THE ANATOMICAL RECORD 300:152 159 (2017) Opposing Extremes of Zygomatic Bone Morphology: Australopithecus Boisei versus Homo Neanderthalensis YOEL RAK* AND ASSAF MAROM Department of Anatomy and Anthropology, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, 6997801, Israel ABSTRACT The lateral margin of the zygomatic bone of Australopithecus boisei flares both anteriorly and laterally. As a result, the bone loses the suspensory bracing of the facial frame and is transformed into a visor-like structure that supports itself and gains its rigidity from its shape. The coronally oriented bony plates and the outline of the facial mask help the A. boisei face resist the effect of the visor-like structure, which tends to pull the bone plates of the face away from the midline. On the other hand, the nearly sagittal orientation of the zygomatic bone in Homo neanderthalensis helps the face resist torque and bending forces, which themselves stem from the positioning of the bite point on the anterior teeth. Although the zygomatic bones of these two taxa are highly specialized, they differ fundamentally from each other. Anat Rec, 300:152 159, 2017. VC 2016 Wiley Periodicals, Inc. Key words: zygoma; zygomatic bone; function; evolution; Australopithecus boisei; Homo neanderthalensis For goodness sake, what more can be said about the zygoma? Anonymous INTRODUCTION The zygomatic bone in modern humans is identical to that of many other primates in both its geometry and its spatial position in the skull. Thus, the anatomy of the human zygomatic bone constitutes a faithful representative of the generalized anatomy. The facial surface of the body of the bone is on the coronal plane, and the bone s superior margin (forming part of the orbital rim) is horizontal and parallel to its inferior margin, which forms part of the zygomaticomaxillary crest. The horizontal orientation of the bone s inferior margin joins with the more vertical and medial portion of the crest (which is part of the maxilla), lending the crest its concave, sometimes even indented, appearance and thus forming the so-called inframalar notch. Extending posteriorly in both a horizontal orientation and a sagittal orientation, the temporal process of the zygomatic bone meets the zygomatic process of the temporal bone to form the zygomatic arch. The sagittal orientation of the temporal process implies that it merges with the body of the zygomatic bone in an approximately right angle (on a horizontal plane). Indeed, the junction between the two can be perceived as a vertical corner (represented as a hinge in Fig. 1) that is sometimes sharply defined and sometimes less defined. This corner, here termed the zygomatic corner, constitutes part of the lateral border of the facial mask, 1 marking the transition between the forward-facing, coronal part of the face and the laterally facing sagittal part, primarily the *Correspondence to: Yoel Rak, Department of Anatomy and Anthropology, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv 6997801, Israel. Fax: 1972 3 640 8287. E-mail: yoelrak@ post.tau.ac.il Received 18 April 2016; Revised 11 July 2016; Accepted 25 July 2016. DOI 10.1002/ar.23491 Published online in Wiley Online Library (wileyonlinelibrary. com). 1 The facial mask is the coronally oriented part of the face excluding the snout, which projects anteriorly from the center of the mask. VC 2016 WILEY PERIODICALS, INC.
