Masticatory stress, orbital orientation and the evolution of the primate postorbital bar

A postorbital bar is one of a suite of derived features which distinguishes basal primates from their putative sister taxon, plesiadapiforms. Two hypotheses have been put forward to explain postorbital bar development and variation in circumorbital form: the facial torsion model and visual predation hypothesis. To test the facial torsion model, we employ strain data on circumorbital and mandibular loading patterns in representative primates with a postorbital bar and masticatory apparatus similar to basal primates. To examine the visual predation hypothesis, we employ metric data on orbit orientation in Paleocene and Eocene primates, as well as several clades of visual predators and foragers that vary interspecifically in postorbital bar formation.A comparison of galago circumorbital and mandibular peak strains during powerful mastication demonstrates that circumorbital strains are quite low. This indicates that, as in anthropoids, the strepsirhine circumorbital region is excessively overbuilt for countering routine masticatory loads. The fact that circumorbital peak-strain levels are uniformly low in both primate suborders undermines any model which posits that masticatory stresses are determinants of circumorbital form, function and evolution. This is interpreted to mean that sufficient cortical bone must exist to prevent structural failure due to non-masticatory traumatic forces. Preliminary data also indicate that the difference between circumorbital and mandibular strains is greater in larger taxa.Comparative analyses of several extant analogs suggest that the postorbital bar apparently provides rigidity to the lateral orbital margins to ensure a high level of visual acuity during chewing and biting. The origin of the primate postorbital bar is linked to changes in orbital convergence and frontation at smaller sizes due to nocturnal visual predation and increased encephalization. By incorporating in vivo and fossil data, we reformulate the visual predation hypothesis of primate origins and thus offer new insights into major adaptive transformations in the primate skull.     

Ontogenetic perspective on mechanical and nonmechanical models of primate circumorbital morphology

Dimensions of the supraorbital torus, postorbital bar, and postorbital septum were collected in an ontogenetic series of Macaca fascicularis and compared with expectations based on models that attribute morphological variation in these features to spatial factors, allometry, anterior dental loading, and neurofacial torsion. Each model was evaluated using correlation, partial correlation, and regression techniques (model I/least squares; model II/reduced major axis) applied to log-transformed and size-corrected data. Results indicate clearly that face or skull size is the primary determinant of variation in circumorbital structures. Strong support is found for the influence of spatial influences on anteroposterior supraorbital torus development (Moss and Young, Am. J. Phys. Anthropol. 18:281-292, 1960). Only minor support is noted for the neurofacial torsion model of Greaves (J. Zool. 207:125-136, 1985), and no support is indicated for the anterior dental loading model. The sexes do not differ significantly in any relevant comparisons of ontogenetic trajectories.     

Strain in the galago facial skull

Little experimental work has been directed at understanding the distribution of stresses along the facial skull during routine masticatory behaviors. Such information is important for understanding the functional significance of the mammalian circumorbital region. In this study, bone strain was recorded along the dorsal interorbit, postorbital bar, and mandibular corpus in Otolemur garnettii and O. crassicaudatus (greater galagos) during molar chewing and biting. We determined principal-strain magnitudes and directions, compared peak shear-strain magnitudes between various regions of the face, and compared galago strain patterns with similar experimental data for anthropoids. This suite of analyses were used to test the facial torsion model (Greaves [1985] J Zool (Lond) 207:125-136; [1991] Zool J Linn Soc 101:121-129; [1995] Functional morphology in vertebrate paleontology. Cambridge: Cambridge University Press, p 99-115). A comparison of galago circumorbital and mandibular peak strains during powerful mastication indicates that circumorbital strains are very low in magnitude. This demonstrates that, as in anthropoids, the strepsirhine circumorbital region is highly overbuilt for countering routine masticatory loads. The fact that circumorbital peak-strain magnitudes are uniformly low in both primate suborders undermines any model that emphasizes the importance of masticatory stresses as a determinant of circumorbital form, function, and evolution. Preliminary data also suggest that the difference between mandibular and circumorbital strains is greater in larger-bodied primates. This pattern is interpreted to mean that sufficient cortical bone must exist in the circumorbital region to prevent structural failure due to nonmasticatory traumatic forces. During unilateral mastication, the direction of epsilon(1) at the galago dorsal interorbit indicates the presence of facial torsion combined with bending in the frontal plane. Postorbital bar principal-strain directions during mastication are oriented, on average, very close to 45 degrees relative to the skull's long axis, much as predicted by the facial torsion model. When chewing shifts from one side of the face to the other, there is a characteristic reversal or flip-flop in principal-strain directions for both the interorbit and postorbital bar. Although anthropoids also exhibit an interorbital reversal pattern, peak-strain directions for this clade are opposite those for galagos. The presence of such variation may be due to suborder differences in relative balancing-side jaw-muscle force recruitment. Most importantly, although the strain-direction data for the galago circumorbital region offer support for the occurrence of facial torsion, the low magnitude of these strains suggests that this loading pattern may not be an important determinant of circumorbital morphology.     

