Comparative microcomputed tomography and histological study of maxillary pneumatization in four species of new world monkeys The perinatal period.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 144:392–410 (2011) Comparative Microcomputed Tomography and Histological Study of Maxillary Pneumatization in Four Species of New World Monkeys: The Perinatal Period Timothy D. Smith,1,2* James B. Rossie,3 Gregory M. Cooper,4 Robin M. Schmieg,1 Christopher J. Bonar,5 Mark P. Mooney,2,6 and Michael I. Siegel2 1 School of Physical Therapy, Slippery Rock University, Slippery Rock, PA Department of Anthropology, University of Pittsburgh, Pittsburgh, PA 3 Department of Anthropology, SUNY Stony Brook, Stony Brook, NY 4 Department of Surgery, Oral Biology, and Bioengineering, University of Pittsburgh, Pittsburgh, PA 5 Cleveland Metroparks Zoo, Cleveland, OH 6 Department of Oral Biology, University of Pittsburgh, Pittsburgh, PA 2 KEY WORDS craniofacial; nasal cavity; paranasal; pneumatization; primate ABSTRACT In anthropoid primates, it has been hypothesized that the magnitude of maxillary sinus growth is inﬂuenced by adjacent dental and soft tissue matrices. Relatively, little comparative evidence exists for the perinatal period when secondary pneumatization is at its earliest stages in some primates. Here, dental and midfacial variables were studied in a perinatal sample of four anthropoid primates, including three callitrichines (Leontopithecus, Saguinus, and Callithrix) and Saimiri boliviensis. In the latter species, the maxillary recess (the ontogenetic precursor to a ‘‘true’’ maxillary sinus) does not undergo secondary pneumatization. Using histological methods and micro-computed tomography, midfacial and dental dimensions and radiographic hydroxyapatite density of tooth cusps were measured. The distribution of osteoclasts and osteoblasts was also documented. Krus- kal–Wallis’s one-way analysis of variance tests indicates signiﬁcant (P < 0.05) differences among groups for dental and midfacial measurements. In particular, the posterior maxillary dentition is relatively larger and more mineralized in Saimiri compared to the callitrichines. At posterior dental levels, Saimiri has the lowest palatonasal index [interdental (palatal) width/width of the nasal cavity] and highest bizygomatic–interorbital index. Distribution of osteoclasts indicates that the inferomedial surfaces of the orbits are resorptive in perinatal Saimiri, whereas, in all callitrichines, these surfaces are depository. Taken together, these ﬁndings suggest that pneumatization in Saimiri is suppressed by an inward growth trajectory of the orbits, relatively large posterior dentition, and a correspondingly compressed nasal region. Am J Phys Anthropol 144:392–410, 2011. V 2010 Wiley-Liss, Inc. Paranasal sinuses are air-ﬁlled mucosal evaginations into the facial skeleton that emanate from smaller ‘‘recesses’’ of the developing nasal fossa (Witmer, 1999; Maier, 2000; Rossie, 2006). Paranasal sinus expansion, also termed secondary pneumatization, may provide a growth mechanism for resolving spatial shifts among skeletal elements of the nasal capsule, orbit, and dentition (Proetz, 1922; Shea, 1985; Enlow and Hans, 1996; Rossie, 2003; Zollikofer and Weissman, 2008). In this view, variations in the nature of sinus expansion are construed primarily to result from speciﬁc craniofacial growth patterns; other functions that have been offered to explain sinus shape and size in adult mammals (e.g., those related to climate or mastication) would be secondary adaptations (Enlow and Hans, 1996). At present, our knowledge of the relationship of the developing sinus to adjacent structures relies disproportionately on evidence from humans and a few nonhuman primates. For example, clinical and comparative evidence demonstrates that the bony wall of the maxillary sinus surrounds tooth crypts during growth and (later) the root tips, such that portions of the teeth create bulges into the sinus cavity (Schaeffer, 1920; Koppe and Nagai, 1999; Rossie, 2006). It is difﬁcult to ascertain the nature of association between the dentition or palatal form and the ﬁnal shape of the maxillary sinus, despite their close proximity in many primates (Koppe and Nagai, 1999; Swindler, 1999). Two factors make this determination difﬁcult. First, because the interrelationship between sinus and dental/osseous morphology changes throughout development, an understanding of the earliest phases of sinus formation is essential; that is, when the sinuses ﬁrst pneumatize beyond the limits of the nasal capsule. Early development of the most precocious paranasal space, the maxillary sinus, is well documented in humans as an event beginning in the second trimester [(Koppe et al., 1994; Smith et al., 1997; and see Sperber (2000)], whereas it is poorly understood across nonhuman primates. Second, a comparative primate perspective at early stages of development has C 2010 V WILEY-LISS, INC. C Grant sponsor: National Science Foundation; Grant numbers: BCS-0820751, BCS-0610514. Present address of Christopher J. Bonar: Dallas World Aquarium, 1801 N. Grifﬁn St., Dallas, Texas 75202, USA. *Correspondence to: Timothy D. Smith, School of Physical Therapy, Slippery Rock University, Slippery Rock, PA 16057. E-mail: firstname.lastname@example.org Received 1 May 2010; accepted 10 September 2010 DOI 10.1002/ajpa.21421 Published online 29 November 2010 in Wiley Online Library (wileyonlinelibrary.com). 393 MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS been elusive, in part, because generating samples of subadult primates is relatively challenging. An even more challenging issue is the apparent ontogenetic differences in the timing of pneumatic expansion among primates that form a maxillary sinus (Rossie, 2006; Smith et al., 2010). Recently, several studies have begun to address the paucity of ontogenetic information. It has been hypothesized that the size and position of the maxillary dentition is critical factor inﬂuencing the extent of pneumatic expansion of the maxillary recess (Smith et al., 2005, 2010; Rossie, 2006). Microanatomical evidence on several species of callitrichines suggests that where the skeletal units bordering the deciduous premolars and developing maxillary sinus form a shared bony lamina, lateral expansion of this paranasal space is inhibited locally by the dental follicle (Smith et al., 2005). In ontogenetic samples of New World monkeys, relative dental size and the proximity of the alveolar and orbital skeletal structures of the maxilla appear to relate inversely to the extent of pneumatization (Rossie, 2006; Smith et al., 2010). Left unclear is the relative importance of each of these neighboring structures and spaces, a question that may be best answered using a comparative sample. Previously, we have emphasized the potential value of New World primate samples as natural experiments to understand pneumatization (Rossie, 2006; Smith et al., 2010). The speciﬁc aims of this study are to test hypotheses generated by bone-cell distribution along the maxillary sinus walls in callitrichines (Smith et al., 2005) and by an ontogenetic comparison of the maxillary recess of Saimiri to the maxillary sinus of Saguinus and Cebuella (Smith et al., 2010). In these studies, we hypothesized that large relative size of the posterior maxillary dentition, a high degree of orbital approximation, and low-palatonasal indices [interdental (palatal) width/width of the nasal cavity] each act to constrain pneumatization from the maxillary recess in Saimiri. Here, to test these hypotheses, we compare these measures among four species with differing degrees of pneumatic expansion within the maxilla, including one species that does not form a maxillary sinus postnatally. MATERIALS AND METHODS Sample We studied four species of New World monkeys that differ in postnasal paranasal morphology. All species form a maxillary ‘‘recess,’’ a space formed by prenatal growth and folding of the cartilaginous nasal capsule (‘‘primary pneumatization’’). These species differ in the extent of secondary pneumatization of this recess, a process by which the mucosa-lined space invades the adjacent maxilla after breakdown of the cartilaginous capsule. In three of these species, Saguinus oedipus, Leontopithecus rosalia, and Callithrix jacchus, a maxillary sinus has been described in adult specimens by numerous authors (e.g., Nishimura et al., 2005; Rossie, 2006; Smith et al., 2010). In the fourth, Saimiri boliviensis, no secondary pneumatization occurs (Smith et al., 2010), as also reported for S. sciureus Rossie, 2003, 2006; Nishimura et al., 2005). Previously, we have distinguished between ‘‘true’’ paranasal sinuses (the ﬁnal product of secondary pneumatization) and paranasal recesses that persist in adults in the absence of secondary pneumati- TABLE 1. Age and sexes of the sample under study Species n (sex) Age range (d) Saguinus oedipus 5 (2 m, 2 f, 1 u) Leontopithecus rosalia Callithrix jacchus 6 (1 m; 5 f) 5 (2 m, 2 f, 1 u) Saimiri