skhu¯l II), each margin of the jaw, each on either side of the specimen possesses some amount of sub- midline, with each reaching its maximum alveolar depression as well as some swelling height well lateral to the midline. Because of the symphyseal region. In all three

Figure 2. Fetal specimens of Homo sapiens to illustrate the early stages of development of the mental trigon. Top right, 5 months (AMNH 99/7321); top left, 6 months (AMNH 99/7322); middle, 7 months (AMNH 99/7323); bottom, 8 months (AMNH 99/7324). The right and left contributions to the central keel, which are seen coalesced or coalescing across the symphysis (from top to bottom) in older specimens, are well-developed in the 5 month fetus. In all, the raised inferior margin and mental fossae are also clearly differentiated. Bar=1 cm. 378  .  .    . 

Figure 3. Mandibles of recent juvenile Homo sapiens to illustrate the essential features of the symphyseal region. Top, left (anterior view) and right (inferior view), 2 years old (AMNH, Education Dept, uncatalogued); bottom, left (anterior view) and right (inferior view), 5 year old (AMNH 99/7799). The central keel, which broadens at the mental tuberosity, fans out inferiorly, terminating bilaterally in blunt ‘‘corners.’’ The inferior views demonstrate the thickness of the mandible in the symphyseal region and the impact of the marginal attributes of the mental protuberance on mandibular shape. Bar=1 cm.   ‘‘’’ 379

Figure 4. Mandibles of late Pleistocene Homo sapiens. Top: Cro-Magnon 3 [Muse´edel’Homme (MH)]; middle: Abri Pataud (MH); bottom: Pr˘edmost´ı 3 (cast, Anthropos Institute, Brno). Although there is variation in degree of expression of the inverted ‘‘T’’-shaped configuration of the symphyseal region, all specimens possess a central keel, thickened inferiorly placed mental protuberance, distended inferior margin, and mental fossae. Bar=1 cm. 380 . .   . 

Figure 5. Mandibles of late Pleistocene H. sapiens from Isturitz. Top: Isturitz 68 [Muse´e des Antiquite´s Nationales (MAN)]; bottom: Isturitz 70 (MAN). Each displays a low central keel, thickened inferior margin, and mental fossae. Each also possesses a cleft indenting the inferior margin at the midline that corresponds to the incompletely fused symphysis seen in fetuses (see Figure 2). Bar=1 cm. specimens, the inferior margin of the sym- right of the midline, as well as by an alveolar physeal region is elevated somewhat relative margin that was reconstructed to overhang to the plane of the inferior margin of the rest the symphyseal region too severely; and to of the corpus. In the less complete Skhu¯lII the left of the midline, the inferior margin is mandible, a definite subalveolar depression also damaged. The mental foramen lies lies immediately below the incisor alveoli; under P2 in Skhu¯l II and V, and under M1 in inferiorly, the bone of the front of the jaw is Skhu¯l IV. In addition, in Skhu¯l V, a distinct gently rounded. In the more complete Skhu¯l and peaked inferior marginal tubercle lies IV and V mandibles, a configuration com- below the level of the first molar and thus parable to that observed in Qafzeh 7 is more posterior to the mental foramen; a similar clearly discernible: there is a subalveolar structure lies slightly more anteriorly in depression, from which emanates a low, Skhu¯l II, beneath the mental foramen. In teardrop-shaped bulge that flows into the inferior view, and regardless of the degree of surrounding bone (rather than an inverted swelling of the symphyseal bulge, the bone ‘‘T’’ delineated by mental fossae). In Skhu¯l of the corpus of each specimen is essentially V, the subalveolar depression is exaggerated uniformly thick buccolingually throughout by crushing, especially above and to the its length.   ‘‘’’ 381

If one were to define the possession of a essentially uniformly thick in inferior view, chin purely on the basis of the development with maximum thickness being achieved of a symphyseal protrusion or bulge, without farther back along the corpus. In addition, taking the details of morphology into con- since on the right side the bone extends sideration, then one would certainly identify externally almost as far as the symphysis as a chin in Qafzeh 7 and all adult Skhu¯l well as superiorly to a point significantly specimens (e.g., McCown & Keith, 1939; below the alveolar crest, it is evident that this Smith, 1984; Stringer et al., 1984; Wolpoff, specimen lacked a mental fossa, at least on 1980, 1996). Clearly, however, while in that side. Since this bony surface is not these specimens the symphyseal bulge can concave with an increasingly everted medial plausibly be assumed to have resulted from contour, but is essentially flat, it is equally bone deposition and the subalveolar depres- clear that the symphyseal region bore sion from bone resorption—because these neither a keel nor a protrusion of any are the only osteogenetic processes available kind. The large mental foramen lies very for skeletal modification—neither Qafzeh 7 posteriorly, under the bony septum that nor the adult Skhu¯l specimens possess the separates the roots of adjacent deciduous details of symphyseal morphology seen in H. molars. sapiens. Since we know that the symphyseal Aside from the impressions created by the morphology of adult H. sapiens is even more roots of the teeth, the symphyseal region of clearly delineated in young individuals, the Skhu¯l I child’s mandible is devoid of similar details in children from Qafzeh and significant morphology. Like the Qafzeh 4 Skhu¯l would be required to demonstrate child, it thus lacks in comparable parts the that the symphyseal bulge of Qafzeh 7 and distinctive and easily identifiable morpho- the Skhu¯l adults are versions of the chin in logical details of juvenile H. sapiens. This modern humans. As discussed above, how- observation is particularly revealing because ever, the juvenile Qafzeh 4 specimen does the diagnostic features of the symphyseal not display the features of juvenile H. region of H. sapiens are more pronounced in sapiens. This also appears to be the case with younger than in older individuals. Conse- the Skhu¯l I child. quently, the absence of these features in the Although the specimen is heavily recon- Skhu¯l I child certainly provides evidence structed in some areas, bone is preserved on of significant developmental differences the right side of the Skhu¯l I child’s mandible between this individual and extant H. sapiens at least as far as the midline of the root of the of similar age, as has also been demonstrated right first deciduous incisor, as well as inter- to be the case with regard to dental growth nally across the symphysis at least as far as rates in the Skhu¯l I child vs. H. sapiens the region of the left di2 (Figure 8). This (Smith et al., 1997). makes it possible to determine accurately If the Skhu¯l I child had become a typical the morphology, thickness, and curvature of Skhu¯l adult, then the developmental differ- the symphyseal region. In contrast to juven- ences between the entire Skhu¯l assemblage ile H. sapiens, the Skhu¯l I mandible is broad (with which, on morphological grounds, we and somewhat arcuate from side to side would include the Qafzeh 7 adult) and H. across the region of the anterior teeth, and, sapiens would have been evident from child- in profile, this region is slightly convex. In hood. However, one cannot accept without further contrast to extant H. sapiens of simi- question the biological affinity of the Skhu¯lI lar age, but as in Neanderthals (see discus- child with the adults from the same site. sion below and Zollikofer et al., 1995), the Indeed, in its general shape, disposition of anterior region of the Skhu¯l I mandible is cusps, and presence and orientation of the 382 . .   .    ‘‘’’ 383

‘‘deflecting wrinkle’’ (elongate centroconid), specimens of H. sapiens, as also appears to the child’sM1 is a good match for the M1 of be the case with regard to all Skhu¯l adults Tabu¯n C1 while, unfortunately, such details and Qafzeh 7. of morphology of adult Skhu¯lM1’s are not preserved. Thus we cannot be certain of the Neanderthals juvenile state from which adult Skhu¯l man- Although variable in its verticality, the dibular morphology emerged. Nonetheless, anterior surface of the adult Neanderthal given that the distinctive features of the mandible is very broad and, if not flat, symphyseal region in H. sapiens, so salient in is often slightly arced from side to side younger individuals, diminish in clarity with (Figure 9; see also e.g., Wolpoff, 1980:285). growth and bone remodeling, we feel that In some specimens, this broad surface may we can at least hypothesize that, in their be adorned inferiorly by a pair of peaked or juvenile mandibles, the Skhu¯l (and Qafzeh angular distentions of the inferior margin of 7) adults would not have been configured as the mandible (identified here as ‘‘inferior in H. sapiens. Rather, we suggest that, as in marginal tubercles’’). When present, the various Neanderthal specimens (see below), inferior marginal tubercle lies below the the bulge observed in the Skhu¯l adult and mental foramen, which, in turn, lies quite