Septa and processes: convergent evolution of the orbit in haplorhine primates and strigiform birds

According to the “nocturnal visual predation hypothesis” (NVPH), the convergent eyes and orbits of primates result from selection for improved stereoscopic depth perception to facilitate manual capture of prey at night. Within primates, haplorhines share additional derived orbital morphologies, including a postorbital septum and greater orbital convergence than any other mammalian clade. While the homology and function of the haplorhine septum remain controversial, experimental data suggest that septa evolved to inhibit mechanical disturbance of the orbital contents by the anterior temporalis muscle during mastication. According to this “insulation hypothesis,” haplorhines are particularly susceptible to disruption of the orbital contents because they have large and highly convergent eyes and orbits. However, comparative tests of the insulation hypothesis have been hindered by the morphological uniqueness of the haplorhine septum among mammals. Among birds, owls (Strigiformes) exhibit an expanded postorbital process that may be functionally analogous to the haplorhine septum. Here we present a comparative analysis of orbital morphology in 103 avian species that tests two hypotheses: (1) large, convergent orbits are associated with nocturnal visual predation, and (2) the strigiform postorbital process and haplorhine postorbital septum similarly function to insulate the eyes from contractions of mandibular adductors. Strigiforms, as nocturnal visual predators, possess relatively large orbits and exhibit the highest degree of orbital convergence in our sample. Notably, orbital convergence does not scale with orbit size in birds as in mammals. Owls are also unique among the birds examined in possessing extensive, plate-like postorbital processes that largely isolate the orbits from the temporal fossae. Furthermore, dissections of four owl species demonstrate that the expanded strigiform postorbital process deflects the path of mandibular adductors around the eye's inferolateral margin. These findings provide further comparative support for both the NVPH and the insulation hypothesis.     

In vivo and in vitro bone strain in the owl monkey circumorbital region and the function of the postorbital septum

