2014 Volume 122 Issue 2 Pages 101-113
Procynocephalus and Paradolichopithecus are large Eurasian papionins from the Middle Pliocene to Early Pleistocene. The two genera are regarded as being phylogenetically close, but their phyletic position is still disputed, in particular regarding to which subtribe, Papionina or Macacina, they are close to. Many fragile structures of the nasal region are well preserved in the type specimen of Procynocephalus wimani from the Xin’an locality in China. Computed tomography scans showed that the Xin’an specimen has no maxillary sinus, an inferior meatus extending medially from the slightly superior portion of the maxillary body, and a thick maxillary body with no maxillary fossa. Morphological variation in the nasal region was surveyed in extant papionins. Our analysis showed that the maxillary sinus is found even in Papio/Theropithecus and that its formation is confirmed for all Macaca. The inferior conchae are suspended from the superior portion of the nasal cavity in Papio/Theropithecus and Mandrillus, and the maxillary fossa is developed by major absorption of the maxillary cancellous bone in Papionina. These findings indicate that a given fossil specimen having a maxillary sinus does not always belong to the Macacina lineage, and that a given specimen having a thin maxillary body is closer to the Papionina. Despite the paucity of evidence definitive of its phyletic position, these morphological examinations suggest that Procynocephalus is closer to the lineage of Macacina though it lacks the maxillary sinus. Whereas Paradolichopithecus arvernensis and Paradolichopithecus sushkini show some morphological similarities to and differences from each other and Procynocephalus, their nasal architecture is similar to that seen in the Macacina rather than in the Papionina. The morphological evaluations of the nasal region in African forms are expected to contribute to our understanding of the phyletic relationships and adaptive radiation of the large Eurasian papionins in the Plio–Pleistocene.
Procynocephalus Schlosser 1924 is a representative of the largest fossil papionins, a tribe of the Papionini, that occurred from the Late Pliocene to the Early Pleistocene of East and South Asia (Delson, 1975; Szalay and Delson, 1979; Pan and Jablonski, 1987; Jablonski, 2002; Takai et al., 2008, 2014). The type species Procynocephalus wimani Schlosser 1924 has been found in China, and its type specimen was discovered from the Early Pleistocene of Xin’an locality, Henan, China (Schlosser, 1924). The specimens from India and Pakistan are usually assigned to other species such as Procynocephalus subhimalayanus von Meyer 1848 or Procynocephalus pinjorii Verma 1969, and the latter is regarded as a junior synonym of the former (Delson, 1975; Szalay and Delson, 1979; Jablonski, 2002).
Another of the largest Eurasian papionins is Paradolichopithecus Necrasov, Samson, and Padulesco 1961 (Delson, 1975; Delson and Nicolaescu-Pliopsor, 1975; Szalay and Delson, 1979; Jablonski, 2002; Takai et al., 2008). Most of the fossil specimens are from the Late Pliocene to the Early Pleistocene of Europe, and are assigned to Paradolichopithecus arvernensis Depèret 1929 or Paradolichopithecus geticus Necrasov et al. 1961 (Delson, 1975; Delson and Nicolaescu-Pliopsor, 1975; Szalay and Delson, 1979; Jablonski, 2002). The type specimen of Para. arvernensis was discovered from the Late Pliocene site of Senèze, France (Depèret, 1929). Outside Europe, two other species are known: Paradolichopithecus sushkini Trofimov 1977 from the Early Pleistocene site of Kuruksay, southern Tajikistan (Trofimov, 1977) and Paradolichopithecus gansuensis Qiu, Deng, and Wang 2004 from the Early Pleistocene of Longdan, Gansu, China (Qiu et al., 2004). This genus is usually regarded as being closely related to Procynocephalus (Delson, 1975; Szalay and Delson, 1979; Takai et al., 2008), and is sometimes regarded as a junior synonym of the latter with few differences in dental morphology and size (Jolly, 1967; Simons, 1970). However, Trofimov (1977) emphasized that there are more sharply ridged molars and a more developed M3 hypoconulid in Paradolichopithecus than in Procynocephalus. Unfortunately, comparable fossil materials are restricted for Procynocephalus (Szalay and Delson, 1979; Takai et al., 2008). Thus, the phyletic relationship between the two genera is still being debated.
The phyletic position of Procynocephalus and Paradolichopithecus is also controversial. There are two subtribes of Papionini: Papionina includes the extant genera Papio, Theropithecus, Lophocebus, Rungwecebus, Mandrillus, and Cercocebus; and Macacina comprises only Macaca (Table 1; Strasser and Delson, 1987; Delson, 2000; Davenport et al., 2006). Procynocephalus has molars with low and rounded cusps, a broad palate, and no maxillary and mandibular fossae (Schlosser, 1924; Szalay and Delson, 1979; Jablonski, 2002). These features support the view that this genus is phyletically closer to the lineage of Macacina (i.e. to macaques) than to the Papionina (Szalay and Delson, 1979; Jablonski, 2002). European Para. arvernensis has many macaque-like features, including a moderately long muzzle, smooth facial profile, and less-developed maxillary and mandibular fossae; it also has some baboon-like features, including postcranial long bones reflecting its terrestrial habits (Szalay and Delson, 1979; Jablonski, 2002). Central Asian Para. sushkini has some characteristics found in modern Papio, including a lacrimal fossa located in the lacrimal bone, moderate maxillary and mandibular fossae, and molars large relative to premolars (Maschenko, 1994; Takai et al., 2008). Thus, the surface features of these extinct taxa do not conform to the distribution of those seen in living papionins.
