Kroeber Anthropological Society Papers, Nos. 71-72, 1990 Neandertals and the Anterior Dental Loading Hypothesis: A Biomechanical Evaluation of Bite Force Production Susan C. Anton Absolutely large anterior occlusal loads have been frequently postulated as the driving mechanism behind the evolution of the Neandertalface. Bite force production capabilities of Neandertals are estimated via mathematical models to test this anterior dental loading hypothesis. Biomechanical analysis in the lateral andfrontal projections indicates that even moderate occlusal loading would have been very costly for Neandertals in terms of condylar reaction force. In the Neandertal model, reaction force at the condyles is always greater than bite force. Absolute bite force values determinedfrom muscularforce estimates are smaller in Neandertals than in modern humans despite absolutely larger muscle force estimates. These results suggest that the typical Neandertal pattern of relatively great anterior dental attrition is not due to absolutely heavy occlusal loading but to continuous loading over time. Likewise, degenerative joint disease of the Neandertal temporomandibular joint is an expected outcome of moderate dental loading given the geometry of the Neandertal masticatory system. Emphasis on the anterior dental loading hypothesis as the driving force in the origin and evolution of Neandertal facial morphology is therefore unwarranted. INTRODUCTION The specific designation of Neandertals and their phylogenetic relationship to Homo sapiens continues to evoke heated debate. Recently, sev- eral workers have approached the question of Neandertal phylogenetic positioning via biome- chanical and functional analyses of the Neandertal face (Smith 1983; Rak 1986; Demes 1987; Trinkaus 1987). The relative continuity or dis- continuity of structure, function and ultimately niche envisioned for the two morphs (Neandertal and modern human) provides fuel for the ques- tion of phylogenetic continuity or discontuity. Smith (1983) and Trinkaus (1987) tend to see continuity while Rak (1986) envisions disconti- nuity. These studies differ on the specifics of mo- delling bending moments in the Neandertal face. However, each author agrees that these moments are in part due to heavy occlusal loads at the Neandertal anterior dentition. Both Smith (1983) and Rak (1986) propose anterior tooth loading to be the driving factor in the evolution of Nean- dertal facial morphology. This hypothesis is referred to as the Anterior Dental Loading Hypo- thesis. Smith (1983) clearly refers to absolutely large occlusal loads as opposed to continuous use over time. Rak (1986) is not as clear in his defi- nition of heavy anterior dental loading. The assumption of large anterior tooth loading in Neandertals has been based on: 1) heavily worn anterior dentition with respect to the degree of wear on the posterior teeth; and 2) consistent ap- pearance of degenerative joint disease (DJD) of the articular eminence of the temporal bone. To test the supposition of absolutely high occlusal loading in Neandertals, the probable maximum force production at the anterior teeth in Neandertals was determined via two-dimensional mathematical models. These models allow for the analysis of the external forces active at the temporomandibular joint (TMJ) and occlusal sur- faces and are frequently used to analyze modern human masticatory forces (Smith 1978; Pruim et al. 1980; Hylander 1985; Osborn and Baragar 1985). The results (possible force production of Neandertals) are then compared with the force production capabilities of extant humans. MATERIALS AND METHODS Materials Casts of cranial and mandibular remains of Amud I from Israel were used to determine cross sectional areas, orientations and moment arms for the Neandertal model (see below). Cranial re- mains from La Ferrassie, France were used in determining the cross-sectional area of the tempo- ral fossa due to damage in this area of the Amud specimen. These specimens were chosen on the basis of their relatively complete states and their reasonably close approximations to one another in size and form. Methods The mandible was modelled as a lever with its fulcrum at the condyle. Because the total mus- 68 cle force vector (Fm) is positioned posterior to the bite point, in static equilibrium both useful bite force (Fb) and condylar reaction force (Fc) are produced (Figure 1; Hylander 1985). In order to estimate the absolute Fb production in Neandertals it is therefore necessary to under- stand the relationship between Fb, Fm and Fc and to be able to