Anthropoids and tarsiers are the only vertebrates possessing a postorbital septum. This septum, formed by the frontal, alisphenoid, and zygomatic bones, separates the orbital contents from the temporal muscles. Three hypotheses suggest that the postorbital septum evolved to resist stresses acting on the skull during mastication or incision. The facial-torsion hypothesis posits that the septum resists twisting of the face about a rostrocaudal axis during unilateral mastication; the transverse-bending hypothesis argues that the septum resists caudally directed forces acting at the lateral orbital margin during mastication or incision; and the tension hypothesis suggests that the septum resists ventrally directed components of masseter muscle force during mastication and incision. This study evaluates these hypotheses using in vitro and in vivo bone strain data recorded from the circumorbital region of owl monkeys. Incisor loading of an owl monkey skull in vitro bends the face upward in the sagittal plane, compressing the interorbital region rostrocaudally and “buckling” the lateral orbital walls. Unilateral loading of the toothrow in vitro also bends the face in the sagittal plane, compressing the interorbital region rostrocaudally and buckling the working side lateral orbital wall. When the lateral orbital wall is partially cut, so as to reduce the width of its attachment to the braincase, the following changes in circumorbital bone strain patterns occur. During loading of the incisors, lower bone strain magnitudes are recorded in the interorbital region and lateral orbital walls. In contrast, during unilateral loading of the P³, higher bone strain magnitudes are observed in the interorbital region, and generally lower bone strain magnitudes are observed in the lateral orbital walls. During unilateral loading of the M², higher bone strain magnitudes are observed in both the interorbital region and in the lateral orbital wall ipsilateral to the loaded molar. Comparisons of the in vitro results with data gathered in vivo suggest that, during incision and unilateral mastication, the face is subjected to upward bending in the sagittal plane resulting in rostrocaudal compression of the interorbital region. Modeling the lateral orbital walls as curved plates suggests that during mastication the working side wall is buckled due to the dorsally directed component of the maxillary force which causes upward bending of the face in the sagittal plane. The balancing side lateral orbital wall may also be buckled due to upward bending of the face in the sagittal plane as well as being twisted by the caudoventrally directed components of the superficial masseter muscle force. The in vivo data do not exclude the possibility that the postorbital septum functions to improve the structural integrity of the postorbital bar during mastication. However, there is no reason to believe that a more robust postorbital bar could not also perform this function. Hypotheses stating that the postorbital septum originally evolved to reinforce the skull against routine masticatory loads must explain why, rather than evolving a postorbital septum, the stem anthropoids did not simply enlarge their postorbital bars. © 1996 Wiley-Liss, Inc.     

A model for comparison of masticatory effectiveness in primates

A model is presented to express how effectively animals increase the exposed surface area of their food by chewing. It includes a coefficient of masticatory effectiveness (E) the value of which increases with effectiveness of exposing new food surface area with each chew. Humans and other species of primates differ significantly in their values of E; among the nonhuman primates studies, Lemur catta has a higher coefficient than Lemur fulvus, and both have higher coefficients than either Varecia variegatus or Galago crassicaudatus argentatus. The differences among the coefficients to these prosimians are correlated with variations in specific features of the molar morphology. Of six lower molar shearing crests considered, the relative length of the post-metacristid correlates most highly with the coefficient of masticatory effectiveness for the prosimian species. Also, among comparable-sized prosimians, E correlates significantly with the absolute postmetacristid length. Both these findings indicate that the relative size of molar shearing crests is related significantly to how effectively an animal chews its food. There are also implications for an adaptation to a high-fiber diet.     

Masticatory-stress hypotheses and the supraorbital region of primates

The purpose of this study is to test various masticatory-stress hypotheses about the evolution and function of well-developed browridges of higher primates. This was done by measuring and analyzing patterns of in vivo bone strain recorded from three-element rosette strain gages bonded to the supraorbital region and to other portions of the bony face of Macaca fascicularis and Papio anubis during mastication and incision. The magnitude and direction of the principal strains recorded support Endo's hypothesis that the supraorbital region during mastication and incision is bent in the frontal plane (Endo, 1966). Our data do not, however, support his hypothesis that the supraorbital region is bent more during incision than during mastication. The data also demonstrate that overall levels of supraorbital strain are not larger in more prognathic subjects. Most importantly, the data indicate that the supraorbital region of nonhuman catarrhines is strained very little during mastication and incision. This indicates that there is much more supraorbital bone than is necessary both to counter masticatory loads and to provide an adequate safety factor to failure for these loads. This in turn suggests that the macaque and baboon browridges can be considerably reduced in size and still maintain these required structural characteristics. Thus, our experiments provide no support whatsoever for those hypotheses that directly link browridge morphology to masticatory stress (cf. Endo, 1966; Russell, 1983, 1985). A recent review of Endo's original work indicates that this latter statement is also true for humans (Picq and Hylander, 1989). We conclude, therefore, that there is no good reason to believe that enlarged browridges in living and/or fossil primates are structural adaptations to counter intense masticatory forces. The evolution of browridge morphology in primates is best explained on the basis of factors related to the position of the brain relative to the orbits (Moss and Young, 1960). When these structures are widely separated, as in gorillas, the large intervening space must be bridged with bone. In addition, enough bone must be present within the supraorbital and bridged regions to prevent structural failure due to non-masticatory external forces associated with highly active primates (e.g., accidental traumatic forces applied to the orbits and neurocranium). This requirement results in both pronounced browridges and in much more supraorbital bone than is necessary to counter routine cyclical stress during mastication and incision. This in turn explains why bone strains recorded from the supraorbital region are extremely small relative to other portions of the primate face during mastication and incision.     

