1. The placement of the Maxillo-Zygomatic suture in the primate midfacial
skeleton: An investigation on Old World Monkeys and New World Monkeys
Qian Wang 1, Jana Makedonska 2, Craig Byron 3, David Strait 2
1Division of Basic Medical Sciences, Mercer University School of Medicine, 2Department of Anthropology, University at Albany , 3 Department of Biology, Mercer University
Craniofacial sutures are major sites of bone expansion during postnatal skull growth
(Opperman, 2000). It is generally thought that the growth and morphology of craniofacial
sutures reflect their functional environment (Rafferty and Herring, 1999). Craniofacial
sutures can be viewed weak bone “growth” sites characterized by very different material
properties than the surrounding rigid skull bones; thence they must be shielded from
unduly high stresses so as not to disrupt vital growth processes and skeletal functions.
Thus, it is hypothesized that the placement of craniofacial sutures should maximize their
growth potentials yet minimize their negative biomechanical impacts, especially in areas
under high stress during dietary activities, such as the midface. Specifically, for any given
suture, it is hypothesized that suture position would be different in skulls of different form
and/or adapted to different dietary ecology.
In this pilot study, we investigated the position of the Maxillo-Zygomatic suture (MZS) and
the covariance patterns between zygomatic bone 3D sutural and geometric landmarks in a
sample including Old World Monkeys (OWMs) and New World Monkeys (NWMs) with the aim
to identify phylogenetic (including allometric) effects and distinctions among dietary
groups. Further, we studied the association of the zygomatic bone with maxillary and
premaxillary traits and neurocranial traits.
Results: Hypotheses # 1 is supported. Hypothesis # 2 not fully supposed by PC 2. Yet, zygomaxillare inferior is slightly more
laterally located in the robust NWM.
The comparison of species’ mean
configurations will certainly be more
suitable for assessing potentially
diet-driven variation in the position
of the zygomaxillare landmarks.
This research was supported by NSF-HOMINID grant BCS 0725126, BCS 0725183 and by Doctoral Dissertation Improvement Grant BCS 1028815.
Klingenberg C. 2011. MorphoJ: an integrated software package for geometric morphometrics. Molec Ecol Res 11: 353 – 357.
Makedonska J, Wright BW, Strait DS. 2012. The Effect of Dietary Adaption on Cranial Morphological Integration in Capuchins (Order Primates, Genus Cebus). The Effect of Dietary Adaption on Cranial Morphological
Integration in Capuchins (Order Primates, Genus Cebus). PLoS ONE 7(10): e40398. doi:10.1371/journal.pone.0040398
O’Higgins P, Jones N. 2006. Tools for statistical shape analysis. Hull York Medical School. http://sites.google.com/site/hymsfme/resources.
Opperman LA. 2000. Cranial sutures as intramembranous bone growth sites. Dev Dyn 219: 472 – 485.
Rafferty KL, Herring SW. 1999. Craniofacial sutures: morphology, growth and in vivoa masticatory strains. J Morphol 242: 167 – 179.
Wiley DF, Amenta N, Alcantara DA, Ghosh D, Kil YJ, Delson E,, Harcourt-Smith W, Rohlf FJ, St. John K, Hamann B. 2005. Evolutionary morphing. Proc IEEE Visualiz..
Summary and Discussion
Here we report a correlated shape change between the facial and temporal segments of the zygomatic bone, which
characterizes the distinction between OWMs and NWMs. In particular, OWMs have more medially placed zygomaxillare
superiors and more medially and anteriorly placed zygomaxillare inferiors. Although visual observations suggest that hard-
object feeding, robust NWMs might have more laterally placed zygomaxillare inferiors than the non-durophagous, gracile
NWMs, further analyses, and specifically the comparison of species’ means are needed. Finally, the zygomatic portion, located
in a highly strained area that supports the masseter muscle and involved in feeding function, is consistently more strongly
correlated with the rostral and molar units than with the zygomatic portion containing landmarks located on structures
supporting the brain and the eye. Thus, the placement of facial sutures and the covariance patterns involving the zygomatic
bone warrant careful ontogenetic, phylogenetic and biomechanical studies.
Materials and Methods
33 3D landmarks were digitized using Landmark Editor (Wiley et al., 2005) on NextEngine surface
models of 158 crania (for details on methodology see Makedonska et al., 2012) of adult
anthropoid specimens housed at the Caribbean Primate Research Center, the Field Museum, the
American Museum of Natural History and the National Museum of Natural History.
Sample used in this study:
Landmarks used in this study, digitized on a
virtual C. apella s.s. cranial model.
Hypotheses # 1 and # 2 tested through Principal Components Analysis (PCA) carried out on the
Procrustes superimposed 3D coordinates of 22 zygomatic landmarks, not regressed on centroid
size and not pooled within species, parvorder or sex.