| Family Cercopithecoidae |
| Subfamily Colobinae |
| Subfamily Cercopithecinae |
| Tribe Cercopithecini |
| Tribe Papionini |
| Subtribe Papionina |
| Papio, Theropithecus, Lophocebus, Rungwecebus, Mandrillus, Cercocebus |
| Subtribe Macanina |
| Macaca |
Contributing to the complexity of this issue, the morphological homogeneity of the specimens assigned to Paradolichopithecus has also been challenged. As described above, differences in the topology of the maxillary fossa between Para. arvernensis and Para. sushkini confuse phyletic interpretations for this genus. In addition, Para. sushkini possesses a maxillary sinus—a hollow in the maxilla (Nishimura et al., 2007)—whereas Para. arvernensis shows no evidence of this feature (Nishimura et al., 2009). Para. gansuensis probably has no maxillary sinus (Nishimura et al., 2010). The maxillary sinus is a paranasal sinus that communicates with the middle meatus of the nasal cavity and pneumatizes the maxilla; this feature is considered to form only in macaques among the extant cercopithecoids including papionins (Koppe and Ohkawa, 1999; Rae et al., 2002; Ito et al., 2009). It is believed that the paranasal sinuses—including the maxillary sinus—were lost in the common ancestor of extant cercopithecoids, and that the maxillary sinus arose again in the crown lineage of macaques (Rae et al., 2002). Thus, such morphological heterogeneities contribute to confusion in arguments about the phylogenetic placement of Procynocephalus and Paradolichopithecus.
The nasal anatomy, including that of the maxillary sinus, has become important for the phylogenetic analysis of fossil nonhuman primates. Knowledge about the morphological variation in extant macaques has contributed greatly to our understanding of the phyletic position of fossil species and the adaptive radiation of this genus in Asia (Rae et al., 2003; Ito et al., 2009, 2014a). Many fragile structures in the nasal region are preserved intact in the type specimen of Pr. wimani (see Results). Unfortunately, knowledge is limited about any morphological variation in the nasal region of Papionina (Koppe and Ohkawa, 1999). The maxillary sinus seems to form partially in the genus Papio (Nishimura et al., 2007). Thus, surveys of the morphological variation in papionins, especially Papionina, are needed to understand the phyletic position of the fossil papionins better.
In this paper we examine the morphological variation in the nasal architecture of extant papionins. We then use this information to diagnose the type specimen of Pr. wimani by comparison with published materials for Paradolichopithecus. We also discuss the phyletic relationships and paleobiogeography of the large Eurasian papionins, including Procynocephalus and Paradolichopithecus.
We examined the type specimen of Pr. wimani from the Xin’an locality, Henan, China (PMU M3652a–d, Table 2, Figure 1), which is currently housed at the Museum of Evolution, Uppsala University, Sweden (PMU). The specimen comprises the posterior parts of the right and left maxillae with M2–3 (PMU M3652a, Figure 1A–D; PMU M3652c, Figure 1E–I), an anterior part of the left upper jaw with a canine and P3 (PMU M3652b, Figure 1J–M), and an isolated left M1 (PMU M3652d, Figure 1N). These are associated with a nearly complete mandible lacking a left M3 (PMU M3513, Figure 1O, P). The similar size and no overlap of parts do not contradict the diagnosis that these maxillary specimens (PMU M3652a–d) belong to the same individual. The fully erupted M3 and a small canine indicate that the maxillary specimens belonged to an adult female individual (Figure 1). The bony surface of the nasal cavity is exposed in the left posterior maxilla (PMU M3652c), but is covered by the matrix in the right (PMU M3652a). The matrix probably filled in the nasal cavity, and it is regarded as an endocast of the intact nasal cavity (Figure 1B–D).
| Species | Specimen* | Locality | Reference |
|---|---|---|---|
| Procynocephalus wimani | PMU M3652a–d** | Xin’an, Henan, China | present study |
| Paradolichopithecus | |||
| P. arvernensis | FSL41336** | Senèze, France | Nishimura et al. (2009) |
| P. sushkini | PIN3120-523**, 524 | Kuruksay, Tajikistan | Nishimura et al. (2007) |
| P. gansuensis | HMV1142** | Longdan, Gansu, China | Nishimura et al. (2010) |
The type specimen of Procynocephalus wimani from Xin’an, Henan, China. (A–D) PMU M3652a, a posterior part of the right maxilla with M2–3; (E–I) PMU M3652c, a posterior part of the left maxilla with M2–3; (J–M) PMU M3652b, an anterior part of the left upper jaw with a canine and P3; (N) PMU M3652d, an isolated left M1; (O, P) PMU M3513, a nearly complete mandible lacking the left M3. The scale is in centimeters.
PMU M3513a–d were scanned separately using a peripheral quantitative computed tomography (pQCT) scanner (XCT Research SA+, Stratec Medizintechnik GmbH, Pforzheim, Germany) at the Primate Research Institute, Kyoto University, Japan. The scans were made with a tube voltage of 50.1 kV, a tube current of 0.504 mA, a voxel of 0.04–0.10 mm, and a slice interval of 0.05–0.22 mm, depending on the size of the subjects.
