Broadly speaking, eastern Asia can be topographically divided into two regions by the Himalayan and Qinling mountain ranges, which effectively run through western and central China. The major uplift of the Himalayas occurred sometime during the Neogene (Fort, 1996; Dennell, 2009), which served as a formidable barrier throughout the Quaternary. The lower-lying Qinling mountain range, which is east of the Himalayas, would also have formed a barrier. To the east of the Qinling mountains, the region of eastern China is low-lying, with much of the region less than 1,000 m above sea level (Norton and Jin, 2009; Norton et al., 2010a, b). Because of this unique topography, eastern Asia can be divided into two primary regions: 1) Siberia, Mongolia, northern China, Korea, and Japan; and 2) southern China and mainland and insular Southeast (SE) Asia. In this article, I refer to the former area as Northeast Asia and the latter region as SE Asia. The primary distinction between the two areas is that the northern part falls within the Palearctic biogeographic region and the southern part represents the Oriental biogeographic region. Biogeographically, insular SE Asia is separated from Australasia by Wallace's Line. I have argued elsewhere (Norton et al., 2010a, b) that the low-lying region of eastern China would have served as a corridor where cold-adapted faunas would have migrated southward during stadials and warm-adapted taxa northward during interstadials. Indeed, presence of normally Oriental region-restricted nonhuman primates in Middle Pleistocene localities in northern China and Korea seem to support this argument (Jablonski et al., 2000; Norton, 2000; Norton et al., 2010b).
Because of the presence of fossil hominins in two distinct biogeographic zones (Palearctic and Oriental), the question that might arise is should we expect to find regional variation in hominin morphology as studies of eastern Asian human fossil materials have shown (e.g., Turner, 1990; Wu et al., 2007; Pietrusewsky, 2010)? Namely, might we distinguish the Middle Pleistocene hominin fossils from SE Asia (including southern China) and Northeast Asia, with morphologically transitional specimens present along the fluctuating Palearctic/Oriental boundary? Morphometric analysis of the Middle Pleistocene Tangshan, Huludong H. erectus crania (Liu et al., 2005), which is located in Nanjing, Jiangsu Province (central-east China), right along the Palearctic/Oriental boundary suggests this may be the case (see also Anton, 2002). The Liu et al. (2005) study found that the Tangshan H. erectus fossils share many features of the coeval Zhoukoudian (ZKD) Locality 1 hominins, but also had a few characters similar to the Indonesian H. erectus. A similar discussion has been presented for the nearby Hexian H. erectus cranium that seems to display character traits that could align it with the late H. erectus Ngandong hominins (Huang et al., 1982), but more likely ZKD H. erectus (Pope, 1992; Wu et al., 2006). Because of this biogeographic distinction, I divide this review of the hominin fossil record into Northeast and SE Asia (Fig. 1 and Table 1).
Important Northeast Asian hominin fossil localities
In addition to the well-known H. erectus fossils from ZKD Locality 1, a number of important Middle Pleistocene archaic H. sapiens sites exist in Northeast Asia, primarily from China (Pope, 1992; Etler and Li, 1994; Wu and Poirier, 1995; Etler, 1996). Middle and Late Pleistocene hominin fossils have also been reported from Mongolia and Korea (Norton, 2000; Coppens et al., 2008). The best-described NE Asian archaic H. sapiens are from Dali, Jinniushan, and Xujiayao, which are located in present-day China.
Dali is an open-air locality situated on the third terrace of the Lohe River located in Shaanxi Province, which is in the Chinese Loess Plateau (Wu and Poirier, 1995). Dali is represented by one almost complete skull that was discovered in 1978. The skull has been described as a robust male approximately 30 years of age. The cranium was found in association with a variety of typical Palearctic taxa and small stone flake tools. Interestingly, warm, humid-adapted faunal elements (e.g., Bubalus) were present, in association with taxa usually found in cooler, more forested environments (e.g., Palaeoloxodon). The only reported uranium-series date for Dali is 209 ± 23 ka (Chen et al., 1984) and a more recent paleomagnetic dating study brackets the hominin fossil between 300 and 260 ka (Yin et al., 2002). Pope (1992), noting the rolled and abraded condition of the paleontological and archaeological materials, suggests that the collection accumulated as the result of fluvial activity.