OPPOSING EXTREMES OF ZYGOMATIC MORPHOLOGY 153 Fig. 1. A schematic representation of the generalized primate face. zygomatic arch (Fig. 1). Positioned vertically on the zygomatic corner, the frontal process of the bone (the lateral margin of the orbit) extends the peripheral border of the face upward and joins the horizontal supraorbital element of the frontal bone. In this manner, the typical superolateral right angle of the facial mask (in a frontal view) is formed. The presence of this right angle, along with a right angle on the facial mask s inferolateral end (an angle formed primarily by the horizontal inferior margin of the body of the zygomatic bone and the vertical orientation of the root of the zygomatic), lends the generalized facial mask a squarish outline. The laterally facing surface of the upright frontal process helps define the straight lateral border of the face. In the generalized face, the inferior margin of the infraorbital bone region, on which the major part of the masseter muscle is suspended, lies on a single coronal plane. As shown in Figure 1, this plane intersects M2, the bite point, producing a configuration that clearly has great biomechanical implications. The degree to which the snout and the more anterior teeth that it carries extend beyond that coronal plane varies from one primate to another. Baboons are an extreme example, at one end of the spectrum of snout protrusion beyond the
154 RAK AND MAROM facial mask, and modern humans lie at the other end. Nevertheless, except for the snout s protrusion, both of these taxa have what we have defined as a generalized face. 2 The shape and position of the generalized zygomatic bone in modern humans help us comprehend the extent of the bone s specialization in two hominids that represent opposing extremes: Australopithecus boisei and Homo neanderthalensis. The differences between the shape and position of their zygomatic bones are only examples of some of the farreaching modifications throughout the facial skeleton, modifications that are necessary for addressing specific demands imposed by two highly specialized masticatory systems. Indeed, each modification constitutes an upheaval in the basic architecture of the generalized face because of different loading (masticatory) regimes. The Shape and Position of the Zygomatic Bone in Australopithecus boisei The gradual anterior migration of the masseter muscles, as inferred from the position of the inferior margin of the zygomatic bone, is part of an effort to provide the mandibular lever with additional force and is characteristic of all the so-called robust australopiths. The morphocline and its polarity are clear: in A. africanus, the advancement of the masseter is detectable; in A. robustus, the muscle has moved farther; and in A. boisei, the migration has reached the maximum, exhibiting the most extreme character state (Rak, 1983). In the latter taxon, the inferior margin of the zygomatic bone (the part bearing most of the fibers of the immense masseter) moves so far beyond the coronal facial frame that the bone loses its suspensory bracing on the supraorbital element. That is, the zygomatic can no longer be suspended from the facial frame or depend on the frame s rigidity to oppose the masseteric pull. 3 The biomechanical solution to this problem is realized by the transformation of the zygomatic bone into a self-supporting structure. Apparently, A. boisei chose to transform the zygomatic into a visor-like structure that is not suspended from but rather rests on. From its archlike shape, the structure gains most of its rigidity and its ability to resist masseteric pull (Fig. 2). Furthermore, A. boisei abandoned the basic generalized architecture of the face for another reason related to the aforementioned modification of the zygomatic bone: the rotation of the visor-like structure around its points of support as a result of the powerful masseter s contraction. This swinging has a double effect, in that it not only moves the superior margin of the visor away from the midline but also forces part of the visor downward (Fig. 3) (Rak, 1983). For the generalized facial frame to efficiently handle the diagonal inferolateral pull, the shape and 2 The snout s protrusion is a specialized character in baboons as well as in modern humans, because they both deviate from the common primate pattern. 3 The masseteric origin (the inferior margin of the zygomatic bone) deviates laterally as well. This deviation is caused by the extremely large temporalis muscle, which dramatically increases the size of the temporal foramen through both the medial sinking of the muscle into the lateral wall of the braincase and the lateral deviation (displacement) of the zygomatic arch. orientation of the frame s entire superolateral corner, including the frontal process of the zygomatic bone, must undergo a transformation. As seen in Figure 3, the frontal process and the supraorbital element form a straight, diagonal line that is continuous with the lateral margin of the visor. This radical reorganization changes the superolateral corner into a kind of cord (in a frontal view) that is more efficient than the generalized arrangement in opposing the tendency of the A. boisei zygomatic bone (the visor) to swing under the influence of the contracting masseter. Even the shape of the orbital rim is sometimes affected by this reorganization (Fig. 3). The primarily coronal orientation of the frontal process in A. boisei and the fanning out of the process s base (deviating heavily from the vertical) also contribute to the process s effort in opposing the visor s diagonal pull. Support for our assertion that much of the anatomy of the A. boisei zygomatic bone stems from its deviation from the basic facial frame can be drawn from a clear case of parallelism with an unrelated primate, the fossil Theropithecus brumpti (Rak, 1983). There, too, the origin of the masseter muscle deviates anteriorly from the facial frame and indeed, the bone s generalized morphology is transformed into a visor (Fig. 4). Accordingly, the base of the zygomatic s frontal process faces forward and broadens as it deviates laterally. The degree of visorization in A. boisei is apparently age and sex dependent. In the female A. boisei KNM-ER 732, a visor is barely noticeable, whereas the frequently overlooked KNM-ER 733, a massive male specimen, exhibits a full-fledged visor. 