Morphological variation in the nasal region including the maxillary sinus was examined in extant papionins using helical computed tomography (CT) scanners. The specimens examined here include 27 crania of Papio, 8 of Theropithecus, 3 of Lophocebus, 13 of Mandrillus, 18 of Cercocebus, and 276 of Macaca (Table 3). The specimens used for the scans are housed at the Primate Research Institute, Kyoto University (KUPRI), Japan; the Department of Anatomy (Macro), Dokkyo Medical University (DMU), Japan; the Japan Monkey Centre (JMC); the Tochigi Prefectural Museum (TPM), Japan; the Shikoku Institute of Natural History (SNH), Japan; the Hakusan Nature Conservation Center (HNCC), Japan; the Kunming Institute of Zoology (KIZ), Chinese Academy of Science; the Kanagawa Prefectural Museum of Natural History (KPM), Japan; and the Laboratory of Physical Anthropology, Graduate School of Sciences, Kyoto University (KAS), Japan. They belong to adult or subadult specimens/individuals with M3 or M2 fully erupted, and have few pathological traits in the skulls including the nasal region. Most of the scans are registered on the website of the Digital Morphology Museum (DMM), KUPRI (http://dmm.pri.kyoto-u.ac.jp/archive/).
| Total | Male | Female | Institutions* | |
|---|---|---|---|---|
| Subtribe Papionina | ||||
| Papio spp. (total) | 27 | 11 | 16 | |
| P. hamadryas | 25 | 10 | 15 | PRI |
| P. anubis | 2 | 1 | 1 | PRI |
| Theropithecus gelada | 8 | 5 | 3 | JMC, PRI |
| Lophocebus albigena | 3 | 0 | 3 | JMC, PRI |
| Mandrillus spp. (total) | 13 | 11 | 2 | |
| M. leucophaeus | 4 | 4 | 0 | PRI |
| M. sphinx | 9 | 7 | 2 | DMU, PRI |
| Cercocebus spp. (total) | 18 | 9 | 9 | |
| C. agilis | 1 | 0 | 1 | JMC |
| C. atys | 4 | 3 | 1 | JMC |
| C. galeritus | 5 | 3 | 2 | JMC, PRI |
| C. torquatus | 8 | 3 | 5 | DMU, PRI |
| Subtribe Macacina | ||||
| Macaca spp. (total) | 276 | 156 | 120 | |
| M. sylvanus | 3 | 3 | 0 | JMC, PRI |
| M. nemestrina | 14 | 7 | 7 | DMU, PRI |
| M. silenus | 5 | 4 | 1 | JMC, PRI |
| M. pagensis | 16 | 8 | 8 | PRI |
| M. hecki | 1 | 0 | 1 | PRI |
| M. maura | 7 | 5 | 2 | DMU, JMC, KPM, PRI |
| M. nigra | 5 | 3 | 2 | KPM, PRI |
| M. ochreata | 1 | 1 | 0 | PRI |
| Macaca sp. (Sulawesi) | 1 | 1 | 0 | PRI |
| M. mulatta | 18 | 10 | 8 | DMU, PRI |
| M. fuscata | 109 | 55 | 54 | DMU, KAS, HNCC, JMC, SNC, PRI, TPM |
| M. cyclopis | 20 | 11 | 9 | DMU, KAS, PRI |
| M. fascicularis | 21 | 12 | 9 | DMU, KAS, PRI |
| M. arctoides | 12 | 9 | 3 | DMU, JMC, PRI |
| M. assamensis | 15 | 11 | 4 | PRI |
| M. thibetana | 9 | 7 | 2 | DMU, KIZ, KPM, JMC, PRI |
| M. radiata | 16 | 8 | 8 | PRI |
| M. sinica | 3 | 1 | 2 | PRI |
| Total | 345 | 192 | 153 | |
The nasal architecture was examined on various cross-sectional images reformatted from the serial CT images using OsiriX software (Rosset et al., 2004) and Amira software (version 6.3; FEI Visualization Sciences Group, Burlington, MA, USA). The examinations were performed using specimens of both sexes together, unless any given features showed distinct sexual dimorphism.
Comparisons with the species of Paradolichopithecus were based on the published descriptions (Table 2): the type specimen of Para. arvernensis from Senèze, France, FSL41336, an adult female (Nishimura et al., 2009); the type and another specimens of Para. sushkini from Kuruksay, Tajikistan, PIN3120-523 and 524, an adult female and a juvenile male (Nishimura et al., 2007); and the type specimen of Para. gansuensis from Longdan, Gansu, China, HMV1142, an adult male (Nishimura et al., 2010).
The CT scans visualized the internal morphology of the fossil materials well (Figure 2, Figure 3). The scans of PMU M3652b show that the canine root is short and almost vertical (Figure 2), reflecting that the specimen belonged to a female animal. The fragile structures, including the thin nasal conchae, are well preserved within the attached matrix of the right posterior maxilla (PMU M3652a, Figure 3A–E).