estimate Fm. Muscle Force Determination Force is a vector quantity. In order to esti- mate muscle force it is therefore necessary to define both the direction and the magnitude of the force for each muscle modelled. The muscles used here (superficial masseter, medial pterygoid and temporalis) to model the external forces at the TMJ are considered power muscles (Osborn and Baragar 1985). Power muscles produce Fb. Control muscles (lateral pterygoid and parts of the temporalis) have poor moment arns for producing useful Fb. Instead, they act to stabilize the condyle. Hence, the lat- eral pterygoid muscle was not used in this model. Experimental studies indicate that the muscles modelled here are all functionally hetero- genous pinnate muscles (Moller 1966; Herring et al. 1979). Depending on its fiber angle, a pinnate muscle may produce a different contractive force than is predicted from its cross-sectional area (Gans and Bock 1965; Josephson 1975; Gans and de Vree 1987). However, due to the complex nature of the pinnation, the inherent dif- ficulties of determining fiber angle from fossil remains, and, most importantly, the lack of deter- mination of fiber angle of the masseter, medial pterygoid and temporalis muscles in modern humans, these muscles have been modelled as simple parallel-fibered muscles. If parallel- Figure 1. Lateral projection analysis. Fc Fb Schematic representation of lateral projection of extemal forces acting on the human mandible during biting. Fc equals condylar reaction force. Fm equals total muscle resultant force (summed left and right muscle forces of the medial pterygoid, temporalis and masseter). Fb equals bite force. X equals distance between Fb and Fm. Y equals Fm moment arm. Z equals Fb moment arm. In conditions of static equilibrium these forces are related in the following manner: (Fb) (z) Fm = y (Fm) (x) Fc= or y (Fb) (x) Fc = z Twisting and bending moments due to the more lateral placement of Fm relative to Fb are not considered in determination of Fb. Therefore, Fb is likely to be overestimated. (Redrawn from Hylander 1975). 69 fibered muscles are assumed, muscular orienta- tions and force vector directions can be inferred from bony morphology (Klaauw 1963; Gans and de Vree 1987). Muscle Force Vector Direction Muscle force vector direction was modelled as a single vector for each of the three muscles. This vector was positioned as the central fiber (essentially the centroid; Hiiemae 1971) within the body of the muscle. Vector direction was then determined by joining the areas of origin and insertion by this central line (Figure 2). The greater surface area under the temporal line and the further posterior extension of this line indicates a larger posterior temporalis component in Neandertals than in modern humans. The tem- poral fossa is similarly elongated posteriorly, Figure 2. Estimated Nea Fm Fpt . Fm providing an increased advantage to some of these posterior fibers over the condition in mo- dern humans. For these reasons the resultant temporalis muscle vector direction has been placed slightly more obliquely than in modern hu- mans (Figure 2). This placement is not critical to the determination of maximum occlusal and con- dylar reaction force magnitudes. The direction would, however, significantly affect the direction of the condylar reaction force vector (Throckmor- ton 1985). The resultant vectors of the masseter and medial pterygoid muscles are quite similar to those of modern humans. Reconstruction of Muscle Force Vector Magnitude Assuming simple muscle architecture, the maximum force a muscle can exert is equal to the ndertal muscle force vectors. Vectors represent combined right and left muscle forces for the masseter (Fma), medial pterygoid (Fpt), and temporalis ;, (Ft). Fm equals the sum of Fma, Fpt and Ft. 1 mm = 20 N and Fm = 2010 N. (Neandertal silhouette after Trinkaus 1983). 70 physiological cross-sectional area of that muscle (i.e., an estimate of the number of muscle fibers firing in unison) multiplied by the stress in the muscle (Weijs 1980; Dul et al. 1984). Muscle cross-sectional areas (Table 1) were determined from bony insertions as follows: 1) the area en- closed within the temporal fossa was used as an estimate for temporalis cross-sectional area; 2) the triangular area formed by the mandibular corpus and ramus at the goneal angle and the mylohyoid groove was used as an estimate for medial pterygoid cross-sectional area; 3) the pro- duct of the length of the masseteric origin on the zygomatic arch and the distance between the lat- eral edge of the zygomatic