boliviensis 6 (5 m, 1 u) 0 (stillborn or died on P0) 0–9 0 (stillborn or died perinatally) 0–10 d, days; f, female; m, male; u, unknown sex. zation (Rossie, 2006; Smith et al., 2010). Thus, according to our terminology, Saimiri lacks a maxillary sinus. Twenty-two specimens were studied (Table 1). All specimens were acquired from cadaveric remains preserved in 10% formalin. This opportunistic sample included animals that were stillborn or died within 10 days of birth in captivity at the New England Primate Center, the University of South Alabama Primate Research Center, the Wisconson National Primate Research Center, or the Cleveland Metroparks zoo. When possible, both males and females of each species were acquired. However, only male perinatal Saimiri (and one of unknown sex) were available for study. Becasue adults of the genus Saimiri are described to have only minimal dimorphism (Corner and Richtsmeier, 1992), it is not expected that Saimiri boliviensis is dimorphic at birth. The specimens are not all assumed to be neonatal in stage of somatic maturity and likely represent a mix of premature and early postnatal stages (thus, these are referred to as a ‘‘perinatal’’ sample). Perinatal specimens of known age were either stillborn or died from ages P0 to P10 (Table 1). Although facial and dental development and growth are presumably rapid in the interval of ages that are available for study, this sample was considered a relatively cohesive group for two reasons. First, endochondral ossiﬁcation of the lateral nasal wall is underway in all specimens. In all species, only remnants of cartilage are observed along the lateral borders of the nasal fossa, including the maxillary recess [see Smith et al. (2005, 2010)]. Thus, all species have passed the stage in which the ‘‘fetal template’’ of paranasal morphology (i.e., the stage when cartilage forms continuous boundaries around the lateral nasal wall) is apparent [see Smith et al. (2008)]. Second, the samples appeared to be relatively consistent in head and midfacial size within each species. Regarding upper limits in the age range, the specimens of oldest known age (9-day-old Leontopithecus and 10-day-old Saimiri) were within, not at the upper limit, the overall range in cranial or palatal length. Microcomputed tomography and histological methods Sixteen of the perinatal specimens (four of each species) were scanned using microcomputed tomography (micro–CT; 19-lm voxel size; VivaCT40 scanner, Scanco Medical, USA). The scans were used to generate threedimensional reconstructions. The threshold for reconstructions was set at 200 dpi, a level selected based in our ability to detect thin bones such as the turbinals or individual septa that bordered the paranasal spaces. Radiographic hydroxyapatite (HA) density was measured to assess the rate of mineralization of the dentition from American Journal of Physical Anthropology 394 T.D. SMITH ET AL. TABLE 2. Distribution of osteoclasts and osteoblasts along surfaces of the ethmoid and maxillary bones Genus/dental level Saguinus Dp2 Dp3 Dp4 M1 Leontopithecus Dp2 Dp3 Dp4 M1 Species/level Callithrix Dp2 Dp3 Dp4 M1 Saimiri Dp2 Dp3 Dp4 M1 Orbit medial Orbit ﬂoor Inferomedial Orbit Medial alveolar wall 1 Variable 2 6b or NA 1 1 1 1 1 1 1 1 Variablea 6b Variable 6b 1 1 Variable 2 1 1 1 NA 1 1 1 1 1c Variable 2d NA 1 1 1c 1 or NA 1 1 1c 1 1 1 1 1 1 Variable Variable NA 2d 2 2 2 1 Variable 2d 2 or NA 1 2 2 2 6b 6b 2d 2 a Predominantly osteoblasts, osteoclasts, or a mix of bone cells in different specimens. In the same individual, there were both osteoclastic and osteoblastic foci in this region. Limited numbers of osteoclasts seen in some specimens. d Limited groups of osteoblasts seen in some specimens. 1, osteoblasts; 2, OC, osteoclasts; NA, not applicable (e.g., ethmoid is cartilaginous or there is no maxillary bone at this level). b c the anterior to posterior limits of the maxillary tooth row. The threshold for measurements was set at 200 dpi. Density was only measured from deciduous dentition, because M1 is poorly mineralized in all species except Saimiri. HA density was obtained for volumetric segments of enamel in slices extending for about 0.48 mm (25 slices). For premolars with multiple cusps, the buccomesial cusp was selected in all specimens. HA density is ultimately determined by calibrating the VivaCT using a phantom that has samples of known concentrations of HA. Then optical density (radiodensity) is measured and interpreted as mg HA/cc (Usas et al., 2009). After micro-CT scanning, the specimens were dissected and processed for parafﬁn embedding. The histological sample also included specimens previously sectioned for studies of other nasal structures [see Smith et al. (2003) and Dennis et al. (2004) for details]. Before embedding, cranial length (prosthion-inion) and palatal length (prosthion–posterior mid-palatal point) were measured with digital calipers to the nearest 0.01 mm. In some specimens, damage during necropsy prevented acquisition of these measurements. However, these samples were still useful for calculation of palatal and nasal breadth indices (see below). All parafﬁn-embedded heads used in this study were serially sectioned at 10 lm and stained for histomorphometric analysis using ImageJ software (NIH). For each species, three specimens were sectioned in the coronal plane. In one specimen of each species, half the head was sectioned coronally, while the contralateral half was sectioned in the sagittal plane. At least every 10th section was mounted on glass slides with serial numbers and stained with hematoxylin-eosin and Gomori trichrome procedures. Intervening sections were saved for future histochemical and immunohistochemical analysis. Using a Leica DMLB microscope, each stained section was examined at magniﬁcations from 325 to 3100 to American Journal of Physical Anthropology assess position of the sinus or recess and adjacent structures such as dental sacs. After the ﬁrst section containing the deciduous canine and the last section containing M1 were identiﬁed, anteroposterior ‘‘start’’ and ‘‘stop’’ points of all intervening dental sacs and the sinus or recess were noted to determine their spatial relationship. At higher magniﬁcation (3200–3630), the distribution of osteoblasts and osteoclasts was observed along osseous contours of the medial and inferior orbit and along the medial side of the alveolar socket for each tooth from dp2 to M1. For each specimen, three sections in the midlevel of each dental sac were used to record whether these surfaces were predominantly lined with osteoblasts and thus showed evidence for bone deposition or lined with multinucleate osteoclasts and thus interpreted as resorptive surfaces. Also noted were surfaces with mixed distribution of osteoblasts and osteoclasts or surfaces that were devoid of bone cells. Based on all specimens of each species, Table 2 lists the bone cell distribution by species and region, with intraspeciﬁc variation noted. Representative sections that showed the characteristic distribution of bone cells were selected for graphical representation. Histomorphometrics Most of the Saguinus and Saimiri sample was previously analyzed in an ontogenetic study of some palatal, nasal, and dental variables (Smith et al., 2010). Some of these data are included in this comparative analysis. However, we have expanded the analysis to include additional dental levels, more species, more variables, and one additional specimen of Saguinus, prepared speciﬁcally for this study. For most variables, data were obtained at mid-cross-sectional levels for each of the maxillary tooth follicles from the deciduous canine to the ﬁrst permanent molar (M1). Palatonasal indices (see MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS 395 gin of alveolar bone was often indistinct in perinatal animals, alternative landmarks were used: the medial margin of the right and left dental follicles. As per Rossie (2006), nasal width was measured as the distance between the right and left sides of lateral nasal wall at the base of the nasal fossa (this corresponds to the wall of the inferior meatus at the level of the last deciduous premolar). The distance between right and left sides was measured as palatal width. The index was then calculated as interdental width/nasal width. In specimens where only half of the face was sectioned in the coronal plane, the ratio was calculated by measuring one-half of the distances from the dental sac or lateral limit of the nasal fossa to the midline (using the center of the palate as the midline). Two additional indices were calculated for this study: 1. Bizygomaticonasal indices (Fig. 1b) were calculated as maximum bizygomatic width/nasal width. This index assesses the overlap of nasal cavity width relative to facial width. In specimens where only half of the face was sectioned in the coronal plane, this ratio was calculated by measuring one-half of the distances from the maximum lateral limit of the zygoma or lateral limit of the nasal fossa to the midline (center of the palate). 