AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 145:615–628 (2011) Dietary Effects on Development of the Human Mandibular Corpus Megan A. Holmes* and Christopher B. Ruff Center for Functional Anatomy and Evolution, Johns Hopkins University School of Medicine, 1830 East Monument Street, Baltimore, MD 21205 KEY WORDS ontogeny; mandible; cross sectional properties; functional adaptation ABSTRACT The extent to which the mandibular corpus exhibits developmental plasticity has important implications for interpreting variation in adult and juvenile mandibular morphology in the archaeological and paleontological record. Here, we examine ontogenetic changes in mandibular corpus breadth, rigidity, and strength in two population samples with contrasting diets: late prehistoric Tigara from Point Hope, Alaska, characterized by a very demanding masticatory regime, and proto-historic Arikara from the Sully Site in South Dakota, with a less demanding regime. A total of 52 juvenile and 11 adult Tigara, and 32 juvenile and 10 adult Arikara were included in the study. Juveniles ranged in age from 1 to 17 years, with good representation of younger (1–6-year-old) juveniles (20 Arikara, 18 Tigara). Superoinferior and buccolingual external and cortical bone breadths of mandibles were measured at the Pm4M1 and M1-M2 junctions using calipers and biplanar radiographs, respectively. An asymmetrical hollow beam model was employed to reconstruct cross sections and calculate bending rigidities and strengths in the sagittal and transverse planes. Among adults, Tigara have greater transverse corpus width, bending rigidity, and strength, and ratios of transverse to sagittal dimensions than Arikara. This shape difference develops gradually during growth, with only weak trends among young juveniles, increasing to near-adult contrasts among adolescents. These results support a role for functional mechanical loading of the mandible during growth in producing adult differences in mandibular corpus morpholV 2011 ogy. Am J Phys Anthropol 145:615–628, 2011. Variation in mandibular morphology has long been related to variation in masticatory loadings (e.g., Hylander, 1977, 1979; Bouvier, 1986; Daegling, 1989; Biknevicius and Ruff, 1992a; Ravosa, 1996); based on this, albeit complex relationship (Daegling, 2007), differences in mandibular form are often used to infer dietary differences between past species or populations of the same species (e.g., Carlson and Van Gerven, 1979; Hylander, 1988; Daegling and Grine, 1991; Ravosa, 1996). In this context, one question that arises is to what extent mandibular morphology reﬂects developmental plasticity, i.e., a direct growth-related response to masticatory loadings, rather than (or in addition to) heritable (genetic) variation. The question is important in that it bears on both broader phylogenetic interpretations of craniofacial morphological variability (e.g., Antón et al., 2010), including ontogenetic variability (e.g., Mallegni and Trinkaus, 1997; Crevecouer et al., 2010), as well as narrower behavioral reconstructions in the archaeological record (Larsen, 1997; Paschetta et al., 2010). Experimental studies have demonstrated that the mandible can exhibit developmental plasticity in response to dietary manipulation, with animals fed harder diets during growth developing thicker and stronger mandibular corpora (among other changes) (Bouvier and Hylander, 1981; Lieberman et al., 2004; Organ et al., 2006). In this respect, the mandibular corpus behaves in a way analogous to that of long bones subjected to increased mechanical loading during life (Ruff et al., 2006 and references therein). However, there is also evidence that some population differences in relative long bone strength may be present very early during ontogeny (Cowgill, 2010; also see Wallace et al., 2010). Fukase and Suwa (2008) reported similar ﬁndings for samples of mandibles from prehistoric Jomon and modern Japanese. Adult Jomon mandibular corpora had previously been shown to have thicker cortical bone than those of modern Japanese, which was interpreted as a result of a tougher diet among Jomon (Kanazawa and Kasai, 1998). Fukase and Suwa showed that this was also true throughout development at the mandibular symphysis. They concluded that their results ‘‘reject the hypothesis that bone mass and deployment in the mandibular symphysis are primarily determined by stress generated by mastication,’’ ‘‘although a common and coherent pattern of growth-related changes in cortical bone thickness. . . suggest(s) that mechanical stress has a signiﬁcant regulatory role during growth remodeling of the symphysis’’ (Fukase and Suwa, 2008: 451-452). They also found that some other, external dimensions of the mandible, including buccal-lingual breadth of the corpus and symphyseal height and breadth, were consistently larger in Jomon mandibles beginning in infancy, suggesting that they were ‘‘manifestations of genetic patterning,’’ (P. 451) rather than a result of environmental inﬂuences during development. C 2011 V WILEY-LISS, INC. C Wiley-Liss, Inc. *Correspondence to: Megan A. Holmes, Center for Functional Anatomy and Evolution, Johns Hopkins University School of Medicine, 1830 East Monument Street, #301, Baltimore, MD 21205. E-mail: firstname.lastname@example.org Received 21 December 2010; accepted 18 April 2011 DOI 10.1002/ajpa.21554 Published online 23 June 2011 in Wiley Online Library (wileyonlinelibrary.com). 616 M.A. HOLMES AND C.B. RUFF The samples studied by Fukase and Suwa (2008) were, however, relatively small (total preadolescent N 5 20/19 for Jomon/modern), especially in infancy (1–3 years, N 5 4/6). Also, internal bone dimensions were only measured (using computed tomography) at the symphysis, not under the tooth rows, and no actual cross sectional measures of bone strength or rigidity (second moments of area and section moduli) were calculated. As shown previously (Daegling, 1989; Biknevicius and Ruff, 1992b), external breadths are of only limited usefulness and accuracy in assessing mechanical characteristics of the mandible, particularly when the aim is to identify relatively subtle differences. Structural interpretation of the mandibular symphysis is also complicated by a number of factors, including variation in mandibular curvature and symphyseal inclination (Hylander, 1984, 1988; Daegling, 2001), neither of which was assessed in the Fukase and Suwa (2008) study. In contrast, the more distal mandibular corpus is relatively straight and may be more easily modeled (and interpreted) using standard beam theory (Hylander, 1988; Daegling, 1989; Biknevicius and Ruff, 1992a). Here, we carry out a comparison