arch and the lateral edge of the mandibular ramus was used to es- timate cross-sectional area for the masseter. Cross-sectional areas for these muscles in Nean- dertals are given in Table 1. These same measurements were taken on a modem human skull and compared to physiologi- cal cross sections for these muscles (Table 2). Correction factors were determined by assuming a linear relationship between physiological cross- sectional area and bony cross-sectional area. The correction factor was greatest for the temporalis. The cross-sectional area of this muscle was over- estimated by a value of 6. The medial pterygoid was overestimated by the bony cross section by a factor of 1.5. The masseter value was not correc- ted. The Neandertal cross-sectional areas were corrected using the same correction factors. The force and cross-sectional areas reported by Schumacher (1961) for modem humans were used to calculate the stress in each muscle (Tables 1 and 2). Given the close phylogenetic relation- ship of Neanderals and modem humans, stress was assumed to be the same in both groups. Individual muscle force magnitude was cal- culated using the product of the corrected cross- sectional areas for Neandertal muscles and the modem human muscle stresses. Empirically de- rived modem human muscle forces are given in Table 3. Position, direction and magnitude of the combined muscle force (Fm) was detennined by simple vector analysis after projecting all vectors onto the same plane (Figure 2). Biomechanical Analysis Extemal forces at the TMJ were analyzed by determining moments about the mandibular condyle, assuming static equilibrium, in lateral projection with the jaw in closed position and with a fixed center of rotation. Simple lateral projection analysis is adequate only when the for- ces on the two halves of the mandible are equal (Hylander 1975, 1985; Smith 1978). Such con- ditions are met during bilateral biting. During unilateral biting, the worldng (biting) side and the balancing side condylar reactions are not equal. In order to completely analyze unilateral biting, an analysis in the frontal projection was also per- formed (see Hylander 1985). In both analyses, bending and twisting moments produced by the positioning of the Fm lateral to the Fb were not considered. As such, Fb is likely to be overesti- mated, as all components of Fm were considered to produce useful Fb. Lateral projection analyses derive two of the three variables (Fm, Fb and Fc) from the third (known) variable. In the model used here, Nean- dertal Fb is determined from Fm (see below and Table 1. Estimated Neandertal muscle cross-sectional areas and forces (for one side). Uncorrected Cross section 2 (cm2) Muscleb Corrected Cross section (cm2) Human Stressa (kgm ls2 ) T 3.2 5.5 8.4 X 105 4.6 x 102 Ma 3.7 3.7 8.4 x 105 3.1 x 102 MPt 5.5 3.8 9.3 x 105 3.5 x 102 a Human muscle stress detemined fnro Schunachex (1961). b T = Temporalis, Ma = Masseter, MPt = Medial PterygoidL Force (N) 71 Table 4). For the system to be in static equili- brium, the summation of the moments around any point is equal to zero (Figure 1; Hylander 1975; Smith 1978). That is: (Fb) (z) = (Fm) (y) where: Fm = muscle force Fb = bite force y = muscle moment arm (Fb) (z) = bite force moment arm From these conditions it follows that: (Fm) (y) Fb = z and from analyzing moments about Fm it follows that: (Fb) (x) Fc = y where: Fc = condylar reaction force x = the distance between Fb and Fm Fm and Fc represent the sum of left and right muscular and condylar reaction forces, re- spectively. Fb is measured perpendicular to the occlusal plane. Forces were calculated for both molar (first molar, Ml) and incisal (lateral in- cisor, 12) biting and the forces were considered point loads (Table 4). To analyze unilateral biting, a frontal projec- tion analysis followed the lateral projection analy- sis for both molar and incisal biting. In such an analysis (Table 5): Fc=Cw+Cb Fm = Fmw+ Fmb where Cw and Cb are Fc on the working and balancing sides, respectively, and Fmw and Fmb are Fm on the working and balancing sides, re- spectively. The Fm resultant was positioned as if the working and balancing side musculature were equally active. The mandible was analyzed as a stationary beam with a point load applied to it (Figure 3). Given conditions of static equilibrium it follows that (Fm) (d) - (Fb) (z) Cw = w Cb= (Fm) (w-d) - (Fb) (w-z) w where: w = bicondylar width d = distance from Fm to Cw z = bite force moment arm RESULTS Muscle Force Determination Neandertal muscle force vector direction is shown in Figure 2. Corrected muscle cross-sec- tional areas for Neandertals are slightly larger Table 2. Modem human muscle bony and physiological cross-sectional areas. Bony Cross Section (cm2) Physiologic a Cross Section 2 (cm) Correction Factor T 2.4 x 101 4.2 6 Ma 3.4 3.4 ---- MPt 2.8 1.9 1.5 Muscleb a Physiological cross-sectional areas taken from Schumacher (1961). b Muscles abbreviated as in Table 1. 