2. Bizygomatic–interorbital indices (Fig. 1c) were calculated as maximum bizygomatic width/minimum interorbital width. This index assesses relative interorbital width. It is a measurement relating to the extent of orbital approximation. In specimens where only half of the face was sectioned in the coronal plane, this ratio was calculated by measuring one-half of the distances from the maximum lateral limit of the zygoma or lateral limit of the nasal fossa to the midline (center of the palate). Fig. 1. Indices of midfacial dimensions used in this study. BZW, bizygomatic width; IOW, interorbital width; NW, nasal width; PW, palatal width. below) were only measured from the levels of the ﬁrst deciduous premolar (dp2) to M1, because the maxillary sinus/recess is posterior to the mid-level of the deciduous or adult canine in each of these species (Smith et al., 2005, 2010). Bizygomatic and interorbital distances (see below) were only measured at the level of the last deciduous premolar (dp4) and M1, because the zygoma is not in the plane of section anterior to dp4 in all species. To acquire data from the histological sample, each stained section ranging from the deciduous canine to M1 was photographed using a Leica DMLB microscope and Cat-Eye digital camera. Measurements were then taken from digital micrographs of histological sections using ImageJ software (NIH). Palatonasal index (Fig. 1a), which measures the proportional relationship between palatal and nasal cavity dimensions, was modiﬁed from Rossie (2006). Rossie measured palatal width as the distance between right and left medial alveolar margins of the maxilla (2006, see Fig. 42). Because the medial mar- Using ImageJ software, dental sac volumes were measured for all maxillary teeth (from deciduous canine to M1) on the right side of the nasal fossa, except in two specimens where damage to the follicles prevented measurements. Among specimens in which right and left sides were both well-preserved for measurement, asymmetry was not biased toward one side (in about half of the specimens, the right dental sacs were larger and in the other half the left sacs were larger). Therefore, the left side was used for analysis in the two specimens where the right side was damaged. Volumes were acquired by tracing the cross-sectional area of the dental sac as estimated by the outer edge of the outer enamel epithelium. Once all sections containing a cross-section of the tooth were measured, the cross-sectional areas in millimeter square were multiplied by the distance between sections for a segmental volume of the tooth. This was repeated for each section of the tooth up to the penultimate section (the last section containing the tooth was regarded as the end). Then, all segmental volumes were summed to obtain total volume of the tooth. The cube root of these tooth volumes was divided by palatal and cranial lengths to obtain relative size of the tooth follicles. Statistical analyses SPSS 15.0 (SPSS) was used to analyze the data. Data were compared between groups using a Kruskal–Wallis one-way analysis of variance test. All differences were considered signiﬁcant at P 0.05. Differences between species were then assessed using a post hoc test for mulAmerican Journal of Physical Anthropology 396 T.D. SMITH ET AL. Fig. 2. Schematic graph illustrating the spatial overlap of the maxillary sinus or recess (gold) and the dental follicles in the four genera. The anteroposterior space is indicated from left to right, with 0% indicating the sectional level of the ﬁrst histological section in which the deciduous canine appears, and 100% indicating the last section in which M1 appears. The graph illustrates an average of all specimens for each species. The graph is intended to show anteroposterior overlap only. Note that superoinferior separation is exaggerated; the sinus sac is actually more closely adjacent to the dentition, especially at posterior levels. dc, deciduous canine; M1, ﬁrst permanent molar. tiple comparisons between treatments (Siegel and Castellan, 1988, p 213), with differences considered different at P 0.05. American Journal of Physical Anthropology Fig. 3. Parasagittal sections at the level of the mucosal lining of the maxillary sinus (ms) in Saguinus (a), Leontopithecus (b), and Callithrix (c) and the maxillary recess (mr) of Saimiri (d). Note that the paranasal space intervenes between the orbit (or) and the anterior maxillary dentition in all species. Even though orbits appear proportionately large in Saguinus and Callithrix, the posterior part of the sinus occupies a space that is relatively uncrowded compared to the other species, due to relatively smaller premolar follicles. However, in Leontopithecus, the sinus appears to be the most expansive, perhaps due to proportionately small orbits (Fig. 3b). dc, deciduous canine; dp2, ﬁrst deciduous premolar; dp3, second deciduous premolar. Scale bars, 1 mm. The background in low-magniﬁcation images (11a, c, e, g) was digitally altered, because the lighting was unbalanced; portions of the image containing tissue were not altered. MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS 397 Fig. 4. Micro-CT reconstruction showing the facial skeleton of Saguinus (a), Leontopithecus (b), and Callithrix (c) and Saimiri (d) in the frontal view. Note the proportionally small interorbital distance in Saimiri (d). Also note that the greater degree of frontation of the zygomatic bones (Z) in Saimiri. M, maxilla; F, frontal bone. Scale bars, 1 mm. RESULTS Relative position of the maxillary sinus or recess to dentition Figure 2 shows the position of the maxillary sinus or recess, speciﬁcally, the mucosal sac, relative to dental level in the four species of primates. Anteroposterior space of the deciduous maxillary dentition, from the ﬁrst serial histological section containing the deciduous canine to the last section containing M1, establishes the limits of the space shown in the graph. This ﬁgure graphically depicts the sinus sac above the maxillary teeth for clarity, whereas the actual position of the mucosal sac is superomedial or medial to the dental follicles (see below). In all callitrichines, the ostium of the sinus overlaps dc and dp2, and the enclosed sinus completely overlaps dp2 and dp3 in anteroposterior extent. The level of dp4 is partially overlapped in anteroposterior space, especially in Leontopithecus, where the sac extends to overlap approximately one-half of this dental follicle. In contrast, the mucosal sac of the maxillary recess in Saimiri is restricted to the anteroposterior space occupied by dp2 and dp3. Figure 3 depicts parasagittal sections at the level of the maxillary sinus or recess in the four primate species. The anterior portion of the sinus in the callitrichines (Fig. 2a–c) is oriented in a posterior and obliquely inferior direction from its ostium (ostium, not shown, is at anterior limit) toward dp3. In this respect, this portion of the sinus is positioned similarly to the entire maxillary recess in Saimiri (Figs. 2 and 3d). At the level of anterior part of dp3, the maxillary sinus of all callitrichines changes orientation to follow the palatal plane (Figs. 2 and 3a–c). Interrelationship of the orbit, nasal fossa, and dentition in the tamarins, marmoset, and squirrel monkey Micro-CT reconstructions of skulls of the four primate species, in anterior view, are shown in Figure 4. This view reveals differing interorbital proportions among species, appearing to be relatively wide in tamarins in relation to the overall facial breadth (Fig. 4a,b), proportionately more narrow in Callithrix (Fig. 4c) and narrowest in Saimiri (Fig. 4d). Figure 5, also in an anterior view, shows the same specimens cross-sectioned at the level of dp2. At this level, the osseous maxillary sinus/ recess has a wide opening into the middle meatus in all species. The ﬂoor of the maxillary recess in Saimiri is positioned at same plane as the orbital ﬂoor (Fig. 5d), whereas it extends more inferiorly in callitrichines, nearly reaching the same horizontal plane as the ﬂoor of the nasal fossa in Leontopithecus (Fig. 5b). Figure 6 shows the same specimens cross-sectioned at the level of dp3. At the level, the sinus cavities are proportionately American Journal of Physical Anthropology Fig. 5. Micro-CT reconstruction showing the facial skeleton of Saguinus (a), Leontopithecus (b), Callithrix (c), and Saimiri (d) in the frontal view cut at the level of the ﬁrst deciduous premolar (dp2). The position of the maxillary sinus or recess (white arrows) is shown near its ostium. The inferior meatus is indicated (black arrows). The sinus is most inferiorly expanded in Leontopithecus. Scale bars, 1 mm. Fig. 6. Micro-CT reconstruction showing the facial skeleton of Saguinus (a), Leontopithecus (b), Callithrix (c), and Saimiri (d) in the frontal view cut at the level of the second deciduous premolar (dp3). The position of the maxillary sinus or recess (white arrows) is shown. The inferior meatus