Qafzeh 7 specimens was created through a posteriorly, beneath either M1 or the bony process of bone deposition upon a morpho- septum separating the mesial alveolus for logically featureless symphyseal region, this tooth from the alveolus for P2 (see while the subalveolar depression would have Krapina 59 and Spy 1 in Figure 9; see also been produced by bone resorption in the Wolpoff, 1980:285). In some specimens, the region of the incisor root tips. However, inferior margin of the anterior mandibular even if one were to claim that the symphy- plane is elevated. There may also be a shal- seal bulge could have been derived from low subalveolar depression coursing below, the configuration seen in the fetuses and if not also across, the entire region of the juveniles of H. sapiens (i.e., the inverted ‘‘T’’ outwardly bowed incisor roots; the more and attendant mental fossae), one would procumbent the incisor roots, the more they have to admit that the subsequent and overhang and emphasize the subalveolar total obliteration of this configuration was depression. In some specimens, the region achieved by a process of development that below the subalveolar depression may be differed significantly from the extant slightly and broadly swollen, providing a hominid in terms of the regions and inter- smooth curve from side to side in frontal actions of the domains of bone resorption view and a gently convex profile (e.g., see and deposition. Spy1inFigure 4; see also Grotta Guattari 2 On the basis of morphological dissimilar- in Wolpoff, 1980:285). Clearly, although ity, then, we must conclude that Qafzeh 4 there is variability within the Neanderthal and Skhu¯l I are, morphologically, not sample in the presentation of the features

Figure 6. Mandibles of adult specimens from Qafzeh with the symphyseal configuration of H. sapiens. Top: Qafzeh 11 [Tel Aviv University (TAU)]; middle: Qafzeh 9 (TAU); bottom: Qafzeh 8 (TAU). Qafzeh 11 possesses a thin central keel that is bound on each side by a mental fossa and which arises from below the alveolar region and expands into a well developed, inferiorly placed mental protuberance that expands somewhat laterally into a thickened inferior margin. Qafzeh 9 preserves a thickened inferior margin, above which the bone becomes everted toward the midline, suggesting that a central keel had been present and bounded by mental fossae. The more fragmentary Qafzeh 8 displays a thickened inferior margin. Bar=1 cm. 384  .  .    . 

Figure 7. Mandibles of adult specimens from Skhu¯l and Qafzeh with similar symphyseal configurations. Top row left, Skhu¯l V [Harvard Peabody Museum (HPM)]; top row right, Skhu¯l IV (TAU); bottom row left, Skhu¯l II (HPM); bottom row left, Qafzeh 7 [Institut de Paleontologie Humaine (IPH)]. Although differing from one another in degrees of expressions of features, all mandibles are anteriorly broad and display a subalveolar depression below. Skhu¯l IV and V and Qafzeh 7 clearly possess a subalveolar bulge that is protrusive above the inferior margin and which lacks a central keel and attendant mental fossae. Skhu¯l II lacks mental fossae and an inferiorly thickened margin. Bar=1 cm.   ‘‘’’ 385

Figure 8. Mandible of Skhu¯l I (TAU); top (anterior view), middle (left lateral view), and bottom (inferior view). The reconstruction of the front of the jaw is based on the right side being preserved to the midline: it is very broad and somewhat arcuate from the side. In contrast to juvenile H. sapiens, Skhu¯l I lacks a central keel and mental fossae, and is thin anteroposteriorly in the symphyseal region. Bar=1 cm. 386 . .   . 

Figure 9. Mandibles of adult Neanderthals to illustrate features in common and differences between individuals. Top, left (anterior view) and right (left lateral view), Tabu¯n C1 [Natural History Museum, London (NHML)]; middle, left (anterior view) and right (left lateral view), Krapina 59 [Hravatsku Prirodoslovni Muzej, Zagreb (GPM)]; bottom, left (anterior view) and right (left lateral view) of Spy 1 (Institute Royal des Sciences Naturelle de Belgique, Brussels). In all, the front of the mandible is very broad and somewhat arcuate from side to side; also, all clearly display a retromolar space between the last molar and the ramus. Individual variation is noted in differences in depth of the mandible, development or lack of inferior marginal tubercles (noted beneath the very posteriorly placed mental foramina of Krapina 59 and Spy 1, but lacking in Tabu¯n C1), development or lack of subalveolar depression along the incisor roots (tall and shallow in Krapina 59, shorter in Spy 1, absent in Tabu¯n C1), presence or absence of slight subalveolar swelling (only somewhat pronounced in Spy 1), and elevation of the inferior margin at the front of the jaw (noted in Krapina 59 and Spy 1, but not Tabu¯n C1). Bar=1 cm. described above, no adult Neanderthal fossae that are otherwise seen in H. sapiens. specimen possesses, even vaguely or subtly, In addition, the Neanderthal symphyseal the inverted ‘‘T’’ and attendant mental prominence, when present, does not have   ‘‘’’ 387 the same configuration as the symphyseal noticeably relatively thin from front to back, bulge seen in the adult Skhu¯l specimens and with the corpus being thickest in the region Qafzeh 7. of the first deciduous molar (see also Study of the mandibles of juveniles Zollikofer et al., 1995). As noted in the identified as Neanderthal on the basis of Amud child and in the occasional various cranial features demonstrates that Neanderthal adult, there may be a discrete, none in our sample displays the inverted peaked inferior marginal tubercle lying ‘‘T’’ and mental fossae of H. sapiens. For below the posteriorly placed mental foramen instance, the symphyseal region of the (see Gibraltar 2 in Figure 11). mandible of the 10-month-old child from Among the Neanderthal children, the Amud (Rak et al., 1994)(Figure 10)is Gibraltar 2 (Devil’s Tower) mandible is of essentially smooth from side to side across further interest in that a shallow midline the midline; there is no trace of a long, cleft indents the inferior border (Figure 11). raised central keel, distended inferior mar- This indentation lies at the mandibular sym- gin, or mental fossae. Surface topography is physis, which is where fusion of the right confined to the swelling of the region above and left mandibular halves occurs. In all the very laterally and posteriorly placed other details, this specimen is recognizably mental foramina due to the expanding similar to the other Neanderthal children. If crypts of developing tooth germs. The we accept that early symphyseal fusion, mandible is very broad and gently arced which is synapomorphic of Anthropoidea frontally from side to side, and, in profile, is (e.g., see review by Schwartz, 1986), slightly arced from top to bottom. When occurred in Neanderthals as it does in H. viewed from above, and in contrast to sapiens—from the top down—then the juvenile H. sapiens, the broad (from side to indentation in the midline of the inferior side) front of the Amud mandible is thinner margin of the Gibraltar 2 mandible is simply anteroposteriorly than the rest of the corpus a vestige of incomplete symphyseal fusion, is more posteriorly. Below the large mental as is obviously the case in the Isturitz foramen on the right side, the inferior specimens (Figure 5). This, of course, is the margin of the mandible bears a long, most parsimonious explanation of this blunted inferior marginal tubercle which is in the Gibraltar 2 child. But even if bounded on either side by a slight upward this is not a completely accurate interpret- depression. ation, it would be incorrect to identify this As would be expected from a growth specimen’s inferior marginal indentation as trajectory that is predicted upon the con- constituting a ‘‘hint’’ of a ‘‘chin.’’ figuration seen in the Amud child in con- The morphology of the Archi child’s trast to juvenile H. sapiens,3–4-year-old mandible, which is generally assumed to Neanderthals (Roc de Marsal, Gibraltar 2, represent a juvenile Neanderthal (Ascenzi Pech de l’Aze´) also lack the inverted ‘‘T’’ & Sergi, 1971a,b; Stringer et al., 1984; and mental fossae seen in extant humans Mallegni & Trinkaus, 1997) (see Figure 12), (Figure 11; see also Tillier, 1983). In frontal can be better understood in light of the view, the mandibles of these children are Gibraltar 2 specimen. In terms of relative broad and somewhat arcuate from side to chronological age, the Archi individual, in side. In profile, the symphysis may be which only the deciduous teeth had erupted, smoothly vertical or slightly convex and would have been younger than the Gibraltar retreating. In contrast to extant H. sapiens of 2 child but comparable to the Roc de Marsal similar age, the Neanderthal symphyseal and Peche de l’Aze´ children. In overall man- region, when viewed from below, is also dibular shape, the uniform anteroposterior 388 . .   . 