72 than the values determined by Schumacher (1961) for modem human muscles (Tables 1 and 2). The largest difference in cross-sectional area is in the medial pterygoid, which is twice as large in the Neandertal estimates as in modem humans. Consequendy, muscle force magnitude estimates are slightly larger for Neandertal masseter and temporalis muscles and two times as large for the medial pterygoid estimates than those of Schu- macher (1961). Several authors have calculated individual muscle forces in modern humans from Fm and physiological cross-sectional area. In all cases the combined force of the masseter and medial pterygoid was slightly greater than that of the temporalis (Carlsoo 1952; Schumacher 1961; Pruim et al. 1980). This is also the case with the Neandertal results. Muscle force magnitude estimates calculated here are comparable to those of Pruim et al. (1980) and lower than those of Carlsoo (1952). Pruim et al. (1980) explained the difference between their force estimates and those of Schu- macher (1961) and others as being due to the fact that Pruim et al. (1980) did not take into account soft tissue inhibitions to force production. This was also the case with this study. Carlsoo's (1952) larger force determinations were due largely to his use of a higher muscle stress value (1.1 x 106 N/m2) than used here. Table 3. Modern human muscle forces (in Newtons). Temporalis Masseter Medial Pterygoid Pruim et al. 5.6 x 102 6.4 x 102 1980 Carlsoo 1952 8.2 x 102 6.1 x 102 3.0 x 102 Schumacher 3.6 x 102 2.9 x 102 1.8 x 102 1961 Table 4. External forces (in Newtons) at the TMJ in molar and incisal biting determined in the lateral projection. (See text for abbreviations). Molar Biting Fb Fm Fc Modern Human 2 3 (Pruim et al. 9.6 x 10 1.6 x 10 7.1 x 102 1980) Neandertal 7.8 x 102 2.0 x 103 1.2 x 103 Incisal Biting Fb Fm Fc Modern Human (Pruim et al. 7.0 x 102 1.6 x 103 9.9 x 102 1980) Neandertal 5.5 x 102 2.0 x 10 1.5 x 103 73 Biomechanical Analysis Lateral Projection Analysis Bite force production at both molar and inci- sal bite points is absolutely smaller in Neandertals than in modern humans (Table 4). This is despite the more than 15% increase in muscle force pro- duction capability in Neandertals as compared to modern humans. At I2 the values are 550 New- tons (N) for Neandertals and 700 N in modern humans. The Fb values at Ml are 780 N in Neandertals and 960 N in modern humans. However, the relative proportion of Fb at I2 relative to Fb at Ml is nearly identical (approxi- mately 70%; Table 5). Condylar reaction force is substantially greater at both molar and incisal bite points than bite force in the Neandertal model. At Ml the Fc is 158% of Fb and at I2 the Fc is 264% of Fb (Table 5). In contrast, Fc in modern humans is 74% of Fb at Ml and 145% at I2. The absolute values of Fc are 30% to 40% greater than those for modern humans (Table 4). In Neandertals, bite force is only 39% and 28% of Fm at Ml and I2, respectively, while it is 57% and 42% of Fm in modern humans (Table 5). Likewise, Fc is 61% and 72% of Fm at Ml and I2, respectively, in Neandertals. Fc is only 42% and 59% of Fm in modern humans. Thus, Neandertals are producing much less useful Fb per Fm than modern humans at a much greater expense to the TMJ. Frontal Analysis Frontal projection analysis showed that in Neandertals the balancing side condyle is more Figure 3. Frontal projections analysis. Cw Schematic representation of external forces acting on the mandible in the frontal projection. Fm and Fb are defined as in Figure 1. Cw and Cb are Fc on the working and balancing sides, respectively. W equals becondylar width. Z equals Fb moment arm and d equals Fm moment arm (distance to working side condyle). In static equilibrium: (Fm) (d) - (Fb) (z) Cw = w (Fm) (w-d) - (Fb) (w-z) Cb= w Cb is greater than Cw. (Redrawn from Hylander 1985). 