is indicated (black arrows). The sinus is most inferiorly expanded in Leontopithecus. The maxillary recess or Saimiri is mostly above horizontal plane of dp3 and above the ﬂoor of the nasal fossa (see inferior meatus). The sinus in Saguinus and Callithrix is parallel to the ﬂoor of the inferior meatus. Scale bars, 1 mm. MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS 399 Figures 6 and 7 emphasize differing orbital proportions and interorbital dimensions at the level of dp3. Qualitatively, the orbits appear proportionately smallest, and the interorbital breadth is proportionately greatest in Leontopithecus (Figs. 6b and 7a). Orbits are proportionately larger, and the interorbital breadth is proportionately smaller in Callithrix (Figs. 6c and 7b) and Saimiri (Figs. 6d and 7c). However, relative to biorbital breadth, the bialveolar breadth appears greater in callitrichines than Saimiri (Figs. 6 and 7). Figure 8 shows the relationship of the orbits and nasal cavity in the four species at a horizontal plane approximately mid-way through the orbit. This view conﬁrms visual impressions from the anterior view (see Fig. 4) showing that the nasal cavity is proportionally widest in the tamarins (Fig. 8a,b) compared to Callithrix (Fig. 8c) and Saimiri (Fig. 8d). The position of the orbits relative to the nasal cavity and orbital orientation also differs among species. In Callithrix (Fig. 8c) and especially Saimiri (Fig. 8d), the orbits encroach on the posterior nasal cavity to a greater degree than in the tamarins (Fig. 8a,b). However, the orientation of the orbital cones is more lateral in all callitrichines (Fig. 8a–c) compared to Saimiri (Fig. 8d). Figure 9 shows a horizontal section through the perinatal skulls of Saguinus (Fig. 9a) and Saimiri (Fig. 9b) closer to the ﬂoor of the orbit. Here, the ﬂoor of the cavity for the maxillary sinus or recess is shown medial to the orbit. The relatively large orbit of Saimiri compresses the paranasal space posteriorly. As a result, the maxillary recess of Saimiri is constricted posteriorly and oriented obliquely in a medial direction from its anterior to posterior limit (Fig. 9b). In Saguinus, the maxillary sinus is more nearly parasagittal in its orientation (Fig. 9a). Comparative development of the maxillary dentition and alveolar bone Fig. 7. Micro-CT reconstruction showing the facial skeleton of Leontopithecus (a), Callithrix (b), and Saimiri (c) in the posterior view cut at the level of the second deciduous premolar (dp3) and tilted inferiorly. The position of the maxillary sinus or recess (white arrows) is shown (in Saimiri, this level is just posterior to the recess, but its position is indicated). This view emphasizes the greater degree of encroachment of the orbits on the nasal fossa and potential paranasal space. Scale bars, 1 mm. most expanded in tamarins (Fig. 6a,b). Inferiorly, the sinus cavity reaches the same plane as the ﬂoor of the nasal fossa in all callitrichines (Fig. 6a–c). In Saimiri, the maxillary recess is positionally consistent with the sinuses of the callitrichines at this level, but is less expanded inferiorly toward the level of the nasal fossa ﬂoor (Fig. 6d). Micro-CT reconstructions revealed that, in all species, the most anterior maxillary teeth are partially erupted through alveolar bone, whereas posterior maxillary teeth, especially dp4 and M1, are positioned more deeply within alveolar crypts or had no detectable bone immediately surrounding them (see Fig. 10). The cusps of the deciduous canine and dp2, in particular, are partially erupted in all species (Figs. 5 and 7). Across the entire maxillary tooth row, deciduous eruption was most advanced in Callithrix (Figs. 5c, 6c, and 7b). The extent of ossiﬁcation of the posterior portion of the maxillary bone appears to be more progressive in Leontopithecus and Saimiri (Fig. 10b,d) compared to Saguinus and Callithrix (Fig. 10a,c). Micro-CT reconstructions indicate that the maxillary teeth are least mineralized in Saguinus in that the cusps of dp3 and dp4 are not united, and only a single cusp of M1 is mineralized (Fig. 10a). The other species show a greater degree of mineralization between cusps of dp3 and dp4. Saimiri is remarkable for the extent of mineralization of M1, which has two or three cusps partially mineralized (Fig. 10d). Although the threshold selected for micro-CT reconstructions may not permit detection of some poorly mineralized bone, histological preparations support these observations. Figure 11 shows sections through the parasagittal plane that transect all maxillary tooth follicles. These trichrome-stained preparations clearly illustrate American Journal of Physical Anthropology 400 T.D. SMITH ET AL. Fig. 8. Micro-CT reconstruction showing the facial skeleton of Saguinus (a), Leontopithecus (b), Callithrix (c), and Saimiri (d) in the superior view cut approximately at the superoinferior mid-level of the orbit. Note that the inferior contour of the orbit, from the maxilla to the zygomatic bones, is oriented more laterally in the callitrichines compared to Saimiri. This suggests a greater degree of convergence of the orbit in Saimiri. Scale bars, 1 mm. that maxillary bone barely extends to the level of dp4 in Saguinus and Callithrix (Fig. 11a,b,e,f). In Leontopithecus and Saimiri, alveolar bone is beginning to surround the anterior extent of M1 (Fig. 11c,d,g,h). Distribution of osteoblasts and osteoclasts along orbital, nasal, and alveolar surfaces of the maxillary bone and orbital surface of the ethmoid bone At all dental levels, the orbital surface of the maxilla is lined with rows of osteoblasts in all callitrichines (Fig. 12a–e), except for local areas of resorption associated with the infraorbital neurovascular bundle (Table 2). In Saimiri, in contrast, the orbital surface of the maxilla is lined with osteoblasts at the level and dp2 and, to some extent, at dp3 (Table 2). However, at dp4 and M1 levels, the orbital surface of the maxilla has scattered osteoclasts and Howship’s lacunae (Fig. 12f,g and Table 2). The outer (orbital) surface of the nasal fossa is formed by the maxillary bone, inferomedially, and more superiorly by the orbital surface of the ethmoid lateral plate. In all species, a small extent of the ethmoid lateral plate is still cartilaginous or is ossifying. Inferomedially, the orbital surface of the lateral nasal wall is lined with American Journal of Physical Anthropology osteoblasts in all callitrichines (Fig. 12a and Table 2). In Saimiri, this region is lined with osteoblasts at dp2, but has mostly osteoclasts and Howship’s lacunae at more posterior levels (Fig. 12f and Table 2). The medial wall of the orbit is formed by the ethmoid lateral plate (anteriorly) or mesethmoid alone. In Saimiri, the medial orbit was composed of only the mesethmoid at the M1 level; at dp4 level, the lateral ethmoid is reduced. In most specimens of Callithrix and Saguinus, the medial orbit is composed of mesethmoid alone at M1 level. The medial orbit is mostly lined with osteoblasts in Leontopithecus and Callithrix, but has more osteoclasts posteriorly in Saguinus (Table 2). In Saimiri, the medial orbit is only lined with osteoclasts and Howship’s lacunae at posterior dental levels (Table 2). Distribution of bone cells along the medial alveolar margins of maxillary tooth crypts is mostly similar in Saguinus, Leontopithecus, and Saimiri. Foci of osteoclasts or rows of osteoblasts are observed at all sectional levels except M1, which has little or no alveolar bone surrounding it in most specimens. Osteoclasts appear to frequently occur opposite medial swellings of the dental follicles. Osteoclasts are scattered along the superior and lateral margins of alveolar crypts in Saguinus, Leontopithecus, and Callithrix. In Callithrix, the medial alveolar walls of the deciduous premolars are nearly uniformly MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS Fig. 9. Micro-CT reconstruction showing the facial skeleton of Saguinus (a) and Saimiri (b) in the superior view cut just above the level of the ﬂoor of the nasal fossa. The position of the maxillary sinus or recess (white arrows) is shown. The osseous root of the maxilloturbinal, separating the maxillary sinus or recess from the inferior meatus, is indicated (black arrows). Note the sinus is roughly parallel to the nasal fossa in Saguinus. The maxillary recess of Saimiri is compressed and more diverted in a medial direction as it extends posteriorly. Scale bars, 1 mm. lined with osteoblasts except at M1 (the maxilla does not extend to this level). Saimiri is distinct from all callitrichines in having mostly osteoblastic surfaces of bone along the superior limit of the alveolar crypts (Fig. 12g). Quantitative results Kruskal–Wallis tests indicate signiﬁcant (P \ 0.05) differences among species for relative dental follicle size for nearly all comparisons (Table 3). The only exception is the size of M1 relative to cranial length ratio (P [ 0.05). For the ratios of cube root of dental follicle volume to palatal length, post