Figure 10. Amud Neanderthal child. The mandible is broad and relatively featureless across the anteroposteriorly thin symphyseal region, which is adorned below the posteriorly placed mental foramina by distinct inferior marginal tubercles. This specimen (TAU) lacks a central keel, mental fossae, and thickened inferior margin. Bar=1 cm.   ‘‘’’ 389

Figure 11. Mandibles of 3–4-year-old Neanderthals. Left column, Gibraltar 2 (Devil’s Tower) (NHML); top (anterior view), middle (right lateral view), bottom (inferior view). Right column, Roc de Marsal (Muse´e National de Prehistoire, Les Eyzies); top (anterior view), middle (right lateral view), bottom (inferior view). As in adults, the front of the mandible is broad and somewhat arcuate from side to side. With the exception of a slight upward indentation at the midline in the Gibraltar 2 child (which probably represents incomplete symphyseal fusion), both mandibles are essentially featureless across the antero- posteriorly relatively thin symphyseal region. The Gibraltar 2 specimen bears well developed inferior marginal tubercles below the extremely posteriorly placed mental foramina. Bar=1 cm.

thickness of bone across the broad symphy- the dm2. Anterior tooth crowns in the seal region, and the thickening antero- process of growing within their crypts swell posteriorly of the corpus posteriorly, the out the mandible externally, creating a Archi child is quite similar to the Gibraltar depression just superior to the inferior 2 child, and thus to other Neanderthal margin of the bone. In profile, the relatively children as well (Ascenzi & Sergi, 1971a). A straight midline of the jaw is tilted somewhat large inferior marginal tubercle lies beneath forward. The central incisor roots notice- each large mental foramen, hence beneath ably overhang the symphyseal region. The 390 . .   . 

Figure 12. Archi 1 Neanderthal child. As in the mandibles of other juvenile Neanderthals, this specimen (Istituto Italiano di Paleontologia Umana) is broad across the symphyseal region, which lacks a central keel, mental fossae, and thickened inferior margin (for view of the anteroposteriorly thin symphyseal region see Mallengi & Trinkaus, 1997). The slight indentation of the inferior margin at the midline probably reflects incomplete fusion of the symphysis. Anterior to each mental foramen and extending to the thinned inferior margin is a longitudinal depression, whose posteroinferior extent is bounded by a distinct inferior marginal tubercle. Bar=1 cm. inferior margin is slightly elevated anterior Sergi (1971b) thought that ‘‘the slightly to the inferior marginal tubercles. inward curve of the outline which corre- Although clearly lacking the general sponds to the alveolar part suggests an initial mandibular and specific symphyseal con- incurvatio mandibulae anterior’’ and that figurations of juvenile H. sapiens, the Archi ‘‘[i]n the basal margin a low triangular specimen has been described as having elevation in the median line, the trigonum ‘‘some chin development’’ (Ascenzi & Sergi, mentale, is visible.’’ And Mallegni & 1971a,b; Stringer et al., 1984:79; Mallegni Trinkaus (1997:657) even ventured to state & Trinkaus, 1997). In particular, Ascenzi & that while ‘‘the discrete traits that contribute   ‘‘’’ 391 to the development of a mental trigone, and its attendant mental fossae. The only the tuber symphyseos, the fossa mentales, features that could possibly be invoked as the tubercula lateralia and the incurvatio demonstrating similarity between these two mandibulae, are clearly present on most hominids—the occasional subalveolar swell- immature Neandertal mandibles . . . Archi 1 ing and the equally occasional inferior exhibits generally weak development of marginal tubercle—are just as clearly not them.’’ As is evident from the foregoing comparable either in detail or overall mor- description of immature human and phological context. But, exactly as adult Neanderthal mandibles, however, modern H. sapiens retain the fetally established mor- humans and Neanderthals are not at all phology of the symphyseal region, so, too, similar in symphyseal morphology or general do adult Neanderthals retain features mandibular proportions. And the same seen in their juveniles. In particular, adult holds true for modern children and the Neanderthals retain the broad, somewhat contrasting Archi child, although the latter arcuate shape of the symphyseal region, nonetheless does display an interesting which remains anteroposteriorly uniformly developmental feature in this region. thick. Furthermore, just as the differences Compared to the same region in the among adult H. sapiens, in say, the develop- Gibraltar 2 child (Figure 11), the ‘‘chin’’ of ment of the inverted ‘‘T’’ and mental fossae the Archi child becomes recognizable as reflect individual differences within the what it is: a slight indentation at the midline species, so, too, do differences between of the inferior margin of the mandible. In adult Neanderthals reflect variations contrast to the Gibraltar 2 child, the Archi between individuals of this particular child’s incompletely formed right and left hominid. Such differences occur in the permanent tooth crowns, still in their crypts, degrees of depth from top to bottom, flatness markedly swell out the bone on either side of vs. arcuateness from side to side, and/or the midline beyond the plane of the sur- verticality vs. inclination of the symphyseal rounding external surface of bone. This region. Individual differences between creates a noticeable depression beneath, Neanderthals are also found in the presence as well as another, but more shallowly or absence of inferior marginal tubercles, depressed, area medial to, each swollen subalveolar depressions, low, broad swell- region of bone. The latter depressions ings below these depressions, and/or in the delineate between them a narrow midline elevation of the inferior margin of the front field of bone, at the base of which lies the of the jaw. In fact, a qualitative assessment inferior marginal cleft. The hint of a chin in of symphyseal morphology indicates that the Archi mandible, therefore, results not Neanderthals are probably more frequently from any of the morphologically consistent and noticeably variable in morphological features in H. sapiens that form the inverted embellishment of this region than extant ‘‘T’’ and attendant mental fossae, but from H. sapiens. the combined effects of incomplete coales- As evolutionary kin, Neanderthals and cence of the inferiormost extremity of the H. sapiens—even if not very closely related— and the intrusiveness would be expected to share many, if not of tooth crown formation. most, aspects of their skeletal morphology. Neanderthals obviously differ from extant These would, of course, have been retained H. sapiens in the development and specific from a common ancestry that they shared morphology of the symphyseal region of the to the exclusion of other hominids. The mandible. In no Neanderthal specimen seen additional insight provided by the juveniles by us are there any signs of an inverted ‘‘T’’ in the Neanderthal sample contributes 392 . .   .  significantly to our ability to determine chin in a meaningful morphological and whether apparent similarities shared by phylogenetic sense. adults of these hominids and H. sapiens are, indeed, present by virtue of common The Case of Tabu¯n II ancestry. When juveniles, especially of close The preserved symphyseal region of the related taxa, are distinctly different, as Tabu¯n II mandible has been said to display juvenile Neanderthals and H. sapiens are in some chin development (McCown & Keith, details of symphyseal morphology, it 1939; Quam & Smith, 1996; Rak, 1998). It becomes even more evident that what might thus requires separate discussion. in some adult specimens appear superficially The Tabu¯n II mandible (Figure 13)is as homologous features (such as a sub- anteriorly broad and deep. In even more alveolar depression or bony build-up in the exaggerated form than in the mandibles of symphyseal region) are actually homo- some Neanderthal adults (e.g., Spy 1 and plasies, having arisen through independent Regourdou), this specimen displays a tall, courses of development. broad, and shallow subalveolar depression This realization is true even if the cellular that is overhung considerably by the out- processes—resorption vs. deposition—that wardly curved roots of the permanent inci- created the features of a particular speci- sors. Although there has been some damage men are similarly operative among all indi- to the region around the left central incisor viduals of all taxa. For, after all, as the only root (which is filled in with plaster), bone is active phases of osteogenesis, how else if preserved to the midline on the right side, not through an interactive process of bone where it clearly lacks any sign of a vertical deposition and resorption can skeletal keel. The only elevation of bone in the features achieve their adult size and shape? symphyseal area is a small smooth fragment Thus, although it is commonplace when (delineated by its edges) that has become considering the evolution of later hominids displaced from the plane of the bone’s sur- to arrange adult specimens along the lines . In addition, there are no signs of any of a scale naturae asareflection of changes structures that might be identified as mental that might have occurred in a transfor- fossae anywhere within the expected areas of mation sequence leading to the morphol- the otherwise well preserved symphyseal ogy of H. sapiens (Gregory, 1922; McCown region. The inferior border of the most & Keith, 1939; Wolpoff, 1980, 1996; central portion of the symphyseal region is Bra¨uer, 1984; Smith, 1984; Lieberman, missing and has been reconstructed with 1995), we suggest that this is not a produc- plaster. However, more posteriorly, the tive way in which to address the situation. inferior border of the jaw is present on each Actually, it is rather Haeckelian. Instead, side, extending to some extent medially following a von Baerian approach, we from a blunt inferior marginal tubercle. In advocate that, especially when they are these regions, the inferior border is not available for study, juveniles should be everted. The inferior marginal tubercles lie compared in order to elucidate develop- beneath the large mental foramina, which mental commonalities. For it is only at this are situated below the bony junctions ff level of comparison that the di erences and between P2 and M1. similarities among the adults can be fully Clearly, had this specimen possessed the understood. In light of this, it is patently inverted ‘‘T’’ and mental fossae that charac- obvious from comparisons between juvenile terize H. sapiens, there is sufficient bone H. sapiens and Neanderthals that only one preserved to demonstrate their presence. In of these species of hominid possesses a their absence, therefore, we cannot ascribe   ‘‘’’ 393