of cross sectional geometric properties of the distal mandibular corpus in two growth series derived from populations with contrasting dietary regimes and adult mandibular morphologies (see below)—Tigara Inupiats and Arikara Native Americans. The null hypothesis is that juveniles and adults from these samples will exhibit the same distinctions in mandibular rigidity and strength. If true, this would support Fukase and Suwa’s (2008) contention that mandibular morphology is largely genetically controlled, with little direct effect of contrasting dietary regimes. Disproving the null hypothesis would support a role for masticatory loadings during development in producing variation in adult mandibular morphology. MATERIALS AND METHODS Samples The populations employed in this study represent two genetically and regionally distinct groups with contrasting dietary practices. Tigara specimens represent a circum-polar population which occupied Point Hope, Alaska between 1200 and 1700 A.D (Larsen and Rainey, 1948). Archaeological and ethnographic data indicate a diet consisting of dried meats exploited from marine resources such as ﬁsh and sea mammals, which would require heavy masticatory loading for consumption (Rainey, 1941; El-Zaatari, 2008, 2010). In addition, there is evidence that the Tigara population utilized their dentition as a tool for gripping, stripping, and softening leather (Merbs, 1968; Costa, 1980). This behavior was also witnessed in the native Alaskan populations during the ﬁrst half of the 20th century (de Poncin, 1941). These behavioral observations, as well as other evidence, such as high frequencies of dental chipping, fracturing, and root resorption, as well as direct measurements of high bite forces, have led researchers to treat arctic populations as an exemplar of adaptation to strenuous masticatory and paramasticatory forces (Turner and Cadien, 1969; Hylander, 1977; Scott and Winn, 2010). In fact, these populations have been repeatedly utilized in comparative studies involving fossil hominids as they are thought to emulate in some ways earlier dietary regimes (Hylander, 1988; Spencer and Demes, 1993; Nicholson and Harvati, 2006). They, thus, represent a modern extreme for mastiAmerican Journal of Physical Anthropology TABLE 1. Sample size for Arikara and Tigara populations arranged by age group and cross-section location Pm4-M1 Age (yr) 1–6 6–12 12–18 Adults Total M1-M2 Arikara Tigara Total Arikara Tigara Total 20 6 6 10 42 18 23 11 11 63 37 29 17 21 105 9 5 6 10 30 5 23 11 11 50 14 28 17 21 80 catory loading. A total of 63 individuals—52 juveniles and 11 adults—were obtained from the collections of the American Museum of Natural History (AMNH), New York (Table 1). The second comparative sample, Arikara from the Sully Site, South Dakota, provide a model for a less strenuous diet. This skeletal population was excavated from a proto-historic site dating from 1650 to 1750 A.D (Jantz, 1973; Jantz and Owsley, 1984; Owsley and Jantz, 1985). Seasonal climates allowed for a relatively varied subsistence strategy that included horticulture and hunting. Dental wear analyses of protohistoric Arikara populations also suggest that their diet was mixed, consisting of both plant and animal foods (Ungar and Spencer, 1999; El Zaatari, 2008). Though the Arikara are thought to have consumed dried meat like the Tigara (Ungar and Spencer, 1999), direct comparisons of dental microwear patterns between these two populations demonstrate their dietary differences, and are consistent with heavier loading (as well as a more abrasive diet) among the Tigara (El Zaatari, 2008). It may be gathered from this information that the Tigara population possessed a diet that included generally higher occlusal loads and greater masticatory demands relative to that of the Arikara. A total of 42 individuals—32 juveniles and 10 adults—were obtained from the collections of the Smithsonian Institution National Museum of Natural History (NMNH), Washington, DC. The growth series utilized here encompass all developmental stages from infancy to adulthood (Table 1 and subsequent ﬁgures). Individuals were aged using a combination of Liversidge and Molleson’s (2004) deciduous tooth root length regression equations, Smith’s (1991) dental development sequence, and timing of long bone epiphyseal fusion among adolescents (Scheuer and Black, 2000). Those specimens presenting a fully erupted M3 were considered adults. For discrete age group comparisons, juveniles were organized into three subgroups demarcated by the eruption of the ﬁrst two permanent mandibular molars, roughly coinciding with ages of 6 and 12 years, i.e., 1–6, 6–12, and 12–18 years (Table 1). No attempt was made to sex pre-adolescents; older adolescents and adults contained an approximately equal number of male and females in each sample, determined from pelvic morphology (Buikstra and Ubelaker, 1994), to control for possible sexual dimorphism. To limit possible effects of tooth loss, tooth wear, and other consequences of aging, adults were limited to individuals between 18 and 28 years, determined using the presence of a fully erupted M3 and standard postcranial observations (Buikstra and Ubelaker, 1994). In addition, edentulous mandibles, mandibles with extensive damage and/or mandibles that did not meet the requirements for length measurement were excluded from this analysis. When available, measurements were preferably taken on the GROWTH OF THE HUMAN MANDIBULAR CORPUS 617 TABLE 2. List of measurements and deﬁnitions Abbreviation ML Dx Dy Iy Ix Zy Zx Measurement Deﬁnition Mandibular Length Corpus Width Corpus Height Transverse bending rigidity Sagittal bending rigidity Transverse bending strength Sagittal bending strength distance from the center of the condyle to the Pm3-C junction transverse (buccolingual) corpus breadth (mm) sagittal (superior-inferior corpus breadth (mm) second moment of area about the sagittal axis (mm4) second moment of area about the transverse axis (mm4) section modulus about the sagittal axis (mm3) section modulus about the transverse axis (mm3) Fig. 1. Calculation of cross-sectional geometric properties using the asymmetrical hollow beam model. A, B: Buccolingual and superoinferior radiographs of a juvenile mandible showing corpus height, Dy, and breadth, Dx, taken with calipers on specimens, and inferior, Ic, lingual, lc and buccal, bc, cortical breadths, measured using ImageJ on digital radiographs, at the Pm1-M1 location. C: These measurements are then used to calculate total subperiosteal (solid line) and endosteal (dotted line) areas and second moments of area, as well as the centroid for each (solid x, subperiosteal