hoc tests (Table 3) indicate that these results are mainly driven by the large size of the dc, dp2, and dp3 in Leontopithecus compared to one or both of the other callitrichines (but not Saimiri). No signiﬁcant differences were detected between species in a post hoc test for dp4. The M1-palatal length ratio is largest in Saimiri; post hoc comparison shows that this measurement is signiﬁcantly (P \ 0.05) larger in Saimiri 401 than Callithrix. When related to cranial length, cube root of dental follicle volumes reveals a somewhat different pattern. Post hoc tests indicate signiﬁcantly (P \ 0.05) larger dc, dp2, dp3, and dp4 relative to cranial length in Leontopithecus compared to Saimiri. Figure 13 graphically depicts the ﬁndings listed in Table 3. Overall, a similar trend in relative dental follicle size is suggested by all graphs for dc, dp2, and dp3. That is, Leontopithecus has large dental follicles relative to palatal or cranial length. Relative to palatal length, the trend in dental follicle size breaks down posteriorly, where Saimiri equals or exceeds (M1) Leontopithecus in relative size. Relative to cranial length, the trend is markedly different at M1, where Leontopithecus is more similar in relative M1 size to the other callitrichines. The change in trend is graphically depicted in Figure 14. Overall, Figures 13 and 14 indicate that the size of the dentition in Saimiri is much greater relative to palatal length than it is for cranial length. Kruskal–Wallis tests reveal signiﬁcant (P \ 0.05) differences in HA density of the alveolar bone and the tooth cusps from dc to dp4 (no comparisons were made of M1, because the not all species had mineralized cusps in M1). HA density was greatest in Saimiri for all cusps and least in Callithrix (Table 3 and Fig. 15). Post hoc tests show that Saimiri and Callithrix were signiﬁcantly different from each other at each tooth level (Table 2). Kruskal–Wallis tests reveal signiﬁcant (P \ 0.05) differences in palatonasal indices at dp3, dp4, and M1, but no signiﬁcant difference (P [ 0.05) at dp2 (Table 3). Post hoc tests reveal that Saguinus has a signiﬁcantly greater palatonasal index compared to Leontopithecus at dp3, and Saimiri had a signiﬁcantly lower palatonasal index compared to Callithrix at M1. Although few significant differences are detected by the post hoc comparisons, Figure 16 shows certain trends in the palatonasal indices. Aside from dp2 and M1 levels, Saguinus and Callithrix have a larger palatonasal index compared to Leontopithecus and Saimiri. That is, the former had wider palatal widths relative to nasal width. From dp4 to M1, this ratio became progressively lower in Saimiri compared to the callitrichines (Table 2 and Fig. 16). That is, the nasal cavity has greater overlap with the palate. Kruskal–Wallis tests reveal signiﬁcant (P \ 0.05) difference in bizygomaticonasal indices at dp4 and M1 (Table 3). Figure 17 shows the indices at the two levels. At dp4, Saguinus and Callithrix are seen to overlap in the range of standard error, whereas Leontopithecus falls below and Saimiri falls above this range. At M1, Saimiri and Callithrix have the highest indices. Leontopithecus and Saguinus have lower indices, with overlapping ranges. Post hoc tests reveal that Saimiri has a signiﬁcantly greater bizygomaticonasal index compared to Leontopithecus at both dp4 and M1 levels. Note that Saimiri and Leontopithecus have a consistent difference at both dental levels: the former has greater bizygomatic width relative to nasal width. Bizygomatic–interorbital indices were highly variable in all species except Leontopithecus (Table 3). This may be due to variability in the ethmoid bone in Callithrix, Saguinus, and Saimiri at dp4 and M1. In these species, the lateral mass of the ethmoid bone was diminutive (and absent at M1), whereas the lateral mass is prominent at these dental levels in Leontopithecus. Despite the variability in most species, Kruskal–Wallis tests reveal signiﬁcant (P \ 0.05) difference in bizygomatic– American Journal of Physical Anthropology 402 T.D. SMITH ET AL. Fig. 10. Micro-CT reconstruction showing the facial skeleton of Saguinus (a), Leontopithecus (b), Callithrix (c), and Saimiri (d) in the inferior view, showing the extent of mineralization of the maxillary dentition. The dentition appears most mineralized in Saimiri and Leontopithecus and least mineralized in Saguinus. M1 was most mineralized in Saimiri. Scale bars, 1 mm second deciduous premolar; dp4, third deciduous molar. Scale bars, 1 mm. dc, deciduous canine; dp2, ﬁrst deciduous premolar; dp3, second deciduous premolar; dp4, third deciduous molar. Scale bars, 1 mm. interorbital indices at dp4 and M1 (Table 3). At dp4, Leontopithecus has a low index compared to the other species, indicating a wide separation between the orbits. At M1, the bizygomatic–interorbital index of Saimiri remains highly variable based on standard deviation (Table 3), but is larger than all callitrichines (see Fig. 17). Post hoc tests reveal that Saimiri has a signiﬁcantly greater bizygomatic–interorbital index compared to Leontopithecus at the M1 level; the same trend is seen at dp4, though it is not signiﬁcantly different (Table 3 and Fig. 17). DISCUSSION Recent studies have sought a clearer understanding of the process of pneumatic expansion in terms of growth (Koppe and Nagai, 1997, 1999; Rossie, 2006) and microanatomical changes in the mucosa and cartilage or osseous boundaries (Witmer, 1997; Smith et al., 2005, 2008). Evidence on the nature of pneumatization has clear beneﬁts for surgical treatment for pathologies and some congenital disorders of the maxillary dentition and midface, for instance, regarding remodeling subsequent to dental extractions (Sarnat, 1997). In addition, examining the interrelationship of developing paranasal spaces and surrounding structures may explain differing patterns of pneumatization or the lack of pneumatization among different anthropoids. Thus, a ﬁrm developmental understanding informs phylogenetic interpretations (Rossie, 2008). Evidence that the enlargement of paranasal spaces is linked closely to growth of speciﬁc craniofacial structures is supported by some, but not all, studies (Rae and Koppe, 2004; Rossie, 2006). That the dentition obstructs maxillary sinus expansion in a localized manAmerican Journal of Physical Anthropology ner is suggested by all species in which the sinus and dentition interface directly (Koppe et al., 1995; Koppe and Nagai, 1999). However, Rae and Koppe (2004) note that tooth follicles are ‘‘exposed’’ in the maxillary sinus ﬂoor in some but not all macaque species. In addition, Koppe et al. (1996) found no clear association between palatal form and maxillary sinus morphology in orangutans. The scaling of maxillary sinus size relative to body size also differs among anthropoids (Rae and Koppe, 2000; Koppe et al., 2005). Based on variability in scaling, and lack of association with speciﬁc midfacial morphologies, Koppe and colleagues (Koppe and Nagai, 1999; Koppe et al., 1999) doubt that pneumatic expansion can be solely explained by an architectural or structural theory. The structural hypothesis and constraint of pneumatization Recently, we have suggested that valid tests of structural hypotheses may require examination of earlier stages of ontogeny (Smith et al., 2010). The hypothesis that sinuses primarily have an architectural or structural role as opposed to a physiological function has been articulated by numerous authors, but not always with the same precise meaning (e.g., Proetz, 1922; Weidenreich, 1924; Moss and Young, 1960; Preuschoft et al., 2002). Among a list of numerous functions, the hypotheses that pneumatic spaces, either serve to maximize strength of bones with minimal materials or are involved in weight reduction (Witmer, 1997), are the most purely ‘‘structural’’ in that they exclude primary roles in biomechanics or other physiological demands. Sherwood (1999) regards these two functions as interrelated. Here, we refer to bone mass reduction/maximum strength with MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS Fig. 11. Parasagittal sections at the level the maxillary tooth row in Saguinus (a, b), Leontopithecus (c, d), Callithrix (e, f), and Saimiri (g, h). Note that alveolar bone of the maxilla is absent at the level of the third deciduous premolar (dp4) and M1 in Saguinus and Callithrix. Osteogenesis of the alveolar portion of the maxillary bone (arrows) at these levels is more advanced in Leontopithecus and Saimiri. dc, deciduous canine; dp2, ﬁrst deciduous premolar; dp3, second deciduous premolar; or, orbit. Scale bars, a, c, e, f, 1 mm; b, d, f, h, 150 lm. The background in low-magniﬁcation images (a, c, e, g) was digitally altered, because the lighting was unbalanced; portions of the image containing tissue were not altered. [Color ﬁgure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] minimal material as sinus functions postulated in the ‘‘structural hypothesis.’’ Pneumatization is viewed as an active growth process by many investigators (Koppe et al., 1994, 1999; Witmer, 1995, 1997, 1999; Sherwood, 1999; Farke, 2008). Witmer (1995, 1999) asserts that pneumatic mucosae opportunistically progress in the wake of an ‘‘osteclastic front’’ (1999, p 22), and observations of the earliest stages of maxillary sinus pneumatization in primates support this claim (Smith et al., 2005). According to some authors, this reﬂects an obligatory growth process. For example, Koppe and Nagai (1999) suggest that the pneumatic cavities have their ‘‘own developmental potential.’’ Conceptually, this recalls Moss and Greenberg’s (1967) concept of a ‘‘pneumatic skeletal unit.’’ In this view, the sinus mucosa is a functional matrix, a soft tissue structure that exerts an epigenetic inﬂuence on the bones that it borders (i.e., the skeletal unit). 403 In Witmer’s (1997) ‘‘epithelial’’ hypothesis, an idea relating to the mechanism of pneumatization at the tissue level, the sinus mucosa expands opportunistically. Paired with the structural hypothesis, as articulated earlier, this implies that the bone is resorbed by pneumatic mucosa wherever possible. This, in turn, means that there are limits to pneumatization, and these limits appear to include sutural boundaries [see Sherwood (1999)], adjacent parts of teeth or tooth follicles (Smith et al., 2005), or adjacent functional matrices (Schaeffer, 1920; Rossie, 2003; Smith et al., 2010). The cranial functional matrices, dentition, and their supporting skeletal units may then be viewed as speciﬁc regional elements that may either ‘‘permit’’ or constrain pneumatic expansion (Rossie, 2003; Nishimura et al., 2005; Smith et al., 2005, 2010). If grounded in functional matrix theory, the hypothesized ‘‘constraint’’ of sinus expansion by orbital, dental, neural, or other tissues (Smith et al., 2005, 2010; Rossie, 2006) could include primacy of certain matrices. Rossie’s (2006) view that the lack of pneumatization in some New World monkeys (e.g., Saimiri and Cacajo) may be related to extremes in approximation of the orbits or palatal proportions can be rooted in growth patterns of soft tissues or dentition that border the paranasal spaces. For instance, the maxillary dentition exerts localized inﬂuence on paranasal spaces as individual dental follicles form osseous capsules (Marks and Schroeder, 1996); erupting teeth are considered osteogenic structures as well (Schroeder et al., 1992). Some of the alveolar bone formed around teeth or tooth follicles also form part of the lateral and/or inferior wall of the maxillary sinus, thus sharing portions of the hypothetical ‘‘pneumatic’’ skeletal unit. Here, we demonstrate that palatal and nasal proportions and relative tooth size differ signiﬁcantly among several species of New World monkeys. Although sample sizes were likely too small to detect all differences, post hoc tests did not support a hypothesis that relative tooth size was signiﬁcantly larger in Saimiri compared to all the callitrichines. Indeed, follicles of deciduous teeth are relatively larger or nearly as large in Leontopithecus compared to Saimiri, especially relative to cranial length (although cranial length may be a poor proxy for cranial size because of the extreme dolichocephaly exhibited by Saimiri). However, Figures 13 and 14 show a slight shift in the anteroposterior tooth follicle size gradient among species. All species show a similar anteroposterior trend in relative size, with large deciduous canines, gradually increasing deciduous premolar size and a smaller M1. The decreasing gradient of relative size of the teeth, in particular, would likely be associated with more ‘‘unneeded’’ bone in the callitrichines compared to Saimiri. This is in keeping with reports on older-aged specimens of each species. The sinus eventually extends beyond M1 in Saguinus (Smith et al., 2010), Leontopithecus (Rossie, 2006), and overlaps M1 in Callithrix (Nishimura et al., 2005; Smith, unpublished observations). Saimiri diverges from the callitrichines at dp4 and especially M1 (see Fig. 14). Results on HA density establish that the anteroposterior pattern of mineralization is likewise different among the species and suggest that the pace of mineralization in the posterior maxillary teeth may be more rapid in Saimiri compared to callitrichines (see Fig. 15). A comparatively accelerated rate of development at dp4 and M1 could easily explain our previous observation that the maxillary recess of Saimiri fails to expand beyond the level of P4 postnatally (Smith et al., 2010). The relaAmerican Journal of Physical Anthropology 404 T.D. SMITH ET AL. Fig. 12. Distribution of osteoclasts and osteoblasts (Ob) as seen in coronal sections of Saguinus (a, inset), Leontopithecus (b–d), Callithrix (e), and Saimiri (f, g) at the level of the last deciduous premolar. In the callitrichines, the inferomedial surfaces of the orbit were mostly depository on the orbital side (double arrows, rows of osteoblasts). In the contralateral (i.e., nasal), surfaces had numerous osteoclasts (open arrows) and ‘‘scalloped’’ surfaces. In Saimiri, the inferomedial surface of the orbit was resorptive on the orbital surface and depository on the nasal side (inset, f). In the ﬂoor of the orbit, the orbital surface of the maxilla was resorptive as well (g) with deposition on the alveolar side (see inset). CR, remnants of the cartilaginous nasal capsule; Gl, glands; M, maxilla; NC, nasal cavity; NS, nasal septum; Oc, osteocytes. Scale bars: (a) 1 mm (inset 20 lm); (b) 1 mm (inset 200 lm); (c) 200 lm; (d) 20 lm; (e) 1 mm (inset 200 lm); (f) 1 mm (inset 200 lm); (g) 1 mm (inset 200 lm). [Color ﬁgure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] American Journal of Physical Anthropology MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS Fig. 12. (Continued) tively large size of the deciduous maxillary teeth in Leontopithecus appears to weaken this argument, given that the anteroposterior and superoinferior extent of the maxillary sinus is most extreme in this species at birth (Figs. 2 and 5). Midfacial spatial characteristics, in concert with relative size of dp4 and M1, offers a better explanation for the lack of pneumatization in Saimiri. Micro-CT recon- 405 structions suggest that the orbits are less convergent in all callitrichines than Saimiri at this early stage (see Fig. 8), in accord with the greater orbital convergence reported in adult Saimiri compared to callitrichines (Ross, 1995). Hartwig (1995) suggests that orbital approximation is particularly extreme at birth in Saimiri. Thus, the relatively advanced neural development of infant squirrel monkeys (Hartwig, 1995; Leigh, 2004) may amplify orbital encroachment toward midline structures and spaces. The different extent of nasal fossa compression between Saimiri and callitrichines is most apparent from the levels of dp3 to M1 (see Figs. 6–8). Bizygomatic–interorbital indices support the idea that the orbit compresses the nasal fossa to a differing degree among species at the level of M1 (see Fig. 17). Collectively, midfacial indices reveal complexity among the species. Palatonasal indices indicate that position of the tooth follicles is closely adjacent to the nasal fossa at all levels in Saimiri, whereas the palatonasal indices increase in all callitrichines from dp3 to M1 levels. Bizygomaticonasal indices show that the proportions of the nasal cavity relative to overall facial width is most different between Saimiri and Leontopithecus, and bizygomatic–interorbital ratios reveal that orbital contents compress the nasal region to a greater extent in Saimiri than Leontopithecus at M1. Although Callithrix has a relatively compressed nasal fossa at M1, it has the lowest palatonasal index at this level. Taken together, our ﬁndings suggest that large relative tooth size alone is not sufﬁcient to prevent pneumatization. The different developmental outcomes in Leontopithecus and Saimiri are best explained by differences in proximity of the orbit and tooth to the nasal region (which, in turn, is dictated, in part, by size of the orbit and teeth). Orbital cross-sectional area or volumes are factors that we were unable to assess, because the orbits were partially damaged in many of the specimens. Qualitatively, the orbits of Leontopithecus appear relatively to be the smallest of the four species, while largest in Saimiri. Other architectural differences between the two taxa, such as the greater vertical separation of the orbits and palate in Leontopithecus (see Figs. 6 and 7), may offer an additional explanation. More robust sample sizes or alternative methods, such as three-dimensional morphometric comparisons, are needed to conﬁrm these suggestions. The constraining effect of structures such as the dentition and alveolar bone on the maxillary sinus has been discussed many times regarding humans (Schaeffer, 1920). Clinical data on humans support the idea that this is largely due simply to adjacency, as suggested by Koppe and Nagai (1999). For example, pneumatic expansion of the