Figure 13. Mandible of Tabu¯n II (TAU). As is typical of Neanderthals, the front of the jaw is very broad and somewhat arcuate from side to side. As in some Neanderthals, the symphyseal region is very deep, bears a subalveolar depression with a slight bulge below, and is elevated slightly along its inferior margin; there are also inferior marginal tubercles beneath the very posteriorly placed mental foramina. Bar=1 cm. 394 . .   .  this specimen to this species. Further- Other Middle to Late Pleistocene Western Old more, since the Tabu¯n II mandible lacks the World Fossils ‘‘tear-drop’’ shaped symphyseal bulge seen Mauer. As is well known, the Mauer jaw in the adults from Skhu¯l, we cannot include lacks any morphology that could be this specimen with that sample, either. In its described as ‘‘chinlike’’ (e.g., Condemi & bulk, depth, and in various morphological von Koenigswald, 1997). The symphyseal features (e.g., subalveolar depression, region is broad and arcs strongly from side inferior marginal tubercles), however, this to side (Figure 14). In profile, it curves specimen is more strongly reminiscent of the gently inferoposteriorly. Anterior to the mandibles of some western European thickest parts of the mandible, which lie Neanderthals—particularly those from Spy below the bilaterally doubled mental and Regourdou—than it is of the Tabu¯nC1 foramina (which, in turn, are beneath the mandible (cf. McCown & Keith, 1939; P2–M1 alveolar septum), the inferior margin Quam & Smith, 1996; Rak, 1998) (see is elevated well above the plane of the Figure 9). Rak (1998), however, has argued inferior border of the posterior corpus. The that Tabu¯n II should be regarded as H. margin of this elevation can be described as sapiens because, like the latter, it differs from shallowly bow-shaped: prior to reaching the Neanderthals in the depth of the mandibular midline, the bone on each side curves down notch crest, the position of greatest depth of to form a low inverted peak. From the this notch, and the point of termination of subforaminal points of greatest thickening, this crest on the mandibular condyle. While the inferior border thins anteromedially such this is true of the left side, the Tabu¯nII that the broad central portion of the sym- mandible displays the Neanderthal con- physeal region is devoid of embellishment, figurations on the opposite ramus. Rak while posteriorly it has thinned maximally

(1998) also suggests that the Tabu¯nII by the level of M3. When viewed from mandible would not display a marked retro- below, the broad symphyseal region is more molar space, as in Neanderthals, if it did not or less uniformly thick anteroposteriorly, have a pre-angular notch. But since such a though thickening at its lateral most notch becomes defined when the coronoid extremities. process of the ramus is expanded anteriorly, This specimen is obviously distinctly dif- it is not clear that one can therefore ‘‘fill in’’ ferent in its symphyseal region from H. the notch and eliminate the retromolar sapiens and the adults from Skhu¯l (plus space. In our experience, one can find Qafzeh 7). One might claim similarity mandibles of H. sapiens that also display a between extant humans and the Mauer jaw pre-angular notch, but they do not also have in the development of inferior marginal a retromolar space. thickening, but, in detail, the configurations While it is clear to us that the Tabu¯nII differ strongly. In H. sapiens (see above), mandible is not H. sapiens, it is less easily inferior marginal eversion and thickening demonstrable that it is a Neanderthal. occurs anterior to the mental foramina and Nevertheless, this seems highly probable are thus restricted to an area that is well on the basis of the specimen’s broad sym- within the symphyseal region. In contrast, physeal surface, inferior marginal tubercles, inferior marginal thickening is most promi- posteriorly placed large mandibular nent in the Mauer jaw at and posterior to the foramina, slightly procumbent anterior mental foramina, while it is lacking anterior tooth roots, and large medial ptyerygoid to these foramina. Although similar to some tubercle (Schwartz & Tattersall, in prep- adult Neanderthal specimens in display- aration). ing elevation of the inferior border of the   ‘‘’’ 395

Figure 14. Mauer Mandible. Although damaged obiquely along the symphysis, it is still obvious that this specimen (Geologish-Pala¨ontologisches Institut, Ruprecht-Karls-Universita¨t Heidelberg) lacked any morphological embellishment in this region. Inferior marginal thickening, which is highlighted by a horizontal sulcus above, is noted well posteriorly, beginning in the vicinity of the large inferior marginal tubercle that lies beneath the mental foramina. symphyseal region, the Mauer jaw is clearly external embellishment in the Mauer adult, distinguished from Neanderthals in general we would predict that the symphyseal not only in the contour of the inferior region of Mauer juveniles would have been margin, but also in the development of similarly unadorned. inferior marginal thickening, not to mention other details of mandibular morphology not Arago. Arago 2 preserves a laterally broad described here (e.g., Condemi & von and broadly arced symphyseal region Koenigswald, 1997). Given the lack of (Figure 15). In profile, the anterior alveoli 396  .  .    . 

Figure 15. Mandibles of Arago 2 (left column) and Arago 13 (right column) (Universite´ de la Me´diterrane´e). In Arago 2 the mandible is broad and essentially featureless surficially across the front of the inferior margin, which is elevated; the incisor roots had been angled forward, creating a subalveolar depression; an inferior marginal tubercle lies below the posteriorly placed mental foramen. Enough of the symphyseal region of Arago 13 is preserved to demonstrate that it, too, lacked a central keel, mental protuberance, thickened inferior margin, and mental fossa. Bar=1 cm.   ‘‘’’ 397 are inclined forward and thus overhang the because, while Smith (1984) is inclined bone below, which curves smoothly down to see the Ehringsdorf material as and back into the inferior margin. The Neanderthal-like and Vlce˘k (1993) views it inferior margin rises anteriorly, beginning its as H. sapiens-like, our preliminary assess- ascent in front of the first molar on the left ment of the material suggests a different side and the second on the right. Thus the interpretation (Schwartz & Tattersall, in symphysis is short superoinferiorly. On both preparation). The adult mandible Eh sides the large mental foramen lies level with 1009/69 is relatively narrow across the sym- the M1. Although bone missing from the physeal region, which is slightly curved from midline deprives us of information concern- side to side as well as from top to bottom. ing the existence of a keel, the lack of both Between a pair of low but distinct inferior an everted inferior margin and mental fossae marginal tubercles that lie just posterior to bilaterally demonstrates the absence of two the large mental foramina (which, in turn, of the major aspects of the symphyseal are situated beneath M1), the inferior border region in H. sapiens. In addition, the bone of the front of the jaw is elevated relative of the symphyseal region and well along to the posterior corpus. The roots of the the corpus is more or less uniformly thick anterior teeth are inclined forward (those of buccolingually. the central incisors being especially procum- Only a small portion of the symphyseal bent), so that the alveolar margin markedly region to the left of the midline is preserved overhangs the symphyseal region. Bone in Arago 13. Neither a keel nor a mental remodeling due to subperiosteal as well as to fossa is present. In profile, the alveolar periapical lesions has created deep craters region is smoothly confluent with the fairly on either side of the retained left central vertical symphyseal region, which curves incisor. These greatly accentuate the incli- gently back below into the inferior border. nation of the anterior teeth, to which this On the preserved right side, below and pathology may also have contributed. By posterior to the fully exposed canine default, alveolar bone eversion in conjunc- alveolus (which may have been opened as a tion with these massive concavities (particu- result of abscessing), the inferior margin larly the larger of the two on the right side) is thickened into a torus that extends gives the false impression of a faintly posteriorly to reach the level of the front of mounded symphyseal surface near the the second molar. The large mental foramen inferior margin. lies below the anterior root of M1. The In contrast to Eh 1009/69, the anterior inferior border of the corpus is not elevated teeth in the juvenile mandible (Eh 1010/69) anteriorly. When viewed from below, what are relatively orthal, but the entire alveolar there is of the external symphyseal surface is region is shifted slightly anterior to the plane relatively flat across and the bone of the of the symphysis below. Although broken region is uniformly thick anteroposteriorly. and glued inferiorly, and reconstructed The corpus does not begin to thin until below and posterior to the region of the much more posteriorly. left canine, the symphysis is relatively broad Clearly, the Arago mandibles present and somewhat arced from side to side. a morphology different from those so far In profile, it is fairly vertical below the discussed. anteriorly displaced alveolar region as far as its inferiormost extremity, where it curves Weimar-Ehringsdorf. We discuss the two gently posteriorly into the somewhat Ehringsdorf mandibles (1009/69 and 1010/ elevated anterior margin. The expanse of 69; Figure 16)inthe‘‘other’’ category symphyseal region that is perfectly preserved 398  .  .    . 