Hypothesis # 3 tested through within-configuration two-block Partial-Least Squares (PLS)
analyses evaluating the between-block multivariate correlation magnitude (RV coefficient) in two
competing partition scenarios:
PLS 1: developmental scenario: block 1 – a zygomatic unit including all zygomatic landmarks,
block 2 – a non-zygomatic unit including all rostral, molar and other neurocranial landmarks.
PLS 2: functional scenario: block 1 – a feeding unit including anterior zygomatic (zygomaxillare
inferiors, zygomatic roots, zygotemporales), rostral and molar landmarks, and block 2 – a
sensory unit including the zygomaxillare superiors, the fronto-malares, the pterions, the second
sutural landmark on the sphenoid, the external auditory meatus and the asterions.
Both scenarios were evaluated within species, on residuals from pooled-within-sex regression
on centroid size, in a selected subsample of species (those with larger sample sizes).
Morphometric analyses (PCA, PLS, regressions) carried out in MorphoJ (Klingenberg, 2011) and
Morphologika (O’Higgins and Jones, 2006).
Hypothesis # 1 :
Based on prior observations, we hypothesize that catarrhine crania are significantly distinct
from platyrrhine crania in that they concomitantly exhibit:
a. a more medially located zygomaxillare superior and a more anteriorly located
zygomaxillare inferior, resulting in a larger anterior zygomatic portion in the maxillary
b. a more anteriorly located pterion indicating the degree of the posterior expansion of the
zygoma, which results in a smaller zygomatic contribution to the cranial vault.
This hypothesis concerns the development of the facial and the temporal parts of the zygoma,
and implies the presence of a correlated shape change between the anterior portion and
the posterior portion of the zygomatic.
Hypothesis # 2:
Hard-object-feeding, robust taxa (i.e., in this study, the tufted capuchin Cebus (Sapajus) apella
sensu stricto (s.s.) and the bearded saki Chiropotes satanas), are characterized by more
laterally located maxillo-zygomatic sutures than non-durophagous, gracile closely-related
taxa (i.e., in this study, the capuchin Cebus albifrons), since a laterally located MZS lies out
of the areas of highest feeding-related strain.
Hypothesis # 3:
The zygomatic landmarks, being linked through development, form a module or a suite of
autonomous and highly inter-correlated traits whose variation is relatively independent
from other non-zygomatic traits. The alternative hypothesis predicts that the functional
partition of the cranium estabishes correlations between skeletal traits that overrides their
developmental history; thus, it is expected that the anterior zygomatic will share stronger
correlations with the maxillary and premaxillary bones than with the posterior zygomatic.
Hypothesis # 3: In all examined species, the integration between the zygomatic “module” (that is all zygomatic landmarks
lumped in a single block) and the block containing the rest of the landmarks (molar, rostral, cranial vault), is higher than the
correlation between the feeding unit (part of zygomatic landmarks, rostral landmarks and molar landmarks) and the sensory
unit (part of zygomatic landmarks and posterior cranial vault landmarks). Some of the latter correlations are not significant
at the 0.05 level (see table below). Hence, the null Hypothesis # 3 is falsified.
Species Sample size
Cebus albifrons 18 (9♀, 9♂)
Cebus apella s.s. 21 (10♀, 11♂)
Chiropotes satanas 19 (9♀, 10♂)
Colobus guereza 17 (10♀, 7♂)
Colobus angolensis 13 (8♀, 5♂)
17 (11♀, 6♂)
Erythrocebus patas 13 (7♀, 6♂)
Papio anubis 18 (4♀, 14♂)
Macaca mulatta 12 (6♀, 6♂)
Macaca nemestrina 3 (3♀)
Macaca arctoides 7 (6♀, 1♂)
Total 158 (83♀, 75♂)
Vectors at landmarks summarizing
the shape transformation from
NWM to OWM along PC 1
Vectors at landmarks summarizing the
shape transformation from colobines,
grivets and patas monkeys to
baboons along PC 2 (or from non-
durophagous to durophagous NWMs)
Non-pooled regressions of PC scores
on centroid size
Antero-posterior zygomatic boundary
of the variation
predicts 16.41 % of
the variation along
Species Correlation between the zygomatic
landmarks and the rest of landmarks
Correlation between the feeding
unit and the sensory unit
Cebus albifrons 0.526 (p=0.018) 0.445 (p=0.096)
Cebus apella s.s. 0.60 (p<0.001) 0.476 (p=0.01)
Chiropotes satanas 0.6435 (p<0.001) 0.597 (p<0.001)
Cercopithecus aethiops 0.5748 (p=0.002) 0.4865 (p=0.066)
Erythrocebus patas 0.622 (p=0.01) 0.4824 (p=0.179)
Colobus guereza 0.675 (p<0.001) 0.4642 (p=0.016)
Colobus angolensis 0.6646 (p<0.004) 0.582 (p=0.021)
Papio anubis 0.5794 (p=0.003) 0.561 (p=0.001)