centroid; dotted x, medullary centroid). D: Subsequently, cortical area and the composite centroid (x) are calculated, which are then used with periosteal and endosteal properties to calculate composite second moments of area and section moduli using asymmetrical hollow beam formulae (http://www.hopkinsmedicine.org/ fae/CBR.htm; see Biknevicius and Ruff (1992) and O’Neill and Ruff (2004) for discussion). right side of the mandible, however, left sides were measured when necessary to maximize sample sizes. Measurements Both linear measurements and cross sectional geometric properties of the mandible were collected for this study (see Table 2). Linear dimensions, measured using digital calipers, included mandibular length, measured from the center of the articular process to the Pm3-C junction, and corpus height (Dy) and width (Dx), measured perpendicular to the long axis of the mandible in the sagittal (superoinferior) and transverse (buccolingual) planes, respectively (see Fig. 1). Dx and Dy and other cross sectional measurements were determined at two locations, behind dm2 or Pm4 if erupted (Pm4-M1 margin), and behind M1 (M1-M2 margin). Before eruption of dm2 or M1, the developing crowns, when visible American Journal of Physical Anthropology 618 M.A. HOLMES AND C.B. RUFF TABLE 3. Comparison of Arikara and Tigara adult mandibular dimensions Arikara Adults Tigara Adults Propertya Mean SE Mean SE p-valueb ML 98.1 1.47 99.9 1.33 0.392 Dx Dy Iy Ix Zy Zx Dx/Dy Iy/Ix Zy/Zx 14.9 32.2 3425 9898 304 591 0.465 0.356 0.521 0.42 0.89 254 767 14.9 32.8 0.017 0.025 0.022 18.6 31.5 6273 10689 484 651 0.593 0.587 0.743 0.65 0.68 626 763 33.2 33.3 0.020 0.037 0.029 0.000 0.517 0.001c 0.474 0.000c 0.219 0.000 0.000 0.000 Dx Dy Iy Ix Zy Zx Dx/Dy Iy/Ix Zy/Zx 17.3 29.7 4592 8172 364 528 0.581 0.581 0.697 0.36 0.81 220 664 11.9 30.8 0.069 0.046 0.031 20.3 28.6 6854 8025 490 538 0.710 0.855 0.905 0.59 0.52 714 719 43.4 40.6 0.020 0.048 0.031 0.000 0.263 0.011c 0.884 0.017c 0.853 0.000 0.000 Pm4-M1 M1-M2 a b c See Table 2 for property abbreviations. t tests. t test with unequal variance applied. radiographically (see below), were used to locate these positions. Lack of space behind M1 precluded measurements at this location in some of the youngest individuals (\ 3 years). Bi-planar radiography, in combination with external linear measurements, can be used to reconstruct bone cross sectional geometry (Biknevicius and Ruff, 1992a; O’Neill and Ruff, 2004). Radiographic digital images were obtained for each mandible in both the sagittal and transverse planes, with the superior alveolar border used to deﬁne the long axis of the mandibular corpus. Specimens collected at the NMNH were digitally radiographed using a Kevex PXS10-16W Microfocus X-Ray source at 68.9 kV and a PaxScan 4030R Varian System Flat Panel Amorphous Silicon Digital X-Ray. A portable Aribex NOMAD handheld X-Ray device (60 kV) and AFP Digital #2 sensor were utilized to obtain radiographic digital images of specimens housed at the AMNH. Cortical thicknesses were measured on the buccal, lingual, and inferior aspects of the mandible on the digital radiographic images using ImageJ (Abramoff et al., 2004) (see Fig. 1). Alveolar cortical bone was often too thin to visualize radiographically; therefore, a thin straight line (.1 mm thick) was used to complete the alveolar perimeter. The mandibular corpus was modeled as an asymmetrical hollow beam, which has been shown to provide a good approximation to true cross sectional geometry of the corpus (Biknevicius and Ruff, 1992b). Cross sectional properties were derived using a custom macro (‘‘EEM_Macro,’’ available at http://www.hopkinsmedicine.org/fae/CBR.htm; also see O’Neill and Ruff, 2004). Biomechanical variables assessed included measures of bending rigidity (I) and strength (Z) in the sagittal (Zx, Ix) and transverse (Zy, Iy) planes. (Note that rigidities and strengths are measured about, i.e., perpendicular to, the designated axes.) Transverse to supero-inAmerican Journal of Physical Anthropology ferior ‘‘shape’’ indices were also calculated, from both linear measures (Dx/Dy) and cross sectional parameters (Iy/ Ix; Zy/Zx). These are analogous to traditional mandibular ‘‘robusticity’’ indices (Chamberlain and Wood, 1985; Daegling and Grine, 1991). Statistical analyses Mandibular length dimensions are often used as a mandibular size proxy (Hylander, 1979, 1988; Bouvier, 1986; Biknevicius and Ruff, 1992b; Cole, 1992; Daegling, 2007). To assess whether size standardization was necessary in comparisons between the Tigara and Arikara samples, we ﬁrst compared mandibular lengths between the samples. As shown below, mandibular length is nonsigniﬁcantly different between Tigara and Arikara adults; the two samples are also equivalent in mandibular length in the earliest (1–6 year) age group (66.0 6 1.4 mm (SE) in Tigara; 65.2 6 2.0 mm in Arikara; t 5 0.32; P 5 0.76). Therefore, all comparisons between groups were carried out using nonsize-standardized data. This also allows for more direct visualization of actual growth trajectories. All data (except shape ratios) were log-transformed in order to normalize distributions and preserve proportionality over the entire age range examined. Adult dimensions were initially compared between the two samples using t tests, to assess whether signiﬁcant differences in morphology existed at the termination of growth. Because growth itself is nonlinear, however, and because age distributions of the two samples were not entirely equivalent, even within the 6-year-age groups employed, a more complex method was necessary to carry out ontogenetic comparisons. Polynomial regressions were ﬁt to pooled sample distributions for each variable, and regression residuals from these lines were used to compare samples (Pinhasi et al., 2005, 2006; Mays et al., 2009). The correct order of the polynomials was determined using a forward step criterion. It was found that second and third order polynomials matched growth of the mandibular corpus best for almost all properties, except for shape ratios at the M1-M2 junction, which were best ﬁt by linear regressions. Subsequently, nonstandardized residuals were calculated as each individual’s vertical distance from the predicted pooled growth trajectory (polynomial line), and used in t tests between Tigara and Arikara within each of the four developmental age groups (including adults as a group). A probability value of 0.05 was considered statistically signiﬁcant in individual comparisons. However, because multiple comparisons across four groups were employed here, a Bonferroni correction of 0.05/4 5 0.0125 is appropriate. Both signiﬁcance