Figure 16. Mandibles of Ehringsdorf (Eh) 1009/69 (left column) and Eh 1010/69 (Museum fu¨r Vor- und Fru¨hgeschichte Thu¨ingsisches, Weimar). Neither specimen displays a central keel, mental fossa, mental protuberance, or thickened inferior margin, and the symphyseal region in each is broad across. In Eh 1009/69 the central incisors greatly overhang the bone below and the inferior marginal tubercle lies level with the posteriorly placed mental foramen. Bar=1 cm.   ‘‘’’ 399

Figure 17. Rabat (Thomas Quarry) mandible (IPH). The incisors overhang somewhat the essentially featureless symphyseal region. between the canine roots/alveoli and on curved but not broad from side to side. On down to the inferior border is devoid of the intact right side is a shallow subalveolar surface morphology. depression, inferior to which the bone is featureless, bearing neither a central keel nor Rabat. The Rabat partial mandible preserves a mental fossa more laterally. In profile, the the entire symphyseal region, which was superoinferiorly not very tall symphysis broken postmortem below the roots of the curves slightly down and back into the left anterior teeth, and has been repaired unembellished inferior border. More clearly (Figure 17). Despite these attentions, it is seen on the right side, a short longitudinal obvious that the symphyseal region is well groove separates the mental foramen (which 400 . .   .  is level with the P2) from the inferior border, behind is uniformly thick anteroposteriorly, which, in consequence, appears to be thick- becoming thickest in the region of M2. ened in that region. When viewed from The symphyseal region of the most com- below, the bone is relatively uniformly thick pletely preserved mandible, Tighenif 3, is anteroposteriorly throughout the symphy- also moderately broad and strongly arced seal region as well as more posteriorly along from side to side. In profile, the alveolar the corpora. region of the anterior teeth is tilted forward Although taxonomic allocation of this and then up and lingually very slightly, and specimen is uncertain, it clearly is not H. the minimally arced symphysis courses back sapiens or a variation on the European or and down to the elevated inferior margin, Levantine hominids discussed above. which begins its ascent from the region below the first molar. There is the faintest Tighenif (Ternifine). Even the least complete hint of a broad, low, midline keel, which of the three Tighenif mandibles, Tighenif 2, widens as it courses down from the alveolar preserves a good portion of the front of the region to become confluent with the slightly jaw (Figure 18). The symphyseal region of thickened inferior border. As is well pre- Tighenif 2 is narrow and transversely arced. served on the left side, there is a distinct This specimen is particularly interesting in horizontal sulcus that courses beneath the that it bears a distinct—broad and low— region of the teeth to the region of midline keel that emerges just below the the midline keel that serves to emphasize alveolar region and fans out as it courses further the thickening of the inferior toward the inferior margin, thereby creating border. a well-defined, triangular protuberance. On Bra¨uer (1984) described a trigonum each side of this keel lies a small, shallow basale only in Tighenif 2 and 3, but clearly depression, which accentuates the lateral Tighenif 1 is similar to the other two extremities of this protuberance. As pre- specimens in displaying a midline keel served on the left side, there is a distinct that courses to the inferior border of the inferior marginal tubercle, which lies under mandible. Indeed, the keel is actually more the large mental foramen (which, in turn, pronounced in Tighenif 1 than in the other is level with the mesial side of P2). This two specimens. It is of potential phylo- tubercle is separate from, hence not a genetic and systematic importance that all terminus or ‘‘corner’’ of, the triangular three Tighenif specimens are unique among protuberance. In profile, the symphysis is the middle Pleistocene fossils we studied in straight and slopes back quite markedly. displaying some of the features distinctive The symphyseal region of Tighenif 1, of the symphyseal morphology of extant which retains both corpora, is moderately H. sapiens. It is of further interest that one of broad and strongly arced from side to side. these features—the central keel—is observed In profile, it forms a smooth arc, from the in fossil that do not also have the slightly inwardly inclined anterior tooth supposed hallmark of the human chin, a roots to the posteriorly curved and modestly clear-cut mental protuberance. This should elevated inferior border. There is a low but not be surprising when one considers that rather broad central keel coursing from the the mental protuberance of H. sapiens, alveolar region to the inferior margin. The which is itself a variable character, emerges presence of this central keel creates shallow postnatally, as does the occasional symphy- fossae on either side of the midline. When seal bulge in Neanderthals and presumably viewed from below, the bone of the symphy- other fossil hominids in which evidence of a seal region and the portion of the corpora central keel is lacking. Figure 18. Mandibles from Tighenif (Ternifine) (Muse´um National d’Histoire Naturelle, ). Top: Tighenif 2; middle: Tighenif 1; bottom: Tighenif 3. Although with some variation in degree of expression, all display a central keel, with slightly depressed mental fossae on either side, that broadens into an inferiorly placed mental protuberance; the inferior margin of the symphyseal region is somewhat elevated. 402 . .   . 