CT images (top) and corresponding line-drawing schemata (bottom) at the canine–P3 level in PMU M3652b, Procynocephalus wimani: (A) sagittal; (B, C) coronal images. The maxillary body is thick at this level (arrows). Abbreviations: P3m and P3d, mesial and distal region of the P3; C, canine; and ir, inferior recess under the canine. The scale is in centimeters.
Coronal CT images and corresponding line-drawing schemata (below) at the level of M2 to the tubercle (from left to right) in (A–E) PMU M3652a; and (F–J) PMU M3652c, Procynocephalus wimani. The outline of the nasal cavity (dotted lines) is well preserved by the matrix attached to PMU M3652a. The middle meatus expands slightly laterally at the M3 level (black arrows). Abbreviations: M2m, M2mid, and M2d, mesial, middle, and distal region of the M2, respectively; M3m and M3d, mesial and distal region of the M3, respectively; lc, lacrimal canal; gpf, greater palatine foramen; ic, inferior concha; im, inferior meatus; sc, superior concha; and zb, zygomatic bone. The scale is in centimeters.
Maxillary pneumatization is not found in PMU M3652a–c, and there are no traces of an ostium opening to the middle meatus (Figure 2, Figure 3). The middle meatus expands slightly laterally at the M3 level so that the medial wall of the maxillary body curves laterally (Figure 3D, E, I, J). This reduces the distance between the middle meatus and the zygomaxillary suture (Figure 3D, E, I, J).
The inferior concha attaches to the slightly superior portion of the maxillary body at the level of M2–3 to the tubercle and extends medially and downward into the nasal cavity (Figure 3A–E). This makes the inferior meatus slightly tall (Figure 3A–E). The lateral wall of the inferior meatus curves laterally and continuously from the rounded floor of the nasal cavity and shows no trace of excavation laterally into the maxillary body at the level of M2–3 to the tubercle (Figure 3).
The distal parts of the inferior and superior conchae are well preserved. The inferior concha expands slightly, whereas the superior concha is thick and expands in a circular shape (Figure 3A–E).
The lacrimal canal travels almost vertically and opens into the inferior meatus at the M2–3 junction level (Figure 3C). The orbital floor is partially preserved at this level, but most parts, including the lacrimal fossa, were lost in PMU M3652a and c (Figure 3).
The maxillary body is continuously thick and is occupied by cancellous bone at the levels of P3 and the M2 to the tubercle (Figure 2, Figure 3), and there is no trace of a maxillary fossa (Figure 1A, E, F, Figure 3).
The zygomatic bone is sutured widely with the maxilla at the M3 level (Figure 1A, E, Figure 3). The recess, a lateral depression, is developed slightly under the canine root at the P3 level (Figure 2B, C), whereas our examinations were inconclusive about this feature at the P4 level.
Comparisons with Paradolichopithecus specimensThe nasal features that are comparable to PMU M3652a–c (hereinafter, the Xin’an specimen) of Pr. wimani are preserved in FSL41336 (hereinafter, the Senèze specimen) of Para. arvernensis and PIN3120-523 and 524 (hereinafter, the Kuruksay specimens) of Para. sushkini, but are restricted in HMV1142 (hereinafter, the Longdan specimen) of Para. gansuensis.
The maxillary sinus is not formed and the middle meatus expands laterally at the M3 level in the Senèze specimen, as seen in the Xin’an specimen. The maxillary sinus is probably lacking in the Longdan specimen. By contrast, the maxillary sinus is distinctive in the posterior portion of the maxilla, which is located almost under the orbital cavity in the Kuruksay specimens, e.g. at the level of distal M2 to the tubercle in the adult female specimen (PIN3120-523).
The inferior concha extends from the slightly superior portion of the maxillary body to make the slightly tall inferior meatus in the Senèze specimen, as seen in the Xin’an specimen. We note that Nishimura et al. (2009) misidentified a part of the medial surface of the lacrimal canal as a basal portion of the inferior concha, which is labeled ‘ic’ (representing the inferior concha) in Figure 3a. The inferior meatus extends from the middle portion at the premolar level in the Kuruksay specimens. The severe destruction and distortion provide little comparable material for the analysis of this feature at the molar level, but the formation of the maxillary sinus indicates that the inferior concha extends from the middle portion to secure the opening of the ostium of the maxillary sinus in the middle meatus in the Kuruksay specimens. The inferior meatus is estimated to be continuously low in the Kuruksay specimens. There is no trace indicating a lateral excavation of the inferior meatus into the maxillary body in both the Senèze and Kuruksay specimens.
The lacrimal fossa is located at the M3 level both in the Senèze specimen and in the adult female specimen from Kuruksay (PIN3120-523). Most of the regions of interest had been lost, and the route of the lacrimal canal is not examined in both specimens.
The maxillary body is thinner at the M1–2 level in the Senèze specimen compared with the Xin’an specimen. It has a slight maxillary fossa. Nevertheless, the body is thick at the premolar level, and at the distal M2–3 level, i.e. almost under the orbital cavity in the Senèze specimen. Such configuration is also seen in the Xin’an specimen. Whereas the muzzle is distorted slightly in the ventrodorsal direction in the adult female specimen (PIN3120-523) and is severely distorted in the juvenile male specimen (PIN3120-524) from Kuruksay, the maxillary body is also thin, as seen in the Senèze specimen. The specimens show a maxillary fossa at the cheek region, but this is emphasized partially by the ventrodorsal distortion in the Kuruksay specimens. Despite the limited materials for analysis, the distinct mandibular fossa strongly suggests a well-developed maxillary fossa in the Longdan specimen.