Dietary correlates of temporomandibular joint morphology in New World primates

Previous analyses of the masticatory apparatus have demonstrated that the shape of the temporomandibular joint (TMJ) is functionally and adaptively linked to variation in feeding behavior and diet in primates. Building on previous research, this study presents an analysis of the link between diet and TMJ morphology in the context of functional and dietary differences among New World primates. To evaluate this proposed relationship, I used three-dimensional morphometric methods to quantify TMJ shape across a sample of 13 platyrrhine species. A broad interspecific analysis of this sample found strong relationships among TMJ size, TMJ shape, and diet, suggesting that both size and diet are significant factors influencing TMJ morphology in New World primates. However, it is likely that at least some of these differences are related to a division of dietary categories along clade lines.A series of hypotheses related to load resistance capabilities and range of motion in the TMJ were then tested among small groups of closely related taxa with documented dietary differences. These pairwise analyses indicate that some aspects of TMJ morphology can be used to differentiate among closely related species with different diets. However, not all of my predictions were upheld. The anteroposterior dimensions of the TMJ were most strongly consistent with hypothesized differences in ingestive/masticatory behaviors and jaw gape, whereas the predictions generated for variation in entoglenoid and articular tubercle height were not upheld. These results imply that while some features can be reliably associated with increased load resistance and facilitation of wider jaw gapes in the masticatory apparatus, other features are less strongly correlated with masticatory function.     

The circumorbital foramina in primates

This paper documents the diversity and variation in the circumorbital foramina in extant primates. A qualitative and quantitative analysis of the circumorbital foramina, comprising the supraorbital foramen, the infraorbital foramen, and the malar foramen, was carried out using representative species from nine extant families of primates. The information obtained from the study is used to reconstruct ancestral morphotypes, and to make inferences about evolutionary changes that may have taken place in the major primate lineages. In addition, the analysis provides useful comparative data for interpretation of the phylogenetic significance and paleobiological implications of the circumorbital foramina in fossil primates.     

Function of the mammalian postorbital bar

Complete postorbital bars, bony arches that encompass the lateral aspect of the eye and form part of a circular orbit, have evolved homoplastically multiple times during mammalian evolution. Numerous functional hypotheses have been advanced for postorbital bars, the most promising being that postorbital bars function to stiffen the lateral orbit in taxa that have significant angular deviation between the temporal fossa and the bony orbit. Without a stiff lateral orbit the anterior temporalis muscle and fascia potentially would pull on the postorbital ligament, deform the orbit, and cause disruption of oculomotor precision. Morphometric data were collected on 1,329 specimens of 324 taxa from 16 orders of extant eutherian and metatherian mammals in order to test whether the orientation of the orbit relative to the temporal fossa is correlated with the replacement of the postorbital ligament with bone. The allometric and ecological influences on orbit orientation across mammals are also explored. The morphometric results corroborate the hypothesis: Shifts in orbit orientation relative to the temporal fossa are correlated with the size of the postorbital processes, which replace the ligament. The allometric and ecological factors that influence orbit orientation vary across taxa. Postorbital bars stiffen the lateral orbital wall. Muscle pulleys, ligaments, and other connective tissue attach to the lateral orbital wall, including the postorbital bar. Without a stiff lateral orbit, deformation due to temporalis contraction would displace soft tissues contributing to normal oculomotor function.     