4 In addition to these observations, support for age- and sex-related variation in the appearance of the visor comes from the much larger sample of available T. brumpti zygomatic bones (some of which are disarticulated). 5 The Shape and Position of the Zygomatic Bone in Homo neanderthalensis As in A. boisei, the zygomatic bone of Homo neanderthalensis deviates radically from that of the generalized pattern. However, the two taxa diverge from that pattern in opposing ways, apparently for reasons related to the fundamental disparity between the loading regimes of their basic facial architectural frames. In essence, what the Neandertal zygomatic bone exhibits is an alignment of the surfaces of all its components on a single plane. This plane is positioned close to the sagittal plane (Figs. 5 and 6). Hence, the zygomatic corner of the generalized face is obliterated. The surface of the zygomatic bone aligns (becomes continuous) with the surface of the temporal process, which has been 4 Very little of specimen KNM-ER 733 survives. Although the zygomatic bone is isolated (floating), there is no doubt that the visor in this specimen extends forward almost horizontally from the facial frame. Furthermore, apropos variation, the temporal lines in this large mature male do not meet to form a crest. 5 The large T. brumpti collection was available in the mid-1970s in the laboratory of Professor F. Clark Howell at the University of California, Berkeley. Subsequently, the entire collection was returned to Ethiopia and is now housed in the National Museum of Ethiopia.
OPPOSING EXTREMES OF ZYGOMATIC MORPHOLOGY 155 Fig. 2. The transformation of the zygomatic bone into a self-supporting visor-like structure. oriented on the sagittal plane from the outset (that is, in the generalized face); and the medial part of the zygomatic s facial surface shifts laterally. Continuing in this parasagittal orientation, the more medial part of the face (the zygomatic process of the maxilla) aligns itself on the same plane as the zygomatic surface, hence the Fig. 3. Schematic diagrams showing a visor-like zygomatic bone superimposed on a generalized face (left); the visor s swing (under the masseter s contraction) and its effect on the upper part of a generalized face (middle); the necessary adjustment seen in the specialized upper face of A. boisei (right).
156 RAK AND MAROM Fig. 4. The visor-like zygomatic bone in Theropithecus brumpti. formation of a single flat surface extending from the zygomatic arch to the lateral margin of the nasal opening. This topographical reorganization eliminates not only the zygomatic corner of the facial mask but also the canine fossa two iconic structures of the generalized face. Much of the unique morphology of the nasal bridge (the upper part of the midface) in H. neanderthalensis is but a by-product of this reorganization. In 1986, one of us (YR) suggested that the reorganization of the facial bones on the parasagittal plane in the Neandertals is a modification that is necessary for enabling the masticatory system to cope with the infraorbital bone plate s torque. This torque is produced when a hominid with a generalized face chooses to position the bite point on the anterior teeth, resulting in a separation between the coronal planes of the bite point and the facial frame, including the masseter. For reasons unknown to us, Neandertals place the bite point on the anterior part of the dental arcade, far from the facial frame (the coronal plane of the orbital rims). The positioning of the bony plate on the parasagittal plane minimizes the torque, replacing it, to a large extent, with bending. Furthermore, this new orientation of the bony plate, along with the increase in the infraorbital region s vertical height (produced by the elimination of the primitive inframalar notch and the elevation of the nasal apophysis), assists the modified face in resisting the bending on the sagittal plane (Fig. 7). We may note that much more is going on in the Neandertal face, beyond the scope of this paper. Many years after the original claim by Rak, additional observations suggested that the Neandertal masticatory system is also capable of producing a very large gape, which apparently is a feature of great importance to this taxon (Rak and Hylander, 2003, 2007, 2008a, 2009, 2014). The large gape is achieved through a combination of three conceivable strategies: the descent of the mandibular condyle to a point close to the occlusal plane (thus modifying the morphology of the upper part of the ascending ramus), the advancement of the dental arcade (leading to the formation of the retromolar space), and the retention of the masseter in its original position (permitting a larger gape than if a masseter with the same fiber length were to move forward). The large gape is clearly preferable to the Neandertal than the obvious biomechanical advantages of retaining the posterior position of the dental arcade (or
OPPOSING EXTREMES OF ZYGOMATIC MORPHOLOGY 157 Fig. 5. A schematic comparison of the facial mask topography in H. neanderthalensis (upper) and the generalized face (lower). CF, canine fossa; Zy, inferolateral angle formed on the zygomatic bone; Ap, nasal apophysis. The unfolded hinge in H. neanderthalensis represents the obliteration of the vertical facial corner.