As seen in the Xin’an specimen, the zygomatic bone is sutured with the maxilla at the M3 level in all the adult specimens assigned to Paradolichopithecus. The recess that expands under the canine root is well developed at the P3 level in the Senèze specimen, as seen in the Xin’an specimen, and it is slight in the Kuruksay specimens.
Extant genera of PapioninaNo maxillary sinus is formed in any specimens of Mandrillus, Lophocebus, and Cercocebus, and in most of Papio and Theropithecus (Figure 4). Alternatively, the middle nasal meatus often expands laterally at the M3 level, which makes the medial wall of the maxillary body curve laterally (Figure 4A–D). In some of these specimens, the middle meatus adjoins the zygomaxillary suture.
Coronal CT images at the level of P3 to the tubercle (from left to right) in Papionina. (A) Mandrillus sphinx, adult female specimen, KUPRI-3091 (CT data id: PRICT-477). (B) Cercocebus galeritus, adult female specimen, KUPRI-4888 (PRICT-588); (C) Papio hamadryas, adult female specimen, KUPRI-5633 (PRICT-406); (D) Theropithecus gelada, adult female specimen, JMC-5910 (PRICT-1028); (E) Lophocebus albigena, adult female specimen, KUPRI-1141 (PRICT-494). The middle meatus expands slightly laterally at the M3 level (*). The maxillary body is thin in the cheek region because of the severe reduction in cancellous bone (white arrow). Abbreviations: tb, tubercle region of the maxilla; c, canine root; cb, cancellous bone region; ic, inferior concha; im, inferior meatus; ir, inferior recess under the canine; lf, lacrimal fossa; lc, lacrimal canal; me, eminence of the muzzle; sc, superior concha. The scale is in centimeters.
A distinct maxillary sinus is formed at the level of distal M3 to the tubercle under the orbital cavity in one female specimen each of Papio and Theropithecus (Figure 5C, D). In addition, whereas it should be called a maxillary sinus, a hollow space is found lateral to the middle meatus at the distal tubercle level in 13 (six female and seven male specimens) of 27 specimens of Papio (Figure 5E, F). The inferior concha bends sharply at the M3 level and the middle meatus extends laterally and inferiorly to make a hollow at the level of M3 to the tubercle. The bony wall separating the hollow from the nasal cavity was often removed at the tubercle level in the dry bony specimens, but the basal portion of the inferior concha is well preserved at the medial/inferior edge of the hollow in them (Figure 5E). Such a feature indicates that the hollow connects to the middle meatus. Whereas the hollow connects to the middle meatus, the inferior meatus sometimes expands distally into the space (Figure 5G, H). These maxillary pneumatizations occupy the restricted region under the orbital cavity, i.e. the region that cancellous bone should occupy in the other typical specimens (Figure 5).
Coronal and horizontal CT images of Papio hamadryas. (A, B) An adult female specimen, KUPRI-5633 (CT data id: PRICT-406) without maxillary pneumatization. (C, D) An adult female specimen, KUPRI-7868 (PRICT-1025) with a maxillary sinus. (E, F) An adult female specimen, KUPRI-2779 (PRICT-1027), with a hollow in the posterior maxilla; (G, H) an adult male specimen, KUPRI-Ph83 (PRICT-241), with a hollow invaded by the inferior meatus. Maxillary pneumatization is restricted to the region occupied by the cancellous bone in specimens having no pneumatization. The inferior concha bends sharply at M3 level and the middle meatus extends laterally and inferiorly to make a hollow at the M3 to tubercle level. The bony wall (dashed line) separating the hollow from the nasal cavity was often removed in the dry bone specimens. Abbreviations: M3m and M3d, mesial and distal region of the M3, respectively; tb, tubercle region of the maxilla; cb, cancellous bone; hs, hollow space pneumatizing the maxillary body; ic, inferior concha; im, inferior meatus; icb, basal portion of the inferior concha; ir, inferior recess under the canine; lc, lacrimal canal; lf, lacrimal fossa; mm, middle meatus; ms, maxillary sinus. The scale is in centimeters.
The inferior concha is suspended downward from the superomedial portion of the nasal cavity and the bilateral conchae are approximated to each other in large specimens of Mandrillus, Papio, and Theropithecus (Figure 4C, D), whereas the inferior concha usually extends medially and downward from the middle or slightly superior portion of the maxillary body into the nasal cavity in small specimens of these three genera and in Lophocebus and Cercocebus (Figure 4A, B, E). The inferior meatus varies and this variation corresponds to that of the inferior concha: from a high and wide meatus in the first three genera having the inferior conchae approximated and suspended, to a low and narrow one in the last two having the inferior concha extending medially (Figure 4). The lateral wall of the inferior meatus usually curves laterally and continuously from the nasal floor with no lateral excavation into the maxillary body in all five genera (Figure 4), while it is slightly angular against the nasal floor in Papio and Theropithecus (Figure 4C, D).
Although the distal part of the nasal conchae had been lost in many of the cranial specimens used here, the inferior and superior conchae usually expand in all five genera, but the superior concha is often slightly thin in large specimens of Papio, Theropithecus, and male Mandrillus (Figure 4).