Muscular and osseous anatomy of the primate anterior temporal fossa and the functions of the postorbital septum

Many adaptive explanations for anthropoid origins incorporate hypotheses regarding the function of the postorbital septum. Two hypotheses are evaluated here: Cachel's ([1979b] Am. J. Phys. Anthropol. 50:1–18) hypothesis that the anthropoid postorbital septum evolved to augment muscle attachment area in the anterior temporal fossa and Cartmill's ([1980] in RL Ciochon and AB Chiarelli (eds.): Evolutionary Biology of the New World Monkeys and Continental Drift. New York: Plenum, pp. 243–274.) hypothesis that the septum evolved to insulate the foveate eye of haplorhines from movements in the temporal fossa during mastication. Dissections of the masticatory muscles of 55 species of primates, with emphasis on the anatomy of the anterior temporal fossa, reveal that in all anthropoids the temporal muscles take origin from the portion of the septum formed by the frontal bone. In some platyrrhines this muscle is anterior temporalis, and in others it is zygomatico-mandibularis. In tarsiers and most platyrrhines, muscle attachment to the zygomatic portion of the postorbital septum is very restricted (and of possibly varying homologies), whereas in catarrhines the zygomatico-mandibularis arises from the postorbital ridge on the zygomatic portion of the septum. This suggests that, contrary to Cachel's hypothesis, the earliest anthropoids did not have extensive areas of muscle attachment on the postorbital septum, a suggestion supported by the bony morphology of Catopithecus browni. Dissections also indicate that in all haplorhines the anteriormost temporal fibers curve around the postorbital septum between origin and insertion, implying that, were the septum not present, the anterior temporal muscles would disturb the orbital contents when contracting. This suggests that insulation may have been the septum's original function, even in the absence of a retinal fovea. In anthropoids, the rostral migration of the line of action of the anterior temporal muscles relative to the eye is attributed to their possession of extreme degrees of both orbital frontation and convergence; in tarsiers it is attributed to their possession of both massively hypertrophied eyes and moderately convergent and frontated orbits. It is argued that the postorbital septum is most likely to have evolved in a morphological context similar to that exhibited by omomyids. © 1995 Wiley-Liss, Inc.     

Skull of Catopithecus browni,an early tertiary catarrhine

Fossil crania from quarry L-41, Fayum, Egypt, representing Catopithecus browni, a primate similar in size to callitrichids but with a catarrhine dental formula, provide the geologically earliest record of an anthropoidean skull. Catopithecus had postorbital closure developed to the stage seen in extant anthropoideans, with direct contact between zygomatic plate and maxillary tuber, isolating an anterior orbital fissure from the inferior orbital fissure. The auditory region also resembles that of later anthropoideans: The posterior carotid foramen is placed adjacent to the jugular fossa; a large promontory canal crosses the promontorium; and the annular ectotympanic is fused ventrally to the bulla. The incisors and canines show an assemblage of features found only among modern anthropoideans and adapoids. The face is characterized by a relatively deep maxilla, broad ascending wing of the premaxilla, and long nasal bones, yielding a moderate muzzle similar to that of Aegyptopithecus. The small braincase bears an anteriorly broad frontal trigon and a posteriorly developed sagittal crest. The mandibular symphysis is unfused even in mature adults. The encephalization quotient (EQ) probably falls within the range of Eocene prosimians, much lower than the EQs of Neogene anthropoideans. © 1996 Wiley-Liss, Inc.     