levels are indicated in the tabulation of results. Where variances were not equivalent, t tests with unequal variances were applied. One additional statistical analysis was also carried out, again utilizing residuals from the polynomial regressions. ANCOVA was used to compare overall change in residuals (slope) between the population samples with increasing age (as a covariate). Signiﬁcant interaction terms between population and age indicate divergent slope trajectories of the residuals, demonstrating divergence in overall growth patterns. A limitation of this approach is that it ﬁts residuals using a linear model, whereas population divergences during growth do not occur at a constant rate (see below). However, because it GROWTH OF THE HUMAN MANDIBULAR CORPUS 619 Fig. 2. Scatterplots and polynomial regressions on age for each variable, measured at the Pm4-M1 junction. Mandibular breadth dimensions on left, height dimensions on right. See Table 2 for abbreviations. Filled black circles: Tigara; open triangles: Arikara. combines information from all age periods, it is a useful adjunct to the discrete age groups comparisons. All statistical analyses were carried out using SPSS ver. 17.0 (SPSS) while Statistica 7.0 (Statsoft) was used to generate graphical presentations. RESULTS Comparisons between the adult sample dimensions are shown in Table 3. Arikara and Tigara adult mandibles do not differ in mandibular length or corpus height dimensions (Dy, Ix, Zx) at either the Pm4-M1 or M1-M2 locations. However, Tigara adults are signiﬁcantly larger in corpus breadth dimensions (Dx, Iy, Zy) at both loca- tions. As a consequence, cross sectional ‘‘shape’’ ratios, Dx/Dy, Iy/Ix, and Zy/Zx, are all signiﬁcantly greater among Tigara adults, indicating that they have relatively rounder, more mediolaterally buttressed cross sections. Scatterplots and polynomial regressions of mandibular cross sectional dimensions on age are shown in Figures 2 and 3 for the Pm4-M1 and M1-M2 locations, respectively and for the three shape ratios in Figure 4. Comparisons of residuals from the ﬁtted polynomial curves in each of the four age categories are shown in Table 4. Box plots of residuals by age group are given in Figures 5–7 for dimensions measured at Pm4-M1 and M1-M2, and shape ratios, respectively. American Journal of Physical Anthropology 620 M.A. HOLMES AND C.B. RUFF Fig. 3. Scatterplots and polynomial regressions on age for each variable, measured at the M1-M2 junction. Mandibular breadth dimensions on left, height dimensions on right. See Table 2 for abbreviations. Filled black circles: Tigara; open triangles: Arikara. No differences between population samples reach statistical signiﬁcance with Bonferroni correction (p \0.0125) in the 1–6 year age group (Table 4). Some breadth dimensions (Dx and Iy) and the three shape indices reach signiﬁcance at the Pm4-M1 location in noncorrected comparisons (P \0.05), in all cases being greater in Tigara (Figs. 5A,C and 7A,C,E). No differences reach signiﬁcance using either criterion at the M1-M2 location in the earliest age group. In the 6–12 year age group, the mandibular strength ratio, Zy/Zx, at M1-M2 is signiﬁcantly larger among Tigara at the Bonferroni-corrected level (Table 4; Fig. 6E). No other comparisons reach this level of signiﬁcance, although Zy/Zx at Pm4-M1, Dx/Dy (both locations) American Journal of Physical Anthropology and Iy/Ix (M1-M2) are signiﬁcant at the uncorrected level (all larger in Tigara) (Fig. 7A,C,D). All of the shape ratios become highly signiﬁcantly different (p .003) between samples during adolescence (12–18 years), in all cases being larger in Tigara (Table 4; Fig. 7). Mandibular external height (Dy) is signiﬁcantly greater among Arikara adolescents, reaching the Bonferroni-corrected level at M1-M2 (Fig. 6B), and the uncorrected level at Pm4-M1 (Fig. 5B). Differences among adults in polynomial residuals parallel those observed earlier using simple t tests (Table 3). Direct comparisons of individual data points and box plots of age group distributions for shape ratios (Figs. 4 and 7) clearly demonstrate the increasing contrast in GROWTH OF THE HUMAN MANDIBULAR CORPUS 621 Fig. 4. Scatterplots and polynomial regressions on age for shape ratios, measured at the Pm4-M1 (left) and M1-M2 (right) junctions. See Table 2 for abbreviations. Filled black circles: Tigara; open triangles: Arikara. cross sectional shape between Tigara and Arikara that develops throughout ontogeny and which characterizes adults. There is extensive overlap in shape between the populations early in development, but by adolescence they are almost completely nonoverlapping. This is mainly due to relatively greater increases in mandibular breadth dimensions among Tigara, although a transient increase in relative height among Arikara also contributes during adolescence (Figs. 2, 4, 5, and 6). Both samples demonstrate a progressive increase during development in relative parasagittal bending rigidity and strength at both locations (Figs. 2–4). These observations are corroborated by the ANCOVA results for regression residuals on age (Table 5). The external breadth (Dx/Dy) and strength (Zy/Zx) ratios at both mandibular locations show signiﬁcant populationage interactions, reﬂecting the increasing divergence of values between samples with age. As expected for these residual analyses, there are no age effects. There is a marginally signiﬁcant (p 5 0.047) effect of population on the Iy/Ix rigidity ratio, only at the Pm4-M1 location. DISCUSSION Mandibular corpus morphology is clearly distinct in adult Tigara and Arikara, with Tigara exhibiting signiﬁcantly more mediolaterally buttressed corpora mesial and distal to M1. This is evident in both external linear American Journal of Physical Anthropology 622 M.A. HOLMES AND C.B. RUFF TABLE 4. Results of t tests (p-values) between population residuals from polynomial regressions at each age interval. Individually signiﬁcant results (p < 0.05) are italicized Pm4-M1 M21-M2 Propertya 1–6 yr 6–12 yr 12–18 yr Adults 1–6 yr 6–12 yr 12–18 yr Adults Dx Dy Iy Ix Zy Zx Dx/Dy Iy/Ix Zy/Zx 0.019 0.914 0.047 0.544 0.091 0.521 0.040 0.026 0.024 0.291 0.089 0.891 0.181 0.982 0.168 0.027 0.051 0.013 0.031 0.016b 0.563 0.066 0.823 0.089 0.000* 0.000* 0.000* 0.000* 0.193 0.001* 0.801 0.000* 0.454 0.000* 0.000* 0.000* 0.544 0.143 0.399 0.266 0.382 0.451 0.742 0.327 0.515 0.091 0.150 0.789 0.029 0.283 0.023 0.014 0.022 0.011* 0.142 0.003*,b 0.252 0.004* 0.070 0.005* 0.002* 0.002* 0.003* 0.000* 0.092 0.017 0.667 0.029b 0.984 0.000* 0.000* 0.000* All data, except ratios, are log-transformed.