Wu (1981) and others (e.g., Pope, 1992) observed that the Dali skull has a number of traits that can be found in western archaic H. sapiens (e.g., Steinheim, Arago, and Jebel Irhoud), eastern H. erectus, and modern H. sapiens. For example, the thickness of the frontal and temporal squamae and parietal tuberosity fall within the range of ZKD Locality 1 H. erectus. In addition, the orbits are quadrangular, a characteristic common in H. erectus (Wu and Poirier, 1995). However, the degree of postorbital constriction and thickness and concavity of the tympanic plate lie between H. erectus and modern H. sapiens. Additionally, the cranial capacity of 1,120 cm3 falls between H. erectus and H. sapiens. A number of measurements and indices derived from the Dali cranium also fall within the range of archaic H. sapiens and are intermediate between H. erectus and modern H. sapiens. These include maximum length, maximum width, and transverse arc and curvature, which are more characteristic of archaic H. sapiens and outside therange of H. erectus (see Wu and Poirier, 1995: their Table 3.1).
In 1984, a partial hominin skeleton was discovered during excavations at Jinniushan, a karst fissure/collapsed cave site located in Liaoning Province, northeastern China (Lu, 1989, 2003; Wu and Poirier, 1995; Rosenberg et al., 2006). The Jinniushan skeletal collection comprises a more or less complete cranium (fragmentary, but undistorted) and ∼50 assorted partial and complete postcranial elements. Almost all of the fossils were excavated from a small 1.6 m2 area within the cave, suggesting that all of the bones derive from one individual (Lu, 2003). The hominin fossils were found in association with a mixture of Palearctic (e.g., Dicerorhinus) and Oriental (e.g., Macaca) taxa as well as about 200 lithic artifacts produced on locally available quartz and siliceous limestone. An array of U-series and electron spin resonance dates suggests the Jinniushan hominin may date to ∼280–260 ka (Lu, 2003), though some (e.g., Pope, 1992) have questioned the relationship between the hominin fossils and the samples used for dating.
The Jinniushan skeleton was originally interpreted to be an adult male (Wu, 1988a; Lu, 1989). However, Jinniushan has also been determined to be a female, based on the morphology of the pubis which displays a subpubic concavity and the “medial aspect of the ischiopubic ramus [which] is ridged rather than flat” (Rosenberg et al., 2006, p 3552). These traits are considered to be the two primary morphologies of the pubis that distinguish sex (Phenice, 1969). The Jinniushan cranium is considered to be larger, but more gracile than the Dali skull. For example, the cranial walls and brow ridges are less robust than H. erectus or Dali. The Jinniushan estimated cranial capacity of ∼1,300 cm3 is within the range of modern humans.
Rosenberg et al. (2006) conducted the most recent morphometric analysis of the Jinniushan materials and drew a number of interesting conclusions. In particular, they used a series of regression equations to estimate height, weight, and body proportions and determined that the Jinniushan female was ∼168 cm tall and ∼77 kg in weight. Given the fact that Jinniushan is located in a high-latitude region, and given what we know of ecogeographic clines in body size and structure (sensu Bergmann's and Allen's Rules), the results of the Rosenberg et al.'s (2006) study conform to what might be expected of a female hominin living in this type of environment. To date, the Jinniushan hominin is the largest female that predates Holocene modern humans and falls within range of body size reconstructions from other Middle Pleistocene hominins (e.g., Boxgrove, Atapuerca) (Rosenberg et al., 2006).
Xujiayao is an open-air site located in the western part of the Nihewan Basin in Shanxi Province, northern China (Jia et al., 1979; Wu and Poirier, 1995; Norton and Gao, 2008a). The site was discovered during field surveys conducted by the Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences in 1974 and subsequently excavated in 1976, 1977, and 1979 (Jia et al., 1979; Norton and Gao, 2008a). The hominin fossils were found in association with a diversity of Early Paleolithic artifacts (Gao and Norton, 2002) and a faunal collection dominated by Equus, with a smaller number of artiodactyla (e.g., Spiroceros, Gazella). Taphonomic analysis of the associated faunal assemblage suggested abundant evidence of efficient hunting and butchering of horse carcasses (Norton and Gao, 2008a). The chronometric age of the Xujiayao deposit is unclear, with ages ranging from ∼16 ka (14C) to ∼500 ka (magnetostratigraphy) (Norton and Gao, 2008a). The majority of the chronometric studies