maxillary sinus ensues following extraction of maxillary dentition [see (Sarnat, 1997), and references therein]. Moreover, distraction of alveolar bone can be associated with enlargement of the nasal airways (Cappellette et al., 2008), which may indirectly affect paranasal spaces as well. Differences among the New World monkeys in the perinatal period show how extremes in patterns of pneumatic expansion can be explained by positional factors. The osseous components of any given skeletal unit may relate to more than one functional matrix (Moss and Greenberg, 1967), and therefore myriad skeletal growth outcomes may be related to the interplay of growth patterns among different functional matrices. American Journal of Physical Anthropology American Journal of Physical Anthropology 0.162/0.006* 0.133/0.005* 0.142/0.007*y 0.144/0.007 0.087/0.007 0.047/0.002* 0.038/0.001* 0.04/0.001* 0.042/0.002* 0.024/0.001 1168.04/45.07 1149.99/30.0 1138.44/62.85 968.93/20.96 1.4/0.09 1.33/0.06* 1.43/0.122 1.53/0.15 3.41/0.1* 3.65/0.19* 15.69/7.41 44.74/33.50* 1173.33/96.43 1085.75/63.06 1002.48/68.6 916.86/21.36 1.43/0.11 1.63/0.06* 1.57/0.06 1.54/0.07 3.88/0.15 3.83/0.21 82.22/65.59 73.52/48.72 M/SD 0.144/ 0.008 0.117/0.01 0.121/0.007* 0.126/0.015 0.076/0.015* 0.04/0.004 0.032/0.003 0.033/0.001 0.035/0.004 0.021/0.004 M/SD Leontopithecus rosalia (lion tamarin) 59.11/58.85 83.54/50.72 3.93/0.21 4.43/0.25 1.38/0.11 1.52/0.16 1.64/0.15 1.76/0.1* 935.98/96.43* 822.84/116.96* 839.07/108.66* 792.95/120.5* 0.13/0.012* 0.108/0.01* 0.111/0.01y 0.126/0.01 0.08/0.013 0.036/0.038 0.03/0.003 0.03/0.004 0.034/0.003 0.021/0.003 M/SD 90.56/71.51 216.09/108.58* 4.52/0.29* 4.33/0.18* 1.30/0.06 1.39/0.12 1.35/0.2 1.35/0.08* 1267.94/98.12* 1208.75/124.3* 1181.79/74.84* 1163.23/143.17* 0.152/0.008 0.124/0.005 0.13/0.006 0.145/0.008 0.122/0.01* 0.034/0.002* 0.027/0.002* 0.029/0.002* 0.032/0.003* 0.025/0.004 M/SD Species (common name) Saimiri boliviensis (squirrel monkey) M, mean; SD, standard deviation; *,y, pairs of means that are signiﬁcantly (P \ 0.05) different according to post hoc testing. Tooth ratios dc-PL dp2-PL dp3-PL dp4-PL M1-PL dc-CL dp2-CL dp3-CL dp4-CL M1-CL HA density dc dp2 dp3 dp4 Palatonasal indices Dp2 PN index DP3 PN index Dp4 PN index M1 PN index Bizygomaticonasal indices Dp4 ON index M1 ON index Bizygomatic–interorbital indices Dp4 ON index M1 ON index Variable Saguinus oedipus (cotton-top tamarin) Callithrix jacchus (common marmoset) TABLE 3. Mean (6standard deviation) dental and midfacial variables in four species of monkeys with results of statistical analysis X2 5 8.31, P \ 0.05 X2 5 8.44, P \ 0.05 X2 5 11.44, P \ 0.05 X2 5 11.95, P \ 0.01 X2 5 5.06, NS X2 5 10.28, P \ 0.02 X2 5 7.98, P \ 0.05 X2 5 12.74, P \ 0.01 X2 5 9.86, P \ 0.05 X2 5 10.26, P \ 0.02 X2 5 11.66, P \ 0.01 X2 5 11.33, P \ 0.02 X2 5 10.27, P \ 0.02 X2 5 10.35, P \ 0.02 X2 5 12.83, P \ 0.01 X2 5 8.82, P \ 0.05 X2 5 10.83, P \ 0.02 X2 5 10.42, P \ 0.02 X2 5 9.37, P \ 0.05 X2 5 9.86, P \ 0.05 X2 5 8.28, P \ 0.05 X2 5 5.95, NS Kruskal– Wallis test 406 T.D. SMITH ET AL. MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS 407 Fig. 14. Means for ratios of the cube-root of dental follicle volume to palatal length (top) and cranial length (bottom), showing the divergence of Saimiri from the callitrichines in relative size of posterior dentition. Error bars, standard error of mean. Fig. 15. Mean hydroxyapatite density of deciduous dentition in the four species. Fig. 13. Species comparison of means for ratios of the cuberoot of dental follicle volume to palatal length (left column) and cranial length (right column). Error bars, standard error of mean. Bone-cell distribution and remodeling ﬁelds Other growth principles may explain the extent of pneumatization of the maxilla, at least in part. In vertical dimensions, for example, the maxillary sinus of humans and some New World primates may be attributed to downward midfacial and anterior orbital growth trajectories (Mauser et al., 1975; Enlow and Hans, 1996; Smith et al., 2005). In humans, the ‘‘ﬂoor’’ of the maxillary sinus ‘‘drops’’ inferiorly, in tandem with resorption that occurs during downward displacement of the maxilla. The ‘‘roof’’ of the maxillary sinus, and shared boundary with the orbit, shifts superiorly, in tandem with deposition on the orbit surface that accompanies forward remodeling of the orbital cones during growth (the ‘‘V’’ principle—Enlow and Hans, 1996). Observations of bonecell distribution along the roof and ﬂoor of the maxillary sinus in several species of callitrichines suggest that the same growth processes may explain vertical expansion of the sinus space (Smith et al., 2005). Here, we conﬁrm earlier observations on callitrichines, but, in Saimiri, orbital remodeling is markedly different. Major differences in the distribution of bone cells along the inferior and medial orbit, in the alveolar region, and in some adjacent regions, are shown in Figure 18. A schematic view of a coronal section through the level of dp4 is shown, with an illustration of distribution patterns detailed in Table 2. This dental follicle level was selected for representation, because this dental level is ultimately the posterior limit for the maxillary recess in Saimiri (Smith et al., 2010). Distribution of osteoclasts (2) and osteoblasts (1) is indicated to denote resorptive and depositional surfaces, respectively. The American Journal of Physical Anthropology 408 T.D. SMITH ET AL. Fig. 16. Palatonasal indices in the four species from the level of the ﬁrst deciduous premolar (dp2) to M1. Fig. 17. Bizygomaticonasal (left) and bizygomatic–interorbital (right) indices in the four species from the level of the last deciduous premolar (dp4) to M1. American Journal of Physical Anthropology MIDFACIAL MICROANATOMY OF PERINATAL MONKEYS Fig. 18. Schematic diagrams showing patterns of osteoblast (1) and osteoclast (2) distribution at the level of the last deciduous premolar in the four species. Note contrast in the inferomedial orbit between callitrichines (a–c) and Saimiri (d). orbital surface of the maxilla was mostly depositional in callitrichines (Fig. 18a–c), but was mostly resorptive in Saimiri (Fig. 18d). At the level of dp4, the medial wall of the orbit is a depositional surface in perinatal Leontopithecus and Callithrix and is resorptive in Saguinus and Saimiri. However, the inferomedial margin of the orbit is depositional at this level in all callitrichines, in contrast to the nearly uniform resorption typical of medial, inferomedial, and inferior surfaces of the maxilla in Saimiri (Fig. 18d). The pattern of a superior shift in the ﬂoor of the orbit in callitrichines closely resembles the prenatal human pattern (Mauser et al., 1975) and, as in humans, is clearly related to vertical expansion of the maxillary sinus. Such a pattern might be regarded as a ‘‘passive’’ mode of sinus enlargement, a descriptor offered by Weidenriech (1924) regarding the very nature of sinuses. Noteworthy in this regard is Enlow and Hans’ (1996) concept of paranasal sinuses as ‘‘leftover ‘dead’ unused’’ (p 143) spaces created as a byproduct of enlargement by neighboring craniofacial bones. In that view, particular boundaries of the maxillary sinus expand outward by following remodeling ‘‘growth ﬁelds’’ of the maxillary bone. Enlow and Hans (1996) did not discount epigenetic effects of soft tissues as important factors in craniofacial growth. The different remodeling patterns seen in Saimiri are most certainly related to our qualitative observations of the degree of perinatal orbital approximation and high-bizygomatic–interorbital indices. The limitations imposed by orbital encroachment thus provide an extreme example of why pneumatization patterns differ so profoundly among New World monkeys (Rossie, 2006). CONCLUSIONS Previously, we suggested orbitonasal proportions and relative size of the dentition factor prominently in pre- 409 venting pneumatic expansion of the maxillary recess in Saimiri (Smith et al., 2010). Here, differences in relative size of dp4 and M1 among species provide a strong explanation for the observation that the maxillary recess in Saimiri fails to expand beyond the level of P4 (Smith et al., 2010). As previously suggested, relative size of the dentition appears not to be the sole explanation. Saimiri is exceptionally different from callitrichines in degree of orbital approximation and in the pattern of orbital remodeling at birth. Overall, it is notable that tooth-palate size ratios show different patterns than tooth-cranial size ratios: compared to the callitrichines, maxillary teeth in Saimiri are large relative to palate but not cranial length (see Fig. 13). These different patterns indicate that the deciduous teeth are large relative to the midface, but not relative to overall head size in perinatal Saimiri. Thus, orbital contents [see Hartwig (1995)] and dentition are more tightly ‘‘packed’’ in the midface of perinatal Saimiri relative to the callitrichines. Although a constraining inﬂuence of the