* p-values meet Bonferroni correction level (p \ 0.0125). See Table 2 for abbreviations. t test with unequal variance applied. a b Fig. 5. Boxplots of residuals from polynomial regressions for dimensions measured at the Pm4-M1 junction (see Fig. 2) for four age groups in Arikara (open) and Tigara (shaded). See Table 2 for abbreviations. American Journal of Physical Anthropology GROWTH OF THE HUMAN MANDIBULAR CORPUS 623 Fig. 6. Boxplots of residuals from polynomial regressions for dimensions measured at the M1-M2 junction (see Fig. 3) for four age groups in Arikara (open) and Tigara (shaded). See Table 2 for abbreviations. dimensions, as well as measures of rigidity and strength that also take into account cortical bone thickness and distribution. This morphological distinction is not simply a result of size differences between the populations, mandibular length is equivalent, as are mandibular height and sagittal rigidity and strength dimensions. This shape difference, most clearly expressed as ratios between buccolingual and sagittal dimensions, progressively develops throughout ontogeny. There is a slight tendency, not statistically signiﬁcant by more stringent multiple-group comparison criteria, for individuals in our youngest age group, 1–6 years, to exhibit this difference at the Pm1-M1, but not M1-M2 location. However, the difference between groups becomes much more marked in older age groups, reaching adult levels of sig- niﬁcance by adolescence. This is also indicated by the signiﬁcant population-age interaction terms for external breadth and strength ratios at both mandibular locations. The shape difference among adolescents and adults is mainly a result of greater growth in buccolingual dimensions among Tigara. Thus, like Fukase and Suwa (2008) and others (Carlson and Van Gerven, 1977; Spencer and Demes, 1993; Antón et al., 2010), we ﬁnd that adults from populations with contrasting dietary regimes exhibit differences in mandibular morphology. However, unlike Fukase and Suwa (2008), we ﬁnd much less evidence for such differences among the youngest juveniles from the same populations, who are much less clearly differentiated than older juveniles and adults. Thus, our results sugAmerican Journal of Physical Anthropology 624 M.A. HOLMES AND C.B. RUFF Fig. 7. Boxplots of residuals from polynomial regressions for shape ratios measured at the Pm4-M1 and M1-M2 junction (see Fig. 4) for four age groups in Arikara (open) and Tigara (shaded). See Table 2 for abbreviations. gest that functional adaptation of the mandible to masticatory loading during growth and development may contribute to the ultimate morphological differences characteristic of adults. The contrast in results between our study and those of Fukase and Suwa (2008) may be in part attributable to the different mandibular dimensions sampled, as well as the available sample sizes. As noted earlier, biomechanical interpretation of symphyseal morphology, the main focus of the Fukase and Suwa study, is potentially more difﬁcult than that of the midcorpus region sampled here (Hylander, 1984, 1988; Daegling, 2001). It is interesting American Journal of Physical Anthropology in this regard that Fukase and Suwa found no differences in symphyseal bending rigidities between their adult Jomon and modern Japanese samples (their data also imply no difference in torsional rigidity), despite differences in absolute and relative cortical thicknesses. We also had available a larger total sample of juveniles (N 5 83 vs. 54 in the Fukase and Suwa study), particularly in the youngest age ranges (N 5 37 vs. 23 for 1–6 years, N 5 21 vs. 10 for 1–3 years). This allowed ﬁtting of more realistic nonlinear growth curves to the data, as well as more direct comparisons of dimensions within speciﬁc age groups, whereas the Fukase and Suwa study GROWTH OF THE HUMAN MANDIBULAR CORPUS TABLE 5. ANCOVA results (p-values) for residuals from polynomial regressions on age Pm4-M1 M1-M2 Propertya Population Age Pop*age Population Age Pop*age Dx Dy Iy Ix Zy Zx Dx/Dy Iy/Ix Zy/Zx 0.433 0.975 0.448 0.796 0.566 0.831 0.176 0.047 0.079 0.671 0.851 0.828 0.904 0.839 0.928 0.705 0.847 0.719 0.003 0.158 0.153 0.326 0.170 0.458 0.014 0.366 0.027 0.769 0.827 0.934 0.682 0.988 0.832 0.641 0.930 0.679 0.894 0.888 0.965 0.899 0.976 0.931 0.766 0.898 0.898 0.029 0.076 0.469 0.165 0.657 0.307 0.016 0.055 0.025 Signiﬁcant results (p \ 0.05) are italicized. All data, except ratios, are log-transformed.a See Table 2 for abbreviations. relied on comparisons of estimated y intercepts (i.e., downwardly extrapolated birth values) based on linear regressions on age in 1–8-year-olds. It is also possible that the difference in results between the two studies is due at least in part to greater genetic differentiation between Fukase and Suwa’s samples than in ours. That is, Jomon and modern Japanese may have evolved different patterns of mandibular growth, beginning prenatally, in response to long-term dietary (or other) selection, whereas Tigara and Arikara did not. Comparisons of ontogenetic trajectories among other population samples, particularly biologically related populations experiencing known changes in dietary patterns, would help to further clarify this issue. A potentially confounding effect on group comparisons in either the Fukase and Suwa (2008) or present studies is the possibility of unaccounted for sexual dimorphism in mandibular morphology in juvenile samples. The Fukase and Suwa study included only males in their adult samples, whereas all of their Jomon and some modern Japanese juveniles were not sexed. We included an approximately equal number of adult and older adolescent males and females, but could not sex younger juveniles. If any of the juvenile samples included an unbalanced sex distribution, this could have affected age trends if sexual dimorphism of the mandible exists among juveniles. Some earlier studies did in fact report sexual dimorphism in certain features of the mandible in young children (Schutkowski, 1993; Loth and Henneberg, 2001), although these results were later questioned (Coqueugniot et al., 2002; Scheuer, 2002). Recent morphometric studies indicate that very little sexual dimorphism of the mandible exists after early infancy and before midadolescence (Franklin et al., 2007; Coquerelle et al., in press). Thus, this factor is unlikely to bias growth studies of mixed-sex samples in this age range. Inclusion of only adult males in the Fukase and Suwa (2008) samples, however, introduces a potential effect on the adult comparisons in that sexual dimorphism of the mandible is marked among adults, particularly in areas such as the chin (e.g., Buikstra and Ubelaker, 1994; Thayer and Dobson, 2010); thus, population-level