Primary homologies of the circumorbital bones of snakes

Some snakes have two circumorbital ossifications that in the current literature are usually referred to as the postorbital and supraorbital. We review the arguments that have been proposed to justify this interpretation and provide counter‐arguments that reject those conjectures of primary homology based on the observation of 32 species of lizards and 81 species of snakes (both extant and fossil). We present similarity arguments, both topological and structural, for reinterpretation of the primary homologies of the dorsal and posterior orbital ossifications of snakes. Applying the test of similarity, we conclude that the posterior orbital ossification of snakes is topologically consistent as the homolog of the lacertilian jugal, and that the dorsal orbital ossification present in some snakes (e.g., pythons, Loxocemus, and Calabaria) is the homolog of the lacertilian postfrontal. We therefore propose that the terms postorbital and supraorbital should be abandoned as reference language for the circumorbital bones of snakes, and be replaced with the terms jugal and postfrontal, respectively. The primary homology claim for the snake “postorbital” fails the test of similarity, while the term “supraorbital” is an unnecessary and inaccurate application of the concept of a neomorphic ossification, for an element that passes the test of similarity as a postfrontal. This reinterpretation of the circumorbital bones of snakes is bound to have important repercussions for future phylogenetic analyses and consequently for our understanding of the origin and evolution of snakes. J. Morphol. 274:973–986, 2013. © 2013 Wiley Periodicals, Inc.     

Where to draw the nonprimate-primate taxonomic boundary.

The known Cretaceous and Paleocene primates, the Paromomyiformes, although lacking a fully developed postorbital bar, are nevertheless, both cladistically and phenetically, closest to the common ancestor of the living primates. Not only their undisputed phylogenetic ties but also their early experiments inthe arboreal milieu necessitate their balanced evolutionary classification within the order Primates.     

Effects of activity pattern on eye size and orbital aperture size in primates

Among primates, nocturnal species exhibit relatively larger orbital apertures than diurnal species. Most researchers have considered this disparity in orbital aperture size to reflect differences in eye size, with nocturnal primates having relatively large eyes in order to maximize visual sensitivity. Presumed changes in eye size due to shifts in activity pattern are an integral part of theoretical explanations for many derived features of anthropoids, including highly convergent orbits and a postorbital septum. Here I show that despite clear differences in relative orbital aperture size, many diurnal and nocturnal primates do not differ in relative eye size. Among nocturnal primates, relative eye size is influenced by diet. Nocturnal visual predators (e.g., Tarsius, Loris, and Galago moholi) tend to have larger relative eye sizes than diurnal primates. By contrast, nocturnal frugivores (e.g., Perodicticus, Nycticebus, and Cheirogaleus) have relative eye sizes that are comparable to those of diurnal primates. Although some variation in orbital aperture size can be attributed to variation in eye size, both cornea size and orbit orientation also exert a strong influence on orbital aperture size. These findings argue for caution in the use of relative orbital aperture size as an indicator of activity pattern in fossil primates. These findings further suggest that existing scenarios for the evolution of unique orbital morphologies in anthropoids must be modified to reflect the importance of ecological variables other than activity pattern.     

Tooth eruption and browridge formation

One of the most reasonable hypotheses regarding the functional significance of the browridge is that the supraorbital torus forms in response to masticatory stress during development. Oyen, Walker, and Rice (1979) have recently proposed a model that tests this hypothesis: if browridges are functionally related to masticatory stresses on the cranial vault, then changes in the biomechanics of the masticatory system ought to be reflected by changes in the browridge. To test their model they attempted to relate biomechanical discontinuities resulting from tooth eruption to episodes of bone deposition on the supraorbital tori of a developmental series of dry Papio crania. This paper reports on a parallel test of the model on a cross-sectional sample of Australian Aboriginal juvenile crania. This sample showed no relation between tooth eruption and the supraorbital surface morphology thought to be indicative of active bone deposition. It is also demonstrated that no significant relationship between tooth eruption and episodes of bone deposition is shown by the Papio sample. It is concluded that the use of small cross-sectional samples of dry crania does not provide a valid test of the model.     

Neocortical development and social structure in primates

The relationships between the relative size of the neocortex and differences in social structures were examined in prosimians and anthropoids. The relative size of the neocortex (RSN) of a given congeneric group in each superfamily of primates was measured based on the allometric relationships between neocortical volume and brain weight for each superfamily, to control phylogenetic affinity and the effects of brain size. In prosimians, “troop-making” congeneric groups (N=3) revealed a significantly larger RSN than solitary groups (N=6), and there was a significant, positive correlation between RSN and troop size. In the case of anthropoids, polygynous/frugivorous groups (N=5) revealed a significantly larger RSN than monogynous/frugivorous groups (N=8). Furthermore, a significant, positive correlation between RSN and troop size was found for frugivorous congeneric groups of the Ceboidea. These results suggest that neocortical development is associated with differences in social structure among primates.     