Implications serve a triangular mental protuberance, which both tapers superiorly into what From a phylogenetic perspective, if we might have been a central keel and expands accept on developmental grounds that H. inferiorly into bilateral mental tuberosities. sapiens is unique among primates (indeed, When viewed from below, the symphyseal among mammals) in the details of its region of this specimen is relatively straight symphyseal morphology, then it is possible across and somewhat thickened from front to entertain the following hypothesis: The to back. It will be interesting to see if the Tighenif specimens’ possession of a central Klasies River Mouth specimens present a keel and a variably small, low-set triangular picture of diversity as seen in the Qafzeh protuberance (a mental trigon or trigonum hominids. basale) could reflect potential synapo- Morphologically, the primary implication morphy with H. sapiens. Obviously, these of our observations is that a bulging sym- apomorphies would have characterized the physeal region is not equivalent to a chin. last common ancestor of a clade that When humans and elephants can both be included extent H. sapiens as well as those described as having chins (e.g., Enlow, late Pleistocene fossils possessing all of the 1982), it is probably time to reconsider the symphyseal details seen in the extant applicability of the term. Consequently, we species: i.e., at least the Qafzeh 11 and the recommend that the term ‘‘chin’’ no longer Tighenif mandibles. be used in discussions of mandibular mor- In this regard, we look forward to future phology and its phylogenetic and systematic study of the Jebel Irhoud juvenile mandible significance. If, however, one feels the need (Hublin & Tillier, 1981) and the Klasies to retain this term, it should be restricted in River Mouth mandibles (Rightmire & usage only to extant H. sapiens and those Deacon, 1991), all of which have been fossils displaying the constellation of sym- described as having some version of chin physeal features of this species. Since it is formation. As illustrated by Hublin & Tillier obvious that bulges and swellings as well as (1981), the former specimen may possess a subalveolar depressions of differing degrees central keel that expands inferiorly into a of expression and different morphologies thickened inferior margin. Casts of this have come to adorn the symphyseal regions specimen appear to show such morphology. of a diversity of hominids, we further recom- In contrast, the Klasies River Mouth mend that the terms (and the synonyms of) specimens, particularly KRM 13400 and ‘‘mentum osseum,’’ ‘‘incurvatio mandibu- 14695, do not appear to have either a central laris,’’ ‘‘tuberculum laterale,’’ and even keel or a thickened inferior margin, although ‘‘trigonum mentale’’ also be limited, or even the central portion of the symphyseal region dropped from usage. As anatomical terms, is surmounted by an external swelling they are used to refer both to sometimes (Rightmire & Deacon, 1991). A cast of superficially similar and to sometimes KRM 13400 appears to be generally consist- markedly different structures in a diversity ent with this description, although it should of adult specimens that may have arisen be noted that, well lateral to the midline, the from totally different juvenile configur- margin thickens to its greatest extent below ations. As for children, the sometime pres- the doubled right mental foramina, and the ence of a gomphotic scar especially low symphyseal profile is essentially vertical. down along the midline of the mandible is a The AMNH cast of KRM 41815, which is naturally occurring result of symphyseal broken in the original just below the incisor fusion. Unless it is distinctly raised and alveoli, does, as described, appear to pre- part of an everted, inverted ‘‘T’’-shaped   ‘‘’’ 403 structure that is associated with mental Intermediate, transitional morphological states fossae, it is not a ‘‘hint’’ or a ‘‘trace’’ of a We presented our descriptions and compari- mental trigon as seen in H. sapiens. Thus, sons of the specimens as we did in order to these terms have become charged with a keep them separate from our discussion of scenario of ‘‘chin’’ evolution which should the systematic implications of the data. itself be discarded. Often, however, there is a tendency to infuse description and comparison of specimens with interpretation of the potential signifi- cance of the described morphologies given a Additional discussion particular, but not overtly stated, phylo- Throughout this paper, we have attempted genetic scenario. Such scenarios tend to be to keep our description of morphology based on assumptions not only of related- apart from a phylogenetic and, especially, ness, but also of transformation series (in functional interpretation of it. We have, which terms such as ‘‘intermediate’’ are a however, stressed the importance of under- part) and implied functional and/or develop- standing the morphology of the adult by mental explanations (in which discussing starting first with an appreciation of the how morphologies might have come to be in details of juveniles precisely because this is one form or another plays a large role). the most fundamental way in which to Basically, however, the facts are that H. appreciate the similarities or differences that sapiens has a distinctive symphyseal con- exist among individuals and between the figuration from its ontogenetic outset and average morphologies of taxa. Comparing that many other hominids, especially Neanderthals with extant H. sapiens and Neanderthals, do not share any version of those specimens that have come to be these apomorphies. From this it would seem regarded as fossil H. sapiens, we are fortu- reasonable to conclude that Neanderthals nate that there are, indeed, juveniles avail- and those specimens that are morpho- able for study. In other cases, such as with logically similar to them in symphyseal the Mauer or Klasies River Mouth man- morphology are not members of the species dibles, we have tried to inform our com- H. sapiens. But if one was committed to the parisons with insights drawn from cases interpretation that there was only a single based on ontogenetic series. species, which encompassed a diversity Although we could leave the discussion of morphologically distinctive specimens— where we did in the preceding sections, such as all or most of those we discuss comments we received from various anony- here—then one would be obliged to explain mous reviewers on an earlier version of this how members of the same species could have paper have provoked us to pull together the come to be morphologically different from loose ends and to detail more fully our one another. We, however, are trying to theoretical, philosophical, and methodologi- understand as best we can what the details cal position. This might be unconventional, of difference and similarity are from the but it may prove useful in clarifying points beginning. Only afterward can we bring this we raised earlier and in focusing attention information to bear on our phylogenetic and on issues that those of different philosophi- systematic hypotheses, and, if we so choose, cal and methodological persuasions might to offer explanations for how and even also take into consideration. We have perhaps why certain features came to be the organized our comments in the context of way they are. the kinds of topic that often characterize Our appreciation of development also dif- debates on modern human origins. fers significantly from the more prevalent 404 . .   .  one in the literature: essentially, if develop- else’’), we must strongly reiterate that such ment is taken into consideration at all, it arguments merely constitute a defense of a is done secondarily, after specimens of particular hypothesis of relatedness that one adult individuals have been sorted into a has already embraced. They do not contrib- Haeckelian scale naturae that is regarded as ute to the generation of theories of related- reflecting a transformation sequence. This ness. More specifically, in terms of the above license is prevalent in , as example, facial , or its lack, has is indicated in the oft-published diagrams of no bearing on the matter at hand. Juvenile fossil adult crania arranged in a presumed Neanderthals are not overtly facially prog- phylogenetic sequence. In such a scenario, nathic relative to the adult state. The profile for example, the symphyseal morphology of of a juvenile Neanderthal’s mandibular sym- adult Skhu¯l and Qafzeh specimens are seen physis can be vertical or anteriorly inclined. as being intermediate between Neanderthals Yet, especially in cases of the former con- and modern H. sapiens, with this inter- figuration, the juvenile Neanderthal’s sym- mediate morphology being attained develop- physeal region is devoid of any specifically mentally from the nondescript symphyseal H. sapiens-like morphology. region seen in the Skhu¯l I child. In contrast In H. sapiens, juvenile mandibles are to this type of approach, we, like von Baer essentially vertical in profile and yet possess (1828) and Huxley (1863), emphasize the the characters already described above. If a commonalities between juveniles, since mental trigon later develops to become those form the substrate upon which adult noticeably protrusive, it does so as a result of morphology is predicated (see also Schartz deposition of bone in that region (Enlow, & Tattersall, 1996b). In this light, especially 1982). Further accentuation of such a pro- if the Skhu¯l I child represents the juvenile trusive mental trigon can be provided by the state of the Skhu¯l adults, it becomes strik- development of a depression or an incur- ingly clear that the symphyseal bulge of the vatio mandibularis above it, which derives latter is not just a variant of the chin of H. from the process of bone resorption (Enlow, sapiens. Even if this were not the case, it is 1982). Thus, the inward curve above and difficult to sustain a compelling comparison the protruding mental trigon below that we between adult morphologies if their juvenile see in the mandibular profile of H. sapiens states, as in Neanderthals and H. sapiens, are results from the modification of an essen- so clearly different. tially vertical symphyseal face, not, as the One might also argue, on the basis of typical Neanderthal-to-H. sapiens scala arranging specimens in a morphological naturae based on adult specimens would scala naturae of presumed evolutionary prompt one to speculate, from the retreat of change, that a more anterior position of the the face during the emergence of the latter , as seen in the facially prog- hominid. Furthermore, since the degree of nathic Neanderthals and some early Homo, facial prognathism characteristic of adult merely masks the development of H. sapiens’ Neanderthals develops from the more symphyseal morphologies in specimens of orthognathic condition of the juvenile, if these representatives of the genus. Conse- there was any basis for the ‘‘anterior alveolar quently, the symphyseal morphologies of position masking H. sapiens’ features’’ H. sapiens are not truly absent after all. hypothesis, we would expect to observe Although this kind of argumentation is com- these features in the juvenile Neanderthals. monplace in paleoanthropology (e.g., ‘‘it But we do not. looks this way because of such and such, It might also be appropriate here to otherwise it would be identical to something explain why we must discuss features as   ‘‘’’ 405 discrete entities, rather than as stages in a the mutant recessive would be converted to transformational continuum: The genetics the dominant state (Johnson et al., 1998). and properties of development make it With this in mind, the novel configuration increasingly clear that novel characters will of the symphyseal region of H. sapiens can be appear instantaneously and, therefore, appreciated: whatever regulatory changes define new species from the start (see occurred that affected bone deposition and reviews by Schwartz, 1999a,b). As hence shape in this region, these apparently Gru¨neberg (1943) demonstrated decades apomorphic morphologies would have ago, a structure (such as a tooth) must reach emerged without a trail of transitional stages a certain minimal threshold of development preceding them. A similar phenomenon or it will be aborted. The evolution of the would explain the development of the horse presents an unambiguous example autapomorphic medial projection in the (Schwartz, 1999c). In three-toed taxa, the nasal region of Neanderthals (Schwartz lateral digits always consist of three phalan- & Tattersall, 1996a).1 In 3–4-year-old geal elements, even if they are small relative juveniles, this structure is already present as to the size of the central digit. With the a vertically oriented bony projection that appearance of the first single-toed horses, emerges directly from the wall of the nasal the lateral digits are entirely absent. From cavity and that becomes larger in adults. the perspective of regulatory genes, and, No equivalent structure exists in the nasal therefore, of the developmental advent of region of juvenile extant primates, H. sapiens structures, the picture is also of the abrupt included, which retain the primitive appearance or disappearance of features mammalian condition of a low, variably (Schwartz, 1999a,b). For example, in the horizontally oriented conchal crest, along case of the vertebrate eye, studies on the Rx which a separate concha articulates. In gene in the mouse demonstrated that indi- adults of these taxa, the concha may fuse to viduals that were either homozygous or the wall of the nasal cavity. heterozygous for the active gene had both an eyeball and bony socket, whereas homo- Variation zygotes for the experimentally-mutated inactive or recessive Rx gene lacked both As noted in Table 1, the specimens of extant structures. There was no in-between state. H. sapiens we studied represent a global Since, as has been known since Bateson’s sampling. But, this fact aside, we must point (1909) time, nonlethal mutations typically out that we are not engaged in a study of arise in the recessive state, it would only be human variation. We are endeavoring to through a process of heterozygosis leading undertake a systematic analysis of hominids. ff to the production of homozygotes for the The latter concern entails a totally di erent mutation that a new feature would emerge emphasis than documenting variation within in a population (see discussion in Schwartz, 1Subsequent to our publication on the structures of 1999a,b). But when it did, it would appear the nasal cavity in Neanderthals, we restudied the Spy material and discovered that we had erred in our initial as if out of nowhere and in its full-blown evaluation of the Spy 1 specimen. Although we state in a number of individuals (Schwartz, were unable to publish a note of correction prior to 1999a,b). This process has most recently Franciscus’ (1999) article, we were able to eliminate this error from two other publications that were still in been documented in a report on the spon- press (Schwartz & Tattersall, 1999a; Schwartz et al., taneous appearance of a mutation affecting 1999). At present, a medial projection is docu- the Hodx-13 gene in a laboratory colony of mented in the following Neanderthal specimens: Engis, , La Chapelle-aux-Saints, , mice (Johnson et al., 1998). At some point, Monte Circeo 1, Roc de Marsal, St Ce´saire, Shanidar and by a mechanism that remains unknown, 1, and Subalyuk. 406 . .   .  a population or species. It involves the describe an arcuate eminence only in delineation of potential apomorphies, juvenile Neanderthals and one adult which, if autopomorphic, contribute to the Neanderthal specimen, Gibraltar 1. Clearly, delineation of species, or, if synapomorphic, if one chose to retain the term ‘‘arcuate to the delineation of sister taxa and clades. eminence’’ one would be better off referring In this particular study, we have delineated to the area in question as the ‘‘region of the what we believe to be various apomorphies arcuate eminence’’ (which technically would of the symphyseal region of H. sapiens.As be defined as the area capping the superior such, and for systematic purposes, it is semicircular canal), and then describing its therefore irrelevant whether one specimen configuration. With regard to the terminol- has a more robust symphyseal keel than ogy applied to the chin of H. sapiens,a another. What is relevant is how many mentum osseum may refer to a bulge that specimens possess the apomorphic features. straddles the midline of the jaw, but it is It is only after one accepts systematic obvious that a bulge can emerge upon totally hypotheses—how many species there are, different substrates: compare, for instance, and which specimens represent which the bulge of Spy 1, that of Skhu¯l V, and that species—that one can then embark, if one of Cro-Magnon 3. Clearly, while a bulge wishes, on studies of variation. But such may be produced by the same process of studies do not shed light on the origins and bone deposition (which is the only process relationships of the taxa involved, only on for increasing the size or thickness of a bony the ways in which individuals within the structure), this does not de facto mean same taxon differ slightly from one another. that all symphyseal bulges are the same In this light, it hardly matters if the inverted structure. Similarly, while an incurvatio ‘‘T’’ with its mental tubercles is more pro- mandibularis may be enhanced by bone nounced in an Aleut than in a West resorption in that region, as well as the African, or if the symphyseal profile of one development of a bulge below and alveolar Neanderthal is more forwardly sloping than prognathism above, this does not mean that another. But it is surely significant that no the details of the symphyseal regions of this Neanderthal specimen, no Skhu¯l speci- array of hominids are identical. men, and not all Qafzeh specimens display In short, the application of terminology the everted and inverted ‘‘T’’-shaped that had been devised for very specific fea- configuration of the symphyseal region. tures to a field of differently configured details that may be present in the same Terminology general region not only hinders productive During our now multiyear comparative comparisons but, in terms of human evol- study of hominids, we have become increas- ution, continues to obscure the differences ingly impressed by how inadequate and that exist between H. sapiens and other restrictive based on hominids. the morphology of H. sapiens often is when describing other hominids. For instance, Conclusions and summary while H. sapiens does indeed have an ‘‘arcuate eminence’’ that is large prenatally The uniquely configured symphyseal region and persists into the adult state, this con- of H. sapiens forms early in fetal develop- dition is rarely found in specimens of any ment and the basic shape and architecture other hominid, much less any other then established are retained into adulthood. mammal (see Schwartz & Tattersall, The salient features of the symphyseal 1996b). So far, we have been able to region of this extant hominid species are a   ‘‘’’ 407 raised central keel that flows into a dis- morphology, especially as considered within tended inferior margin, a low-lying triangu- a developmental framework, lends support lar mental tuberosity at the confluence of the to previous suggestions of notable taxic keel and the inferior margin, and mental diversity within the genus Homo (e.g., How- fossae that lie on either side of the keel and ell, 1994; Santa Luca, 1978; Schwartz above the distended inferior margin. No & Tattersall, 1996b; Tattersall, 1986; Neanderthal included in this study, whether Zollikofer et al., 1995): a diversity that is at juvenile or adult, possesses these distinctive present substantially underestimated. features. Instead, the mandibles of both juvenile and adult Neanderthals are broad Acknowledgements and somewhat arcuate across the midline. Those features that occur variably in We thank Drs C. Stringer (London), A. Neanderthal adults (e.g., a subalveolar Langaney, J.-J. Hublin, D. Grimaud-Herve´, depression, a swelling below this depression, S. Condemi, H. Thomas, and H. de Lumley inferior marginal tubercles that lie below the (Paris), M.-H. Thiault and S. Tymula very posteriorly placed mental foramina) (St.-Germain-en-Laye), R. Orban (Brussels), represent aspects of individual variation E. Poty (Lie`ge), R. Ziegler (Stuttgart), H.-E. appended during postnatal growth. Joachim (Bonn), N. Farsan (Heidelberg), Among fossils usually considered to A. Hoffman (Berlin), S. Dusek (Weimar), represent H. sapiens, those from the latest J.-J. Cleyet-Merle (Les Eyzies), B. Pleistocene and the earlier Qafzeh 11 (and Vandermeersch (Bordeaux), V. Merlin- possibly Qafzeh 8 and 9) display the sym- Anglade (Perigeux), S. Simone (Monaco), physeal features seen in living humans. In G. Spadea and A. de Lucchessi contrast, all of the Skhu¯l adults and the (Ventimiglia), A. Bieti (Rome), J. Radovcˇic´ other adult Qafzeh individuals differ both and M. Paunovic (Zagreb), V. Dobosi and I. from H. sapiens and from Neanderthals in Pap (Budapest), M. Dobis´ıkova´ (Prague), having symphyseal regions adorned with a M. Teschler-Nicola (Vienna), J. Jelinek variably developed teardrop-shaped bulge. (Brno), Y. Rak (Tel Aviv), J. Zias The Skhu¯l I child has a featureless symphy- (Jerusalem), J. Svoboda (Dolní Ve˘stonice), seal region. With the exception of fossil D. Pilbeam (Cambridge, MA), G. Kordos moderns, in the entire sample under con- (Thessaloniki), A. Akerraz (Rabat Souissi), sideration here it is only in the Tighenif Ben-Ncer Abdelwahed (Rabat Chellah), mandibles that one finds a central keel. and H. Langdon (Pittsburgh), and Direc- The latter observation suggests that the trice Joudia Hassar-Benslimane (Institut development of a central symphyseal keel is National des Sciences de l’Archeologie et du a synapomorphy of a clade that includes Patrimoine, Rabat) for permission to study extant H. sapiens and those fossils similarly and photograph specimens in their charge configured in details of the symphyseal that were essential to this study. The region, plus the Tighenif specimens. Within Spy specimens are in the collection of the this clade, H. sapiens and similarly config- Royal Belgian Institute of Natural Sciences, ured fossil specimens are distinguished by gift of the heirs of Professor Max Lohest the consistent possession of a thickened (1857–1926). All photographs are copy- inferior margin that forms an inverted ‘‘T’’ right  J. Schwartz. We thank Y. Rak, T. with the central keel, plus a variably protru- Harrison, R. D. Martin, and four anony- sive triangular mental protuberance at the mous reviewers for comments on aspects of junction of these structures and well defined this paper. J.H.S. was funded by the L. S. B. mental fossae bilaterally. Clearly symphyseal Leakey Foundation and the University of 408 . .   . 