differences in symphyseal morphology may be greater or less within single-sex rather than mixed-sex samples. The development of a rounder, more buccolingually buttressed posterior mandibular corpus among Tigara suggests that their mandibles were better adapted to resist torsional loadings than were those of the Arikara (Hylander, 1988; Daegling and Grine, 1991). Although several factors can inﬂuence the magnitude of such loadings, powerful masticatory forces are among the most 625 important (Hylander, 1988; Daegling and Grine, 1991). Overall craniofacial morphology in Inupiat and Inuit populations is consistent with the generation of high bite forces, both anteriorly and posteriorly (Hylander, 1977; Spencer and Demes, 1993), as are direct behavioral observations of these populations (see above). Although ontogenetic analyses of these general structural and behavioral characteristics have not, to our knowledge, been carried out among these populations, it seems plausible that young children would not have been exposed to the same levels of masticatory stress (even scaled for body and/or mandibular size) as adults. This certainly would have been the case for young infants who had not yet transitioned to adult diets. Weaning age among recent Alaskan Inupiats has been estimated to be about 3–4 years (Chance, 1990). Use of the teeth for paramasticatory functions, i.e., as tools (Merbs, 1968; Costa, 1980), would also likely have become common only after early childhood. Eruption and occlusion of the adult dentition itself, between 6 and 12 years, may have been necessary to transmit high masticatory loadings (although teeth of Native arctic populations are not relatively large; Hylander, 1977). All of these general timing factors would have applied to Arikara as well as Tigara; the fact that the two populations diverge in morphology during growth is indicative of progressive functional (mechanical) differentiation within this common developmental framework. Previous authors have emphasized the increase in parasagittal bending strength or height of the mandibular corpus in Native arctic samples (Hylander, 1977; Antón et al., 2010), which we did not ﬁnd here (in comparison to Arikara). However, Hylander (1977) based his conclusions on a compilation by Hrdlicka (1940), which did not include mandibular corpus breadth, so it is not possible to assess whether that may have been even more increased in the arctic samples. Furthermore, examination of Hrdlicka’s original data shows that while one particular ‘‘Eskimo’’ sample (from Kodiak Island) had a large mandibular corpus height, ‘‘Eskimos, in general’’ fall well within the range, and even slightly below the median, of the other populations sampled, which were world-wide in distribution. Thus, there is no evidence that they had particularly large corpus heights. Antón et al. (2010) also emphasized the increased parasagittal bending rigidity among their ‘‘Alaskan’’ samples, but bending rigidities were only calculated for an Aleut sample, not their Inupiat sample from Point Hope. Their Point Hope sample had the largest corpus breadth and breadth/height ratio of any of their Holocene samples (the Aleuts were also comparatively high). In pairwise comparisons, corpus breadth showed more signiﬁcant differences between Point Hope and other samples than corpus height (the same was true for Aleuts). Therefore, it appears that this Inupiat sample also exhibits a relatively rounder, i.e., buccolingually buttressed mandibular corpus. It is also worth noting that the present study comparison of Tigara was with another population sample, the Arikara, that was also likely subjected to fairly heavy masticatory loading, at least in comparison with Western industrialized populations (see above). Thus, it may not be surprising that parasagittal bending rigidity and strength were not increased in Tigara relative to Arikara. In Hrdlička’s (1940) tabulation, US whites show considerably lower corpus heights than almost all of his other samples, including ‘‘Eskimos’’ and several lower latitude Native American samples. In addition to biomechanical factors, spatial requirements for the developing dentition have also been sugAmerican Journal of Physical Anthropology 626 M.A. HOLMES AND C.B. RUFF gested to exert an inﬂuence on mandibular size and shape (Wolpoff, 1975; Wood, 1978; Smith, 1983). Thus, it is possible that the relatively larger buccolingual breadth of the mandible in the Tigara sample may be the result of relatively wider buccolingual molar dimensions. Although there is no doubt that dental size may directly affect mandibular size, it is unlikely that this factor explains population differences in corpus shape between the Arikara and Tigara. First, previous comparative analyses have found little support for a direct relationship between molar width and distal mandibular corpus ‘‘robusticity’’ (breadth/height) (Daegling and Grine, 1991; Plavcan and Daegling, 2006). Second, there is no evidence that Inupiats and Inuits in general had particularly large molar dimensions when compared with other Native American groups, including Arikara. The lack of an increase in dental size among ‘‘Eskimos’’, despite evidence for increased mandibular robusticity, has been speciﬁcally remarked upon by Hawkes (1916) and Hylander (1972, 1977). The average M1 mesiodistal and buccal-lingual breadths of a series of 22 ‘‘Eskimo’’ population samples reported by Mayhall (1967) and Turner (1967) were 11.66 mm and 11.20 mm, respectively, very similar (especially buccal-lingual) to those measured for Arikara by Perzigian (1975): 11.54 mm and 11. 22 mm, respectively (sexes combined in both groups). Falk and Corruccini (1982) reported similar results for M1 breadths of Point Hope ‘‘Eskimos’’ compared with those of several other modern human groups. Thus, differences in dental size per se should not account for the differences in mandibular morphology that we found here. Our results also have some broader implications regarding the interpretation of mandibular form in earlier hominins, including among juvenile specimens. If at least some aspects of mandibular morphology are dependent on functional stimuli during development, then this implies that such characteristics may be less useful as phylogenetic markers (e.g., see discussion in Lieberman, 1996; Antón et al., 2010). In this study sample, this appears to apply to relative mandibular corpus rigidity and strength in buccolingual vs. parasagittal planes, a commonly noted distinction between different hominin taxa (e.g., Chamberlain and Wood, 1985; Hylander, 1988; Daegling and Grine, 1991). In this respect, it is interesting that