74 heavily stressed than the working side condyle (Table 6). This is also true of modem humans (Smith 1978) and is corroborated by clinical evidence in which patients with diseased TMJs chew on the diseased side (Hylander 1975). The greater Fc in Neandertals than in modern humans is reflected in the greater percentage of Fb rep- resented by Fc (Table 6). In molar biting in modem humans, Fc is 62% of Fb at the working side condyle and 92% of Fb at the balancing side condyle. During incisal biting Fc is 87% of Fb on the working side and 171% of Fb at the balancing side. Thus, incisal biting is extremely costly. DISCUSSION Neandertal facial prognathism has been sug- gested to be the result of the rearrangement of the infraorbital plates and the anterior migation of the tooth row with respect to the mandibular ra- mus (Rak 1986). Masticatory muscular relations with respect to the TMJ are not greatly altered by this facial arrangement. However, the F b moment arms are elongated resulting in the pro- duction of a less useful bite force and a very large condylar reaction force. The consistent appearance of DJD in Nean- dertals is most likely due to the proportionally larger Fc produced at the TMJ. Even moderate occlusal loading at either the molar or incisal regions inflicts large reaction forces at the condyles leading, over time, to this degeneration. The physiological restrictions imposed by the production of large Fc make it unlikely that the attrition of the anterior dentition is related to absolutely greater occlusal loading. This absence of large anterior occlusal loads is corroborated by the presences of only minor trauma (enamel mi- crofracture and flaking) in Neandertal anterior teeth (Smith 1983). This conclusion contradicts Smith's (1983) idea that the small amount of trauma was due to absolutely larger teeth which were better able to bear greater loading. Heavy anterior dental attition may more Table 5. Relationships between masticatory forces for modem humans and Neandertals. Modern Humans 73% Neandertals 71% Molar Incisal Molar Incisal Fc 74% 145% 158% 264% Fb Fb Fm 57% 42% 39% 28% Fc Fm 42% 59% 61% 72% Table 6. Working and balancing side condylar reaction forces as a percen- tage of useful bite force. (Modem human data from Smith 1978). Work Molar Balance Incisal Work Balance Modern Human 15% 63% 71% 71% Neandertal 62% 92% 87% 171% Fb @ I2 Fb Q Ml 75 likely be due to repetitive usage of the anterior dentition in food preparation or other cultural be- haviors. Similar wear patterns are observable in prehistoric Californian groups which have heavy anterior dentition usage (personal observation). These patterns exist without the elongated facial geometry typical of the Neandertal face. Such paramasticatory behavior is also suggested by microwear studies of the Neandertal anterior den- tition which show labial wear striae indicating the use of the anterior dentition in a vice-like grip (Trinkaus 1983). The assumption that absolutely large anterior occlusal loads contribute to bending moments of the Neandertal face cannot be supported. How- ever, bending moments in the sagittal plane due to increased Fb moment arms are greater than those in modern humans. Additionally, forces generated by continuous use of the anterior denti- tion over time may be a factor in the evolution of facial morphology (see Hylander 1979). The unique facial morphology of the Nean- dertals certainly affected their masticatory force production capabilities. However, the supposi- tion that masticatory forces were the driving forces in the evolution of facial morphology begs the question of the origin of the morphology (see Rak 1986). Force production capabilities of Neandertals are disadvantageous compared to those of less prognathic hominids. Facial elonga- tion (with or without the rearrangement of the infraorbital plates typical of the Neandertal face) is not an effective method of counter-balancing high anterior dental loading and sagittal bending moments when compared to forces incurred by less prognathic hominids. Therefore, it seems unlikely that facial prognathism would have de- veloped as a direct result of heavy anterior dental loading unless the prognathic position of the face were advantageous for some separate reason. CONCLUSIONS The geometry of the Neandertal masticatory system makes the production of even moderate occlusal loads very costly in terms of condylar reaction forces. It is likely that the level of these reaction forces may provide an upper limit to the level of Fb production possible. Estimates of Fb production capabilities for Neandertals are well below modern human capa- bilities. This is true despite greater muscle force production capabilities for the Neandertals. In sum, the attrition of the Neandertal an- terior dentition is not likely to be related to absolutely greater occlusal loads but to consistent usage of the anterior dentition through