The lacrimal canal travels rather horizontally to open into the tall inferior meatus at the M2–3 junction level from the lacrimal fossa on the anteromedial inferior portion of the orbital cavity at the tubercle level in Mandrillus, Papio, and Theropithecus (Figure 4A, C, D). This feature is associated with the suspended and approximated inferior conchae in these specimens (Figure 4A, C, D). It travels almost vertically to the inferior meatus from the lacrimal fossa at the level of M1 to mesial M2 in Lophocebus and Cercocebus (Figure 4B, E).
The maxillary body is thick at the premolar level, to anchor the canine root, and at the portion under the orbital cavity in Papionina (Figure 4). It is quite thin with little cancellous bone in the cheek region between the canine alveolus and the orbital cavity in Papio, Theropithecus, and Cercocebus (Figure 4C, D). In fully adult specimens of Papio and Theropithecus, where the orbital cavity is located distally to the tooth row, the maxillary body is thin even at the level of M3 to the tubercle. It is thin even under the thick anteroinferior rim of the orbital cavity in Cercocebus (Figure 4B). The maxillary body is slightly thick with a cancellous region along the bilateral eminence on the superior portion of the muzzle in Mandrillus (Figure 4A). In addition, it becomes slightly thick in the cheek region under the eminence in small female specimens of Mandrillus (Figure 4A), but it is thin under the eminence in the large male, as seen in Papio and Theropithecus. The maxillary body is slightly thick in the specimens of Lophocebus studied here, in which the orbital cavity is located at the M1 level distally to the canine alveolus (Figure 4E). The maxillary fossa is usually well developed in Papio, Theropithecus, the large male Mandrillus, and Cercocebus (Figure 4B, C, D). It is moderate in the female Mandrillus (Figure 4A), and slight in Lophocebus (Figure 4E).
The zygomatic bone is sutured to the maxilla at the M1–2 level in Lophocebus and Cercocebus (Figure 4). It is usually sutured at the level of distal M2 to M3 in female specimens and at the level of distal M3 to the tubercle in male specimens of Mandrillus, Papio, and Theropithecus specimens (Figure 4). In Lophocebus and Cercocebus, the zygomatic bone is thick and expands anteriorly at the inferior rim of the orbital cavity (Figure 4B, E).
The recess under the canine root varies from low and narrow to tall and wide in configuration, even within a single genus.
Extant genus of Macacina, MacacaThe CT scans used here confirm the formation of the maxillary sinus in all specimens of the genus Macaca (Figure 6). Wide variation in topology is also confirmed even in a single species, from the maxillary pneumatization restricted at the level M3 to the tubercle (Figure 6B) to the large pneumatization fully mesially to the canine alveolus (Figure 6A, C, D).
Coronal CT images at the level of P3 to the tubercle (from left to right) in Macaca. (A) Mac. sylvanus, adult male specimen, KUPRI-5644 (CT data id: PRICT-1020). (B) Mac. nemestrina, adult female specimen, KUPRI-79 (PRICT-719); (C) Mac. nigra, adult female specimen, KUPRI-1527 (PRICT-702); (D) Mac. thibetana, adult female specimen, JMC-5930 (PRICT-1019); (E) Mac. fascicularis, adult female specimen, KUPRI-4434 (PRICT-1026). The maxillary pneumatization is restricted to the region occupied by cancellous bone in the specimens without pneumatization. Abbreviations: tb, tubercle region of the maxilla; ic, inferior concha; im, inferior meatus; lc, lacrimal canal; lf, lacrimal fossa; ms, maxillary sinus.
The inferior concha usually extends medially and downward from the middle portion of the maxillary body into the nasal cavity, making the inferior meatus low and narrow in most species of Macaca (Figure 6A–C, E). This feature sometimes extends from the slightly superior portion of the maxillary body to make the inferior meatus slightly tall at the P4 to M1 level in some specimens of the species having a long face, e.g. Mac. nemestrina and Mac. thibetana (Figure 6D). The inferior meatus shows no excavation laterally into the maxillary body in this genus (Figure 6). The inferior and superior conchae usually expand at the distal part to comprise the inferior and superior surfaces of the middle meatus, respectively (Figure 6).
The lacrimal canal is always located at the anterior aspect of the ostium opening into the maxillary sinus in Macaca, i.e. at the anterior end of the middle meatus (Figure 6). It travels almost vertically to the inferior meatus from the lacrimal fossa of the anteromedial inferior portion of the orbital cavity in this genus (Figure 6).
The maxillary body is usually continuously thick even in the cheek region between the canine alveolus and orbital cavity in Macaca (Figure 6A, C, E). This thickened region is pneumatized by the maxillary sinus (Figure 6C) or is occupied by cancellous bone (Figure 6A, E). Regardless of the presence of a thick body, the maxillary fossa is developed slightly in some species of this genus such as Mac. nemestrina and Sulawesi macaques (Figure 6C). The maxillary body is rarely thin at the M1 level for some species such as Mac. nemestrina, Mac. thibetana, and Mac. fuscata (Figure 6B, D). This thin body contributes to the maxillary fossa in some specimens (Figure 6D), but not in others (Figure 6B).
The recess under the canine root varies from low and narrow to tall and wide in configuration in Macaca (Figure 6).