main nasal chamber on the maxillary recess cannot be discounted, preliminary ﬁndings indicate that the surface area of the recess increases isometrically relative to the main nasal chamber, in apparent contrast to disproportionately higher rate of sinus expansion (relative to the nasal fossa) in Saguinus (Smith et al., 2010). Comparisons among other species would bolster the hypothesis that the packing of dental and orbital matrices is the most important factors in constraint of pneumatization, and a comparison of perinatal cercopithecoids (many of which lack true maxillary sinuses) would be most informative. The divergence of M1 from the pattern seen for the other teeth is interesting in light of the different patterns of brain growth seen in Saimiri compared to Saguinus fuscicollis (Leigh, 2004). Brain growth is more rapid in Saimiri, being complete by six months of age, while brain growth is more prolonged in the tamarin. If these neural differences reﬂect a contrast between Saimiri and callitrichines, in general, it suggests that life history may have an indirect inﬂuence on constraint of pneumatic expansion via the high correlation of M1 development and brain growth. Similarly, Rossie (2006) hypothesized that the selection for precocial neonates may be linked to the loss of the maxillary sinus in Saimiri, which likewise suggests the need for investigation of perinatal cercopithecoids. ACKNOWLEDGMENT We thank L.T. Nash for suggestions regarding post hoc statistical tests. LITERATURE CITED Cappellette M, Cruz OLM, Carlini D, Weckx LL, Pignatari SSN. 2008. Evaluation of nasal capacity before and after rapid maxillary expansion. Am J Rhinol 22:74–77. Corner BD, Richtsmeier JT. 1992. Cranial growth in the squirrel monkey (Saimiri sciureus): a quantitative analysis using three dimensional coordinate data. Am J Phys Anthropol 87:67–81. Dennis JC, Smith TD, Bhatnagar KP, Burrows AM, Bonar CJ, Morrison EE. 2004. Expression of neuron-speciﬁc markers by the vomeronasal neuroepithelium in six primates species. Anat Rec 281:1190–1199. Enlow D, Hans M. 1996. Essentials of facial growth. Philadelphia: W.B. Saunders Company. American Journal of Physical Anthropology 410 T.D. SMITH ET AL. Farke AA. 2007. Morphology, constraints, and scaling of frontal sinuses in the Hartebeest Alcelaphus buselaphus (Mammalia: Artiodactyla, Bovidae). J Morphol 268:243–253. Hartwig WC. 1995. Effect of life history on the squirrel monkey (Platyrrhini: Saimiri) cranium. Am J Phys Anthropol 97:435– 449. Koppe T, Nagai H. 1997. Growth pattern of the maxillary sinus in the Japanese macaque (Macaca fuscata): reﬂections on the structural role of the paranasal sinuses. J Anat 190:533– 544. Koppe T, Nagai H. 1999. Factors in the development of the paranasal sinuses. In: Koppe T, Nagai H, Alt KW, editors. The paranasal sinuses of higher primates. Berlin: Quintessence. p 133–149. Koppe T, Moormann T, Wallner C-P,Röhrer-Ertl. 2005. Extensive enlargement of the maxillary sinus in Alouatta caraya (Mammalian Primates, Cebidae): an allometric approach to skull pneumatization in Atelinae. J Morphol 263:238–246. Koppe T, Rae T, Swindler D. 1999. Inﬂuence of craniofacial morphology on primate paranasal pneumatization. Ann Anat 181:77–80. Koppe T, Röhrer-Ertl O, Hahn D, Reike R, Nagai H. 1995. Growth pattern of the maxillary sinus in orangutan based on measurements of CT scans. Okajimas Folia Anat Jpn 72:37– 43. Koppe T, Röhrer-Ertl O, Hahn D, Reike R, Nagai H. 1996. The relationship between the palatal form and the maxillary sinus in orangutan. Okajimas Folia Anat Jpn 72:297–306. Koppe T, Yamamoto T, Tanaka O, Nagai H. 1994. Investigations on the growth pattern of the maxillary sinus in Japanese human fetuses. Okajimas Folia Anat Jpn 71:311–318. Leigh SR. 2004. Brain growth, life history, and cognition in primate and human evolution. Am J Primatol 62:139–164. Maier W. 2000. Ontogeny of the nasal capsule in cercopithecoids: a contribution to the comparative and evolutionary morphology of catarrhines. In: Whitehead PF, Jolly CJ, editors. Old World monkeys. Cambridge: Cambridge University Press. p 99–132. Marks SC, Schroeder HE. 1996. Tooth eruption: theories and facts. Anat Rec 245:374–393. Moss M, Greenberg S. 1967. Functional cranial analysis of the human maxillary bone. Angl Orthod 37:151–164. Moss ML, Young RW. 1960. A functional approach to craniology. Am J Phys Anthropol 18:281–292. Nishimura TD, Takai M, Tsubamoto T, Egi N, Shigehara N. 2005. Variation in maxillary sinus anatomy among platyrrhine monkeys. J Hum Evol 49:370–389. Proetz AW. 1922. Observations upon the formation and function of the accessory nasal sinuses and mastoid air cells. Ann Otol Rhinol Laryngol 31:1083–1100. Preuschoft H, Witte H, Witzel U. 2002. Pneumatized spaces, sinuses and spongy bones in the skulls of primates. Anthropol Anz 60:67–79. Rae TC, Koppe T. 2000. Isometric scaling of maxillary sinus volume in hominoids. J Hum Evol 38:411–423. Rae TC, Koppe T. 2004. Holes in the head. Evolutionary interpretations of the paranasal sinuses in catarrhines. Evol Anthropol 13:211–223. Ross CF. 1995. Allometric and functional inﬂuences on primate orbit orientation and the origins of the Anthropoidea. J Hum Evol 29:201–227. Rossie J. 2006. Ontogeny and homology of the paranasal sinuses in Platyrrhini (Mammalia: Primates). J Morphol 267:1–40. Rossie JB. 2003. Ontogeny, homology, and phylogenetic signiﬁcance of anthropoid paranasal sinuses, Ph.D. Thesis, Yale University. New Haven, CT. Rossie JB. 2008. The phylogenetic signiﬁcance of anthropoid paranasal sinuses. Anat Rec 291:1485–1498. American Journal of Physical Anthropology Sarnat BG. 1997. Postnatal growth of the nasomaxillary complex. In: Dixon AD, Hoyte DAN, Rönning O, editors. Fundamentals of craniofacial growth. Boca Raton: CRC Press. p205– 224. Schaeffer JP. 1920. The nose, paranasal sinuses, nasolacrimal passageways, and olfactory organ in man. Philadelphia, PA: P. Blakiston’s Son and Co. Schroeder HE, Luder H-U, Bosshardt. 1992. Morphological and labeling evidence supporting and extending a modern theory of tooth eruption. Schweiz Monatsschr Zahnmed 102:20–31. Shea BT. 1985. On aspects of skull form in African apes and orangutans, with implications for hominoid evolution. Am J Phys Anthropol 68:329–342. Sherwood RJ. 1999. Pneumatic processes in the temporal bone of chimpanzee (Pan troglodytes) and gorilla (Gorilla gorilla). J Morphol 241:127–137. Siegel S, Castellan NJ. 1988. Nonparametric statistics for the behavioral sciences. New York: McGraw-Hill. Smith T, Siegel M, Mooney M, Burrows A, Todhunter J. 1997. Formation and enlargement of the paranasal sinuses in normal and cleft lip and palate human fetuses. Cleft Palate-Craniofac J 34:483–489. Smith TD, Bhatnagar KP, Bonar CJ, Shimp KL, Mooney MP, Siegel MI. 2003. Ontogenetic characteristics of the vomeronasal organ in Saguinus geoffroyi and Leontopithecus rosalia with comparisons to other primates. Am J Phys Anthropol 121:342–353. Smith TD, Rossie JB, Cooper GM, Carmody KA, Schmieg RM, Bonar CJ, Mooney MP, Siegel MI. 2010. The maxillary sinus in three genera of New World monkeys: factors that constrain secondary pneumatization. Anat Rec 203:91–107. Smith TD, Rossie JB, Cooper GM, Mooney MP, Siegel MI. 2005. Secondary pneumatization in the maxillary sinus of callitrichid primates: insights from immunohistochemistry and bone cell distribution. Anat Rec 285:677–689. Smith TD, Rossie JB, Docherty BA, Cooper GM, Bonar CJ, Silverio AL, Burrows AM. 2008. Fate of the nasal capsular cartilages in prenatal and perinatal tamarins (Saguinus geoffroyi) and extent of secondary pneumatization of maxillary and frontal sinuses. Anat Rec 291:1397–1413. Sperber G. 2000. Craniofacial development. Hamilton: B.C. Decker. Swindler D. 1999. Maxillary sinuses, dentition, diet, and arch form in some anthropoid primates. In: Koppe T, Nagai H, Alt KW, editors. The paranasal sinuses of higher primates. Berlin: Quintessence. p 191–206. Usas A, Ho AM, Cooper GM, Olshanski A, Peng H, Huard J. 2009. Bone regeneration mediated by BMP4-expressing muscle-derived stem cells is affected by delivery system. Tissue Eng A 15:285–293. Weidenriech F. 1924. Über die pneumatischen Nebenräume des Kopfes. Ein Beitrag zur Kenntnis des Bauprinzips der Knochen, des Schädels und des Körpers. Anat Embryol 72:55–93. Witmer LM. 1995. Homology of facial structures in extant archosaurs (birds and crocodilians), with special reference to paranasal pneumaticity and nasal conchae. J Morphol 225:269– 327. Witmer LM. 1997. The evolution of the antorbital cavity of archosaurs: a study in soft-tissue reconstruction in the fossil record with an analysis of the function of pneumaticity. J Vert Paleontol 17:1–73. Witmer LM. 1999. The phylogenetic history of the paranasal air sinuses. In: Koppe T, Nagai H, Alt KW, editors. The paranasal sinuses of higher primates. Berlin: Quintessence. p 21–34. Zollikofer CPE, Weissman JD. 2008. A morphogenetic model of cranial pneumatization based on the invasive tissue hypothesis. Anat Rec 291:1446–1454.