even young juvenile Neandertal mandibles appear to express an increase in relative corpus breadth (for their age) compared with modern humans, as do adult Neandertals (Mallegni and Trinkaus, 1997; Crevecouer et al., 2010, Walker et al., 2010). It is, of course, likely that a combination of genetic and environmental effects operate during development to produce ﬁnal adult mandibular form. As with all skeletal elements, genetic selection may favor changes in relative strength (robusticity) of the mandible under appropriate environmental conditions (Cole, 1992; Cowgill, 2010; Wallace et al., 2010). However, here we ﬁnd that adult differences in mandibular morphology between two modern human populations with distinct but not extremely different masticatory loading regimes, Tigara and Arikara, appear to develop only after the initiation of adult patterns of masticatory behavior. CONCLUSIONS Tigara and Arikara adults differ in mandibular corpus morphology measured at the Pm4-M1 and M1-M2 junctions, with Tigara mandibles exhibiting greater buccolinAmerican Journal of Physical Anthropology gual bending rigidity and strength, measured radiographically. This structural difference develops throughout growth of the mandible, with very young juveniles (1–6 years) exhibiting a weak, statistically nonsigniﬁcant difference at Pm4-M1 only, older juveniles (6–12 years) exhibiting more consistent differences at both locations, and adolescents (12–18 years) an almost fully developed adult pattern. The greater rigidity and strength of the mandibular corpus among older juvenile and adult Tigara is likely related to their very demanding masticatory and para-masticatory regime. The progressive development of these features supports a role for functional adaptation of the mandible to applied mechanical loadings during growth. ACKNOWLEDGMENTS The authors would like to thank David R. Hunt (NMNH) and Giselle Garcia-Pack (AMNH) for access to the collections used in this study; and the Fishes Department (NMNH) for use of their digital X-ray machine. We are grateful to Evan M. Garofalo for all her assistance in sexing and aging of specimens. We also thank the reviewers and Associate Editor for insightful comments that helped improve this article. LITERATURE CITED Abramoff MD, Magelhaes PJ, Ram SJ. 2004. Image processing with ImageJ. Biophoton Int 11:36–42. Antón SC, Carter-Menn H, DeLeon VB. 2010. Modern human origins: continuity, replacement, and masticatory robusticity in Australasia. J Hum Evol 60:70–82. Biknevicius AR, Ruff CB. 1992a. Use of biplanar radiographs for estimating cross-sectional geometric properties of mandibles. Anat Rec 232:157–163. Biknevicius AR, Ruff CB. 1992b. The structure of the mandibular corpus and its relationship to feeding behaviours in extant carnivorans. J Zool 228:479–507. Bouvier M. 1986. A biomechanical analysis of mandibular scaling in old world monkeys. Am J Phys Anthropol 69:473–482. Bouvier M, Hylander WL. 1981. Effect of bone strain on cortical bone structure in macaques (Macaca mulatta). J Morphol 167:1–12. Buikstra JE, Ubelaker DH. 1994. Standards for data collection from human skeletal remains. Arkansas Archeological Survey research series no. 44. Fayetteville, AR: Arkansas Archeological Survey. Carlson DS, Van Gerven DP. 1977. Masticatory function and post-Pleistocene evolution in Nubia. Am J Phys Anthropol 46:495–506. Chamberlain A, Wood B. 1985. A reappraisal of variation in hominid mandibular corpus dimensions. Am J Phys Anthropol 66:399–405. Chance NA. 1990. The Iñupiat and Arctic Alaska: an ethnography of development. Fort Worth, TX: Holt Rinehart and Winston. Cole TM. 1992. Postnatal heterochrony of the masticatory apparatus in Cebus apella and Cebus albifrons. J Hum Evol 23:253–282. Coqueugniot H, Giacobini G, Malerba G. 2002. L’utilisation de caractères morphologiques dans la diagnose sexuelle des mandibules d’enfants. Application à la collection ostéologique de Turin (Italie). Bulletins Et Mémoires De La Société d’Anthropologie de Paris n.s. 14:131–139. Coquerelle M, Bookstein FL, Braga J, Halazonetis DJ, Weber GW, Mitteroecker P. In press. Sexual dimorphism of the human mandible and its association with dental development. Am J Phys Anthropol, 145: n/a. doi: 10.1002/ajpa.21485. Costa RL, Jr. 1980. Age, sex, and antemortem loss of teeth in prehistoric Eskimo samples from Point Hope and Kodiak Island, Alaska. Am J Phys Anthropol 53:579–587. GROWTH OF THE HUMAN MANDIBULAR CORPUS Cowgill LW. 2010. The ontogeny of Holocene and late Pleistocene human postcranial strength. Am J Phys Anthropol 141:16–37. Crevecoeur I, Bayle P, Rougier H, Maureille B, Higham T. 2010. The spy VI child: a newly discovered Neanderthal infant. J Hum Evol 59:641–656. Daegling DJ. 1989. Biomechanics of cross-sectional size and shape in the hominoid mandibular corpus. Am J Phys Anthropol 80:91–106. Daegling DJ. 2001. Biomechanical scaling of the hominoid mandibular symphysis. J Morphol 250:12–23. Daegling DJ. 2007. Relationship of bone utilization and biomechanical competence in hominoid mandibles. Arch Oral Biol 52:51–63. Daegling DJ, Grine FE. 1991. Compact bone distribution and biomechanics of early hominid mandibles. Am J Phys Anthropol 86:321–339. de Poncins G. 1941. Kabloona. New York: Reynal. El-Zaatari S. 2008. Occlusal molar microwear and the diets of the Ipiutak and Tigara populations (Point Hope) with comparisons to the Aleut and Arikara. J Archaeol Sci 35:2517–2522. El-Zaatari S. 2010. Occlusal microwear texture analysis and the diets of historical/prehistoric hunter-gatherers. Int J Osteoarchaeol 20:67–87. Falk D, Corruccini R. 1982. Efﬁcacy of cranial versus dental measurements for separating human populations. Am J Phys Anthropol 57:123–127. Franklin D, Oxnard CE, O’Higgins P, Dadour I. 2007. Sexual dimorphism in the subadult mandible: quantiﬁcation using geometric morphometrics. J Forensic Sci 52:6–10. Fukase H, Suwa G. 2008. Growth-related changes in prehistoric Jomon and modern Japanese mandibles with emphasis on cortical bone distribution. Am J Phys Anthropol 136:441–454. Hawkes EW. 1916. Skeletal measurements and observations on the Point Barrow Eskimos with comparisons from other Eskimo groups. Am Anthropol 18:203–234. Hrdlicka A. 1940. Lower jaw. Further studies. Am J Phys Anthropol 27:383–467. Hylander WL. 1972. The adaptive signiﬁcance of Eskimo craniofacial morphology. Ph.D. dissertation, University of Chicago. Hylander WL. 1977. The adaptive signiﬁcance of Eskimo craniofacial morphology. In: Dahlberg AA, Graber TM, editors. Orofacial growth and development. Paris: Mouton. p 129–170. Hylander WL. 1979. The functional signiﬁcance of primate mandibular form. J