The morphological evaluations and comparisons for Procynocephalus were performed after confirmation of the morphological variations in nasal features in extant papionins. This survey was successful in providing information about some important variations and differences in the maxillary sinus, inferior concha, inferior meatus, and maxillary body thickness in the two subtribes Papionina and Macacina.
Here we confirmed that the maxillary sinus is formed homogeneously within the genus Macaca, as already known (Koppe and Ohkawa, 1999; Ito et al., 2009, 2014a, b). By contrast, heterogeneity in the presence and absence of maxillary pneumatization was found within two single genera in Papionina, namely Papio and Theropithecus. Although the pathological abnormality (loss or shrinkage) of this feature is known in nonhuman primates including Macaca (Koppe et al., 2006; Nishimura and Ito, 2014), there is no such pathological trait in the specimens used here. We note that maxillary pneumatization often occurs in Papio. These findings suggest that the genetic basis for the formation of the maxillary sinus has been secured since the ancestral loss of this feature in phenotype in cercopithecoids; it is usually silent in these two and other genera of Papionina. Although its biological advantages are still argued (Rae and Koppe, 2004; Rossie, 2008; Smith et al., 2010), the maxillary sinus is believed to develop opportunistically from the middle meatus if there are few structural disturbances as the pneumatization begins and progresses (Zollikofer and Weissmann, 2008; Smith et al., 2010). Whereas such disturbances occurring during the perinatal period are yet to be reported, the heterogeneity discovered here could reflect an anatomical variation that has been retained to release such disturbances in the two genera. The present findings also suggest that the maxillary sinus can also be found in extinct lineages of Papionina and not only in Macacina. Although no sinuses have been reported among extant colobines (Koppe and Ohkawa, 1999; Rae and Koppe, 2003), the maxillary sinus is found in some extinct forms including Cercopithecoides, Libypithecus, and Kanagawapithecus (Rae et al., 2007; Rae, 2008; Nishimura et al., 2012). Thus, whereas the absence of the maxillary sinus supports the view that a given fossil specimen should be excluded from the crown lineage of Macacina, the presence or absence of this feature does not necessarily reflect phyletic differences between given fossil specimens.
The suspended and approximated inferior conchae and the tall and wide inferior meatus are distinct in Papio/Theropithecus and Mandrillus, i.e. papionins having a long muzzle, as is already known in part (Koppe and Ohkawa, 1999). Their large inferior meatus differs in topology from the expanded inferior meatus seen in other primates, including Gorilla of hominoids (Koppe and Ohkawa, 1999) and Chiropotes of the New World monkeys (Nishimura et al., 2005). In those examples, the inferior meatus excavates laterally into the maxillary body. The large inferior meatus contributes to the vortex-like inhaled airflow in Papio (Patra et al., 1986), and this feature might offer evolutionary advantages by means of adjusting the temperature and humidity of inhaled air. The transitional conditions found here in the inferior concha and inferior meatus suggest that these features depend on elongation of the muzzle in Papionina. Such a view is also supported by the present finding that this transition accompanies a transition from a vertical to a horizontal lacrimal canal. Whereas there is usually a vertical canal in short-faced haplorrhines, a horizontal one is often found in long-faced strepsirrhines (Rossie et al., 2006). By contrast, such a transition and trend are not found in Macaca. The present findings thus suggest that the suspended and approximated inferior conchae could have arisen in any clades having an extremely long muzzle as seen in Papionina, such as Papio/Theropithecus and Mandrillus.
The maxillary fossa is one of the distinctive features in Papionina, whereas a slight maxillary fossa is also found in Macaca. The present findings strongly suggest that the maxillary bony materials are reduced and saved in Papionina if there is little functional need for them. The maxillary body remains thick around the canine alveolus to anchor the canine root, and under the orbital cavity to comprise part of the orbital floor. By contrast, cancellous bone is reduced in the long cheek region between the two portions, because of small functional need in Papionina, excluding Mandrillus in which a slightly thick maxillary body underlies the distinct facial architecture of this genus. Maxillary pneumatization, if present, is also restricted in the posterior thick region under the orbital cavity in Papio and Theropithecus. By contrast, the maxillary body is usually thick in the cheek region even in the long-faced Macaca, so the maxillary fossa is not well developed in this lineage. Thus, a distinct maxillary fossa can be regarded as a consequence of the major absorption of cancellous bone in Papionina.
Morphological variation in the nasal region, including the maxillary sinus, inferior conchae, and maxillary body thickness, contributes to our better understanding of the phyletic position and relationships of the Xin’an specimen of Pr. wimani. The lack of the maxillary sinus in the specimen supports the view that Procynocephalus must be excluded from the crown lineage of Macaca. Nevertheless, the thick maxillary body in the cheek region suggests that this genus should also be excluded from the crown lineage of Papionina. While Mandrillus has a bilateral eminence on the muzzle to form a thick maxillary body, the Xin’an specimen does not have such a facial feature. The maxillary body topology of the Xin’an specimen is rather similar to that of Macaca. Such a view does not contradict the topology of its inferior meatus: the inferior concha extends from the slightly superior portion of the nasal cavity as seen in some species of Macaca as well as in some genera of Papionina. Whereas the features preserved in the Xin’an specimen provide little conclusive evidence for its phyletic position, these interpretations do not contradict the view that Procynocephalus represents one of the extinct forms in Macacina.