Pittsburgh, and I.T. by the Ogden Mills of Human Evolution, pp. 167–185. London: Taylor & Fund (AMNH). Francis. Huxley, T. H. (1863). Man’s Place in Nature.New York: D. Appleton. Inke, V. G. (1967). Eigenschaften und Bedeutung der References Granzflache zwischen Basal und Alveolarteil des menschichen Unterkiefers. Nova Acta Leopold. 33, Aiello, L. & Dean, C. (1990). An Introduction to Human 161–195. Evolutionary Anatomy. New York: Academic Press. Johnson, K. R., Hope, O. S., Donahue, L. R., Ward- Ascenzi, A. & Sergi, A. (1971a). Il giacimento Bailey, P., Bronson, R. T. & Davisson, M. T. (1998). con mandibola neandertaliana di Archi (Reggio A new spontaneous mouse mutation of Hoxd13 Calabria). Accd. Naz. Lincei 50, 763–771. with a polyalanine expansion and simi- Ascenzi, A. & Sergi, A. (1971b). A new Neanderthal lar to human synpolydactyly. Hum. Mol. Gen. 7, child mandible from an Upper Pleistocene site in 1033–1038. southern Italy. Nature 233, 280–283. Johnston, T. B. & Willis, J. (Eds) (1954). Gray’s Baer, K. E. von (1828). U} ber Entwicklungsgeschicte Anatomy, 31st edn. London: Longmans, Green. der Thiere: Beobachtung und Reflexion.Ko¨nigsberg: Lam, Y. M., Pearson, O. M. & Smith, C. M. Borntra¨ger. (1996). Chin morphology and sexual dimorphism Bateson, W. (1909). Mendel’s Principles of Heredity. in the fossil hominid mandible sample from Cambridge: Cambridge University Press. Klasies River Mouth. Am. J. phys. Anthrop. 100, Bra¨uer, G. (1984). A craniological approach to the 545–577. origin of anatomically modern Homo sapiens in Lieberman, D. E. (1995). Testing hypotheses about Africa and implications for the appearance of modern from . Curr. Anthrop. Europeans. In (F. H. Smith & F. Spencer, Eds) The 36, 159–197. Origins of Modern Humans: A World Survey from the Mallegni, F. & Trinkaus, E. (1997). A reconsideration Fossil Evidence, pp. 327–410. New York: Alan R. of the Archi 1 Neanderthal mandible. J. hum. Evol. Liss. 33, 651–668. Condemi, S. & von Koenigswald, W. (1997). Der Mathers, P., Grinberg, A., Mahon, K. & Jamrich, M. Unterkeifer von Mauer. In (G. A. Wagner & K. W. (1997). The Rx homeobox gene is essential for Beinhauer, Eds) von Mauer: vertebrate eye development. Nature 387, 604–607. das Auftreten des Menschen in Europa, pp. 200–214. McCown, T. D. & Keith, A. (1939). The of Heidelberg, HVA: Akademischer Verlag. Mount Carmel: The Fossil Remains from the Levalloiso- Daegling, D. J. (1993). Functional morphology of the . Oxford: Clarendon Press. human chin. Evol. Anthrop. 1, 170–177. Dean, M. C., Stringer, C. B. & Bromage, T. B. (1986). Quam, R. M. & Smith, F. H. (1996). Reconsideration Age at death of the Neanderthal child from Devil’s of the Tabu¯nC2‘‘Neanderthal’’ (abstract). Am. J. Tower, Gibraltar and the implications for studies of phys. Anthrop. (Suppl.) 22, 192. general growth and development in Neanderthals. Rak, Y. (1998). Does any Mousterian present Am. J. phys. Anthrop. 70, 301–309. evidence of two hominid species? In (T. Akazawa, K. Enlow, D. H. (1982). Handbook of Facial Growth. Aoki & O. Bar-Yosef, Eds) Neanderthals and Modern Philadelphia: W. B. Saunders. Humans in West Asia, pp. 353–366. New York: Franciscus, R. (1999). Neandertal nasal structures and Plenum. upper respiratory tract ‘‘specialization’’. Proc. Natl. Rak, Y., Kimbel, W. H. & Hovers, E. (1994). A Acad. Sci. 96, 1805–1809. Neanderthal infant from Amud Cave, Israel. J. hum. Gregory, W. K. (1922). Origin and Evolution of the Evol. 26, 313–324. Human Dentition. Baltimore: Williams & Wilkins. Reisenfeld, A. (1969). The adaptive mandible: an Gru¨neberg, H. (1943). The Genetics of the Mouse experimental study. Acta Anat. 72, 246–262. (English translation). Cambridge: Cambridge Rightmire, G. P. & Deacon, H. J. (1991). Comparative University Press. studies of late Pleistocene human remains from Howell, F. C. (1994). A chronostratigraphic and taxo- Klasies River Mouth, South Africa. J. hum. Evol. 20, nomic framework of the origins of modern humans. 131–156. In (M. H. Nitecki & D. V. Nitecki, Eds) Origins of Robinson, L. (1913). The story of the chin. Knowledge Anatomically Modern Humans, pp. 253–319. New 36, 410–420. York: Plenum. Rosas, A. (1995). Seventeen new mandibular speci- Howells, W. W. (1959). Mankind in the Making. mens from the Atapuerca/Ibeas Middle Pleistocene Garden City. NY: Doubleday. hominids sample.. J. hum. Evol. 28, 533–559. Hrdlicˇka, A. (1911). Human dentition and teeth from Santa Luca, A. P. (1978). A re-examination of the evolutionary and racial standpoint. Dom. Dental presumed Neandertal-like fossils. J. hum. Evol. 7, J. 1911, 1–15. 619–636. Hublin, J.-J. & Tillier, A.-M. (1981). The Mousterian Schwartz, J. H. (1986). systematics and a juvenile mandible form Irhoud (Morocco): a phylo- classification of the order. In (D. Swindler & J. genetic interpretation. In (C. B. Stringer, Ed.) Aspects Erwin, Eds) Comparative Primate Biology, vol. 1,   ‘‘’’ 409