Ontogeny and phyletic size change in living and fossil lemurs

Lemurs are notable for encompassing the range of body‐size variation for all primates past and present—close to four orders of magnitude. Benefiting from the phylogenetic proximity of subfossil lemurs to smaller‐bodied living forms, we employ allometric data from the skull to probe the ontogenetic bases of size differentiation and morphological diversity across these clades. Building upon prior pairwise comparisons between sister taxa, we performed the first clade‐wide analyses of craniomandibular growth allometries in 359 specimens from 10 lemuroids and 176 specimens from 8 indrioids. Ontogenetic trajectories for extant forms were used as a criterion of subtraction to evaluate morphological variation, and putative adaptations among sister taxa. In other words, do species‐level differences in skull form result from the differential extension of common patterns of relative growth? In lemuroids, a pervasive pattern of ontogenetic scaling is observed for facial dimensions in all genera, with three genera also sharing relative growth trajectories for jaw proportions (Lemur, Eulemur, Varecia). Differences in masticatory growth and form characterizing Hapalemur and fossil Pachylemur likely reflect dietary factors. Pervasive ontogenetic scaling characterizes the facial skull in extant Indri, Avahi, and Propithecus, as well as their larger, extinct sister taxa Mesopropithecus and Babakotia. Significant interspecific differences are observed in the allometry of indrioid masticatory proportions, with variation in the mechanical advantage of the jaw adductors and stress‐resisting elements correlated with diet. As the growth series and adult data are largely coincidental in each clade, interspecific variation in facial form may result from selection for body‐size differentiation among sister taxa. Those cases where trajectories are discordant identify potential dietary adaptations linked to variation in masticatory forces during chewing and biting. Although such dissociations highlight selection to uncouple shared ancestral growth patterns, they occur largely via transpositions and retention of primitive size‐shape covariation patterns or relative growth coefficients. Am. J. Primatol. 72:161–172, 2010. © 2009 Wiley‐Liss, Inc.     

Cranial morphology of Aegyptopithecus and Tarsius and the question of the tarsier-anthropoidean clade

New crania of the Oligocene anthropoidean Aegyptopithecus provide a test of the hypothesized tarsier-anthropoidean clade. Three cranial characters shared by Tarsius and some modern anthropoideans (apical interorbital septum, postorbital septum, "perbullar" carotid pathway) were examined. 1) An apical interorbital septum is absent in Aegyptopithecus. A septum does occur in Galago senegalensis (Lorisidae) and Microcebus murinus (Cheirogaleidae), so the presence of a septum is not strong evidence favoring a tarsiiform-anthropoidean clade. 2) In Aegyptopithecus and other anthropoideans, the postorbital septum is formed mainly by a periorbital flange of the zygomatic that extends medially from the lateral orbital margin onto or near the braincase. The postorbital plate of Tarsius is formed by frontal and alisphenoid flanges that extend laterally from the braincase to the zygomatic's frontal process, which is not broader than the postorbital bars of other prosimians. Periorbital flanges evolved in Tarsius for support or protection of the enormous eyes, as suggested by the occurrence of maxillary and frontal flanges that cup portions of the eye but do not separate it from temporal muscles. 3) The internal carotid artery of Aegyptopithecus enters the bulla posteriorly and crosses the anteroventral part of the promontorium. The tympanic cavity was probably separated from the anteromedial cavity by a septum stretching from the carotid channel to the ventrolateral bullar wall. In Tarsius, the carotid pathway is prepromontorial, and a septum stretches from the carotid channel to the posteromedial bullar wall. Quantitative analyses indicate that anterior carotid position has evolved because of erect head posture. The cranium of Oligocene anthropoideans thus provides no support for the hypothesized tarsier-anthropoidean clade.