158 RAK AND MAROM Fig. 6. A schematic comparison of paratransversal sections through the midface (at the level of porion Po and the mid-height of the nasal opening) in H. sapiens (gray) and H. neanderthalensis (black). CF, canine fossa; Zy, inferolateral angle formed on the zygomatic bone. The horizontal line represents the position of the supraorbital element, and the elongated oval structures, the orbital rims. The two jagged lines represent the sutures between which lies the zygomatic bone. Fig. 7. Schematic diagram showing the outcome of the occlusal load on the anterior teeth in Neandertals (left) and the generalized face (right).
OPPOSING EXTREMES OF ZYGOMATIC MORPHOLOGY 159 even pulling the arcade back), moving the masseter muscle closer to the bite point, or both. The latter two strategies are the preference of the robust australopiths, as mentioned earlier, and the determining factor of their facial topography (Rak, 1983) and modest gape (Rak and Hylander, 2008b). The questions of why the large gape is so important to Neandertals and why they are willing to pay such a heavy biomechanical price are yet to be answered. Among the researchers who recognized the uniqueness of the zygomatic bone in Neandertals and its effect on the Neandertal face, Sergio Sergi (1942) and Howells (1975) stand out. Both try to quantify the bone s peculiarity and compare it to the zygomatic of the generalized H. sapiens face. Sergi expresses this uniqueness in terms of the area of a triangle that is formed by the three processes of the zygomatic bone. In a lateral view, the triangle s area is much greater in Neandertals than in the generalized face. (This difference stems from the unfolding of the Neandertal bone surface into one plane that is perpendicular to the eye of the observer; see also Rak, 1991.) Howells points out that in a side view of the generalized face (H. sapiens), the difference between the length of the line segment from porion to the upper end of the zygomaticomaxillary suture and the length of the segment from porion to the lower end of this suture is smaller than in Neandertals. Here, too, the difference between Neandertals and H. sapiens stems from the more sagittally oriented bone surface of the zygoma (that is, the elimination of the zygomatic corner) in Neandertals. Although Howells and Sergi are aware of the differing morphology portrayed in their own drawings they focus solely on the taxonomic issue, without relating it to the functional benefits of the bony plates position close to the sagittal plane. Phylogenetically speaking, both of these scholars were right. One of us wrote with admiration: I wholeheartedly agree with Howells (1975) that the facial morphology of Homo specimens preceding the classic Neanderthal is more similar in these respects to the morphology of those following it than either is to the Neanderthal itself (Rak, 1986, p. 163). In the last few years, empirical evidence has accumulated in support of the biomechanical inferences presented here regarding the facial architecture of A. boisei and H. neanderthalensis. This evidence emanates primarily from finite element (FE) studies of these (and other hominid) facial morphologies. Strait et al. (2010) confirmed the hypothesis set forth by Rak (1983) that the derived facial features in A. africanus were adaptations that provide a structural reinforcement to the face against loads imposed by premolar biting. Rak s assertion vis-a-vis A. boisei was corroborated by Smith et al. (2015), who demonstrated that differences in strain distribution observed between FE models of A. boisei (OH 5) and Pan troglodytes