The Senèze specimen of Para. arvernensis has a nasal architecture similar to that of the Xin’an specimen, except for the maxillary body thickness. The maxillary body is thin only at a restricted portion but is slightly thicker as a whole. Such a pattern differs from the distinct architecture in Papionina specimens having a comparable cranial size, and the topology is similar to that seen in Mac. thibetana (Figure 6D). These conditions do not contradict the view that the two specimens can be classified within the same subtribe. In contrast to these two specimens, the Kuruksay specimens of Para. sushkini possess a maxillary sinus, which might indicate that they are closer to the crown lineage of Macacina (Nishimura et al., 2007). However, the present findings do not strongly support such a phyletic view. Nevertheless, the Kuruksay specimens show a slightly thick maxillary body as seen in the Senèze specimen, and such a feature is consistent with the view that they should be excluded from the lineage of Papionina. Whereas the distorted specimens have a long and low muzzle and a moderate maxillary fossa, this species is regarded as being slightly higher and smoother in profile and having a relatively small maxillary fossa than they appear. The specimens of Paradolichopithecus from the two localities shows some morphological similarities and differences with each other and with the Xin’an specimen of Procynocephalus, but the nasal architectures are similar to that in the lineage of Macacina rather than of Papionina.
The present findings describe some morphological features in the nasal region aimed at a phyletic evaluation of the large Eurasian papionins from the Plio–Pleistocene. Many scholars have suggested that these primates arose in the Early Pliocene of western Eurasia and then dispersed eastward based on the geological age and distribution of the fossils (Necrasov et al., 1961; Jolly, 1967; Delson, 1974; Delson and Nicolaescu-Pliopsor, 1975; Aguirre and Soto, 1978; Ardito and Mottura, 1987; Eronen and Rook, 2004; Takai et al., 2008). The present evaluations of the three species do not contradict such a view, but they also support alternative biogeographic scenarios. For example, if the maxillary body topology is emphasized, Procynocephalus is regarded as having been distributed independently into southern and eastern Asia in the Early Pleistocene, and Paradolichopithecus achieved successful radiation in western Eurasia in the Late Plio–Early Pleistocene. The tribe Papionini was separated from the Cercopithecini, and then two subtribes, Papionina and Macanina, arose in the Late Miocene in Africa (Harris, 2000; Tosi et al., 2003; Raaum et al., 2005; Springer et al., 2012). While Macaca dispersed into Europe from Africa during the latest Miocene (Alba et al., 2014), papionins diversified abundantly and achieved successful radiation under diverse ecological environments in Africa in the Plio–Pleistocene, and some of them probably have descendants in Eurasia for each subtribe (Szalay and Delson, 1979; Gilbert, 2013). European Para. arvernensis has been reported to have some similarities to some of the African contemporaneous papionins: it shows low sexual dimorphism as in Gorgopithecus, and less developed maxillary and mandibular fossae as in Dinopithecus (Szalay and Delson, 1979). The paleobiogeographic arguments about the large Eurasian papionins will thus benefit from morphological examinations of the nasal architecture of papionins from the African Miocene to Pleistocene, such as Parapapio, Gorgopithecus, and Dinopithecus, as well as of Para. geticus from eastern Europe and Pro. subhimalayanus from southern Asia. Future evaluations of the phylogenetic relationships of the Eurasian and African forms should improve our understanding of the dispersion of the large papionins from Africa into western and eastern Eurasia in the Plio–Pleistocene, and of the phyletic relationships and adaptive radiation of the large Eurasian papionins.
We express our gratitude to the Museum of Evolution, Uppsala University, Sweden for providing the opportunity to examine the CT scans of the Xin’an specimen; and the Department of Anatomy (Macro) of the Dokkyo Medial University, Japan; Japan Monkey Centre; Primate Research Institute of the Kyoto University (KUPRI), Japan; Tochigi Prefectural Museum, Japan; Shikoku Institute of Natural History, Japan; Hakusan Nature Conservation Center, Japan; Kunming Institute of Zoology, Chinese Academy of Science; Kanagawa Prefectural Museum of Natural History, Japan; and Laboratory of Physical Anthropology, Graduate School of Sciences, Kyoto University, Japan for providing the opportunity to examine the CT scans of the extant species. We also thank V. Berg-Madsen, Y. Hamada, M. Nakatsuasa, and N. Ogihara for help with CT scanning of the Xin’an specimens; and N. Shigehara, H. Takahashi, D. Shimizu, T. Takano, Y. Shintaku, H. Taru, M. Kagaya, S. Kondo, Tet. Hayashi, Ter. Hayashi, S. Yachimori, E. Ashida, J. Xue-Long, S. Li, Thaung-Htike, Zin-Maung-Maung-Thein for help with CT scans of extant species. This research was financially supported in part by a grant by Ito Foundation, Tokyo (to T.D.N.), by JSPS Grants-in-Aid for Scientific Research (Grant 26650171 to T.D.N.; Grant 26304019 to M.T.), by JSPS Fellows (Grant 11J00120 to T.I.), by a JSPS Strategic Young Researcher Overseas Visits Program for Accelerating Brain Circulation (to KUPRI), and by JSPS ITP-HOPE Programs (to KUPRI).