Systematics, Evolution and Anatomy, pp. 1–41. New Smith, F. H. (1984). Fossil hominids from the Upper York: Alan R. Liss. Pleistocene of Central Europe and the origin of Schwartz, J. H. (1999a). Homeobox genes, fossils, and modern Europeans. In (F. H. Smith & F. Spencer, the origin of species. Anat. Rec. (New Anat.) 257, Eds) The Origins of Modern Humans: a World Survey 15–31. of the Fossil Evidence, pp. 137–209. New York: Alan Schwartz, J. H. (1999b). Sudden Origins: Fossils, Genes, R. Liss. and the Emergence of Species. New York: John Wiley & Smith, P., Gomorri, J. M., Spitz, S. & Becker, J. Sons. (1997). Model for the examination of evolutionary Schwartz, J. H. (1999c). Evolutionary provocations: trends in tooth development. Am. J. phys. Anthrop. Paul Sondaar, the evolution of the horse, and a new 102, 283–294. look at the origin of species. In (J. de Vos, Ed.) Stringer, C. B., Hublin, J.-J. & Vandermeersch, B. Festschrift for Paul Sondaar. Deinsea. (1984). The origin of anatomically modern humans Schwartz, J. H. & Houghton, F. (in preparation). The in western Europe. In (F. H. Smith & F. Spencer, cremated skeletal remains from Punic Carthage. Eds) The Origins of Modern Humans: A World Survey In (L. E. Stager & J. Greene, Eds) The American from the Fossil Evidence, pp. 51–135. New York: Alan Excavations at Carthage. R. Liss. Schwartz, J. H. & Tattersall, I. (1996a). Significance of Tattersall, I. (1986). Species recognition in human some previously unrecognized apomorphies in the . J. hum. Evol. 15, 165–175. nasal region of Homo neanderthalensis. Proc. Natl. Ten Cate, A. R. & Mills, C. (1972). The development Acad. Sci. 93, 10,852–10,854. of the periodontium: The origin of alveolar bone. Schwartz, J. H. & Tattersall, I. (1999b). Toward distin- Anat. Rec. 173, 69–78. guishing Homo neanderthalensis from Homo sapiens, Tillier, A.-M. (1979). Restes craniens de l’enfant and vice versa. Anthropologie 43, 129–138. Mouste´rian Homo 4 de Qafzeh (Isra¨el), la mandibule Schwartz, J. H. & Tattersall, I. (1999a). Toward a et les maxillaires. Pale´orient 5, 67–85. definition of Homo sapiens and Homo neanderthalensis: Tillier, A.-M. (1983). Le craˆne d’enfant d’: un I. The nasal region. In (J. Gibert, F. Sa´nchez, L. exemple de distribution des caracte`res juve´niles, Gibert & F. Ribot, Eds) The Hominids and Their primitifs et ne´anderthaliens. Bull. Soc. Roy. Belge Environments During the Lower and Middle Pleistocene Anthrop. Pre´hist. 94, 51–75. of Eurasia, pp. 299–310. Orce: Museo de Prehistoria Vlce˘k, E. (1993). Fossile Menschenfunde von y Paleontolog´ıa. Weimar-Ehringsdorf. Stuttgart: Konrad Theiss Verlag. Schwartz, J. H. & Tattersall, I. (1999b). Variability in Wallis, W. D. (1917). The development of the human hominid evolution: putting the cart before the horse? chin. Anat. Rec. 12, 315. In (M. Antunes, Ed.) Last Neanderthals in Europe: Wolpoff, M. H. (1980). Paleoanthropology. New York: Odontologic and Other Evidence. Lisbon: Academy of Alfred A. Knopf. Sciences. Wolpoff, M. H. (1996). Human Evolution. New York: Schwartz, J. H. & Tattersall, I. (in preparation). Atlas of McGraw Hill. Fossil Hominids, Volume 1: Homo (Europe).New Wolpoff, M., Smith, F., Malez, M., Radovcˇic´,J.& York: John Wiley & Sons. Rukavina, D. (1981). Upper Pleistocene hominid Schwartz, J. H., Tattersall, I. & Laitman, J. L. (1999). remains from Vindija Cave, Croatia, Yugoslavia. Am. New thoughts on : evidence J. phys. Anthrop. 54, 499–545. from nasal morphology. In (H. Ullrich, Ed.) Hominid Zollikofer, C. P. E., Ponce de Leo´n, M. S., Martin, Evolution: Lifestyles and Strategies, pp. 166–186. R. D. & Stucki, P. (1995). Neanderthal computer Gelsenkirchen/Schwelm: Edition Archaea. skulls. Nature 375, 283–285.

Note added in proof: Since this manuscript went to press, we did study the Klasies River Mouth specimens. Our preliminary suggestions were borne out.