time, use which might have entailed paramasticatory beha- viors such as food, tool or leather preparation. Thus, the central role which the heavy anterior dental loading hypothesis has been given in the origin and evolution of the Neandertal face ap- pears unwarranted. ACKNOWLEDGMENTS F.C. Howell and the Laboratory for Human Evolutionary Studies graciously provided access to the Neandertal cast material. M. Koehl, D. Pentcheff, S. Swartz and S. Worchester provided invaluable advice concerning the methods and direction of this analysis. G.D. Richards, K. Dustin and J. Hyland provided insightful com- ments concerning the structure and content of the manuscript. REFERENCES CITED Carlsoo, Sven (1952) Nervous coordination and mechanical function of the mandibular eleva- tors. Acta Odontologica Scandinavica. Suppl. 11:1-129 Demes, Brigitte (1987) Another look at an old face: Biomechanics of the Neandertal facial skeleton reconsidered. Journal of Human Evolution 16:297-304. Dul, J.G., E. Johnson, R. Shiavi and M.A. Townsend (1984) Muscular synergism - II. A minimum-fatigue criterion for load sharing between synergistic muscles. Journal of Bio- mechanics 9:675-684. Gans, Carl and William J. Bock (1965) IV. The functional significance of muscle archi- tecture: A theoretical analysis. Ergebnisse Anatomie Entwicklungsgeschichte 38:115- 142. Gans, Carl and F. de Vree (1987) Functional bases of fiber length and angulation in muscle. Journal of Morphology 192:63-85. Herring, Susan, A.F. Grimm and B.R. Grimm (1979) Functional heterogeneity in a multipin- nate muscle. American Journal of Anatomy 154:563-576. Hiiemae, Karen M. (1971) The structure and function of the jaw. muscles in the rat (Ratus norvegicus L.) III. The mechanics of the muscles. Zoological Journal of the Linnaean Society 50:111-132. Hylander, William L. (1975) The human man- dible: Lever or link? American Journal of Physical Anthropology 43:227-242. Hylander, William L. (1979) Functional anat- omy. In B.G. Sarnat and D.M. Laskdn (eds.), 76 The Temporomandibular Joint: A Biological Basis for Clinical Practice. Illinois: C.C. Tho- mas. Pp.85-113. Hylander, William L. (1985) Mandibular func- tion and temporomandibular joint loading. In D.S. Carlson, J.A. McNamara and K.A. Rib- bens (eds.), Developmental Aspects of Temporomandibular Joint Disorders. Ann Ar- bor: University of Michigan Press. Pp. 19-35. Josephson, Robert K. (1975) Extensive and in- tensive factors determining the performance of striate muscle. Journal of Experimental Zool- ogy 194:135-154. Klaauw, C.J. van der (1963) Projections, deep- enings and undulations of the surface of the skull in relation to the attachment of muscles. Verhandelingen der Koninklijke Nederlandse Akademie van Wetenschappen, Afd. Natuur- kunde 55:1-270. Moller, Eigild (1966) The chewing apparatus: An electromyographic study of the action of the muscles of mastication and their correlation to facial morphology. Acta Physiologica Scandinavica. 69 supplementum 280:1-229. Osborn, J.W. and F.A. Baragar (1985) Pre- dicted pattern of human muscle activity during clenching derived from a computer assisted model: Symmetric vertical bite forces. Jour- nal ofBiomechanics 18:599-612. Pruim, G.J., H.J. de Jongh and J.J. ten Bosch (1980) Forces acting on the mandible during bilateral static bite at different bite force levels. Journal of Biomechanics 13:755-763. Rak, Yoel (1986) The Neandertal face: A new look at an old face. Journal of Human Evolu- tion 15:151-164. Schumacher, Gort H. (1961) Funktionelle Mor- phologie der Kaumuskulatur. Jena: Gustav Fischer Verlag. Smith, Fred H. (1983) Behavioral interpretation of changes in craniofacial morphology across the archaic/modern Homo sapiens transition. In Erik Trinkaus (ed.), The Mousterian Leg- acy: Human Biocultural Change in the Upper Pleistocene. British Archaeological Reports International Series 164:141-163. Smith, Richard J. (1978) Mandibular biomecha- nics and temporomandibular joint function in primates. American Journal of Physical Anthropology 49:341-350. Throckmorton, Gaylord S. (1985) Quantitative calculations of temporomandibular joint reaction forces - II. The importance of the di- rection of the jaw muscle forces. Journal of Biomechanics 18:451-461. Trinkaus, Erik (1983) The Shanidar Neander- tals. New York: Academic Press. Trinkaus, Erik (1987) The Neandertal face: Perspectives on a recent hominid face. Journal of Human Evolution 16:429-443. Weijs, Wilhelm (1980) Biomechanical models and the analysis of form: A study of the mammalian masticatory apparatus. American Zoologist 20:707-719.