are consistent with the hypothesis that the position of the zygomatic root is an important factor influencing load resistance in the facial skeleton. The effect of infraorbital plate position in the generalized vs. the derived face of H. neanderthalensis was addressed by Marom (2013), who generated an FE model of P. troglodytes and transformed its midface into a Neandertal version by changing the position of the infraorbital plate from frontal to sagittal. Anterior biting task simulations in the two models resulted in significant differences between strain patterns, validating Rak s prediction that when one controls for muscle force and bite point location (anterior, in this case), stresses in the modified face are generally lower than in the primitive one. To summarize, the coronally oriented bony plates in A. boisei and the outline of its facial mask help the face resist the effect of the zygomatic bone s modification into a visor, which tends to pull the bone plates away from the midline. On the other hand, the modified Neandertal zygomatic bone, which is oriented on the sagittal plane, helps the face resist torque and bending, which themselves are brought about by the positioning of the bite point on the anterior teeth. Indeed, these two specialized zygomatic bones differ fundamentally from each other. LITERATURE CITED Howells WW. 1975. Neanderthal man: Facts and figures. In: Tuttle RH, editor. Paleoanthropology: morphology and paleoecology. Paris: Mouton. p 389 407. Marom A. 2013. Functional implications of the unique Neandertal facial skeleton. PhD Thesis. Tel Aviv University. Rak Y. 1983. The australopithecine face. New York: Academic Press. Rak Y. 1986. The Neanderthal: A new look at an old face. J Hum Evol 15:151 164. Rak Y. 1991. Sergio Sergi s method and its bearing on the zygomatic bone position in the Neandertal face. In: Piperno M, Scichilone G, editors. The Circeo I Neandertal skull studies and documentation. Rome: Instituto Poligrafico e Zecca Dello Stato. p 301 310. Rak Y, Hylander WL. 2003. Neandertal facial morphology and increased jaw gape [abstract]. Am J Phys Anthropol 120:174. Rak Y, Hylander WL. 2007. The functional significance of the retromolar space in the Neandertal mandible [abstract]. Abstracts of the PaleoAnthropology Society 2007 Meetings. PaleoAnthropology 2007:22. Rak Y, Hylander WL. 2008a. Anatomical correlates for increased gape in the Neandertal face [abstract]. Abstracts of the Paleo- Anthropology Society 2008 Meetings. PaleoAnthropology 2008:25. Rak Y, Hylander WL. 2008b.What else is the tall mandibular ramus of the robust australopiths good for? In: Vinyard C, Ravosa MJ, Wall C, editors. Primate craniofacial function and biology. New York: Springer Science1Business Media. p 431 442. Rak Y, Hylander WL. 2009. Opposing extreme of jaw mechanics: Australopithecus boisei vs. Homo neanderthalensis [abstract]. Am J Phys Anthropol 120:217. Rak Y, Hylander WL. 2014. The gape of Homo neanderthalensis [abstract]. Am J Phys Anthropol 153:216. Sergi S. 1942. Sulle variazioni di posizione dell osso zigomatico nell uomo. Riv. Antr XXXIII:1 48. Smith AL, Benazzi S, Ledogar JA, Tamvada K, Pryor Smith LC, Weber GW, Spencer MA, Lucas PW, Michael S, Shekeban A, et al. 2015. The feeding biomechanics and dietary ecology of Paranthropus boisei. Anat Rec 298:145 167. Strait DS, Grosse IR, Dechow PC, Smith AL, Wang Q, Weber GW, Neubauer S, Slice DE, Chalk J, Richmond BG, et al. 2010. The structural rigidity of the cranium of Australopithecus africanus: Implications for diet, dietary adaptations, and the allometry of feeding biomechanics. Anat Rec 293:583 593.