Morphol 160:223–239. Hylander WL. 1984. Stress and strain in the mandibular symphysis of primates: a test of competing hypotheses. Am J Phys Anthropol 64:1–46. Hylander WL. 1988. Implications of in vivo experiments for interpretating the functional signiﬁcance of ’‘robust’’ australopithecine jaws. In: Grine FE, editor. Evolutionary history of the ‘‘robust’’ Australopithecines, 1st ed. New Brunswick, NJ: Transaction Publishers. p 55–83. Jantz RL. 1973. Microevolutionary change in Arikara crania: a multivariate analysis. Am J Phys Anthropol 38:15–26. Jantz RL, Owsley DW. 1984. Long bone growth variation among Arikara skeletal populations. Am J Phys Anthropol 63:13–20. Kanazawa E, Kasai K. 1998. Comparative study of vertical sections of the Jomon and modern Japanese mandibles. Anthropol Sci 106:107–118. Langenbach GEJ, van Eijden TMGJ. 2001. Mammalian feeding motor patterns. Integr Comp Biol 41:1338. Larsen H, Rainey FG. 1948. Ipiutak and the arctic whale hunting culture. New York: The American Museum of Natural History. Larsen CS. 1997. Bioarchaeology: interpreting behavior from the human skeleton. Cambridge: Cambridge University Press. Lieberman DE. 1996. How and why humans grow thin skulls: experimental evidence for systemic cortical robusticity. Am J Phys Anthropol 101:217–236. Lieberman DE, Krovitz GE, Yates FW, Devlin M, St Claire M. 2004. Effects of food processing on masticatory strain and craniofacial growth in a retrognathic face. J Hum Evol 46:655– 677. 627 Liversidge H, Molleson T. 2004. Variation in crown and root formation and eruption of human deciduous teeth. Am J Phys Anthropol 123:172–180. Loth SR, Henneberg M. 2001. Sexually dimorphic mandibular morphology in the ﬁrst few years of life. Am J Phys Anthropol 115:179–186. Mallegni F, Trinkaus E. 1997. A reconsideration of the Archi 1 Neanderthal mandible. J Hum Evol 33:651–668. Mayhall JT. 1976. The morphology of the permanent dentition of prehistoric and modern Central Arctic Eskimoid peoples: a study of their biological relationships. Ph.D. dissertation, University of Chicago. Mays S, Ives R, Brickley M. 2009. The effects of socioeconomic status on endochondral and appositional bone growth, and acquisition of cortical bone in children from 19th century Birmingham. England. Am J Phys Anthropol 140:410–416. Merbs CF. 1968. Anterior tooth loss in arctic populations. Southwestern J Anthropol 24:20–32. Nicholson E, Harvati K. 2006. Quantitative analysis of human mandibular shape using three-dimensional geometric morphometrics. Am J Phys Anthropol 131:368–383. O’Neill MC, Ruff CB. 2004. Estimating human long bone crosssectional geometric properties: a comparison of noninvasive methods. J Hum Evol 47:221–235. Organ JM, Ruff CB, Teaford MF, Nisbett RA. 2006. Do mandibular cross-sectional properties and dental microwear give similar dietary signals? Am J Phys Anthropol 130:501–507. Owsley DW, Jantz RL. 1985. Long bone lengths and gestational age distributions of post-contact period Arikara Indian perinatal infant skeletons. Am J Phys Anthropol 68:321–328. Paschetta C, de Azevedo S, Castillo L, Martı́nez-Abadı́as N, Hernández M, Lieberman DE, González-José R. 2010. The inﬂuence of masticatory loading on craniofacial morphology: a test case across technological transitions in the Ohio Valley. Am J Phys Anthropol 141:297–314. Perzigian AJ. 1975. Natural selection on the dentition of an Arikara population. Am J Phys Anthropol 42:63–70. Pinhasi R, Shaw P, White B, Ogden AR. 2006. Morbidity, rickets and long-bone growth in post-medieval Britain–a cross-population analysis. Ann Hum Biol 33:372–389. Pinhasi R, Teschler-Nicola M, Knaus A, Shaw P. 2005. Crosspopulation analysis of the growth of long bones and the os coxae of three early medieval Austrian populations. Am J Hum Biol 17:470–488. Plavcan JM, Daegling DJ. 2006. Interspeciﬁc and intraspeciﬁc relationships between tooth size and jaw size in primates. J Hum Evol 51:171–184. Rainey F. 1941. The Ipiutak culture at Point Hope. Alaska. Am Anthropol 43:364–375. Ravosa MJ. 1996. Mandibular form and function in North American and European Adapidae and Omomyidae. J Morphol 229:171–190. Ruff C, Holt B, Trinkaus E. 2006. Who’s afraid of the big bad Wolff? ‘‘Wolff ’s law’’ and bone functional adaptation. Am J Phys Anthropol 129:484–498. Scheuer L. 2002. A blind test of mandibular morphology for sexing mandibles in the ﬁrst few years of life. Am J Phys Anthropol 119:189–191. Scheuer L, Black S. 2000. Developmental juvenile osteology. London: Academic Press. Schutkowski H. 1993. Sex determination of infant and juvenile skeletons: I. morphognostic features. Am J Phys Anthropol 90:199–205. Scott GR, Winn JR. 2010. Dental chipping: contrasting patterns of microtrauma in Inuit and European populations. Int J Osteoarchaeol. DOI:10.002/oa. 1184. Smith BH. 1991. Standards of human tooth formation and dental age assessment. In: Kelley MA, Larsen CS, editors. Advances in dental anthropology. Chichester: Wiley-Liss. p 143–168. Smith RJ. 1983. The mandibular corpus of female primates: taxonomic, dietary and allometric correlates of interspeciﬁc variations in size and shape. Am J Phys Anthropol 61:315–330. Spencer MA, Demes B. 1993. Biomechanical analysis of masticatory system conﬁguration in Neanderthals and Intuits. Am J Phys Anthropol 91:1–20. American Journal of Physical Anthropology 628 M.A. HOLMES AND C.B. RUFF Thayer ZM, Dobson SD. 2010. Sexual dimorphism in chin shape: implications for adaptive hypotheses. Am J Phys Anthropol 143:417–425. Turner CG. 1967. The dentition of Arctic peoples. Ph.D. dissertation, University of Wisconsin. Turner CG, Cadien JD. 1969. Dental chipping in Aleuts. Eskimos and Indians. Am J Phys Anthropol 31:303–310. Ungar PS, Spencer MA. 1999. Incisor microwear, diet, and tooth use in three Amerindian populations. Am J Phys Anthropol 109:387–396. Walker MJ, Lombardi AV, Zapata J, Trinkaus E. 2010. Neandertal mandibles from the Sima de las Palomas del Cabezo Gordo. Murcia, southeastern Spain. Am J Phys Anthropol 142:261–272. American Journal of Physical Anthropology Wallace IJ, Middleton KM, Lublinsky S, Kelly SA, Judex S, Garland T Jr, Demes B. 2010. Functional signiﬁcance of genetic variation underlying limb bone diaphyseal structure. Am J Phys Anthropol 143:21–30. Wolpoff MH. 1975. Some aspects of human mandibular evolution. In: McNamara JA Jr, editor. Determinants of mandibular form and growth. Monograph No. 4, Craniofacial Growth Series. Center for Human Growth and Development. Ann Arbor: University of Michigan. p 1–64. Wood BA. 1978. Allometry and hominid studies. In: Bishop WW, editor. Geological background to fossil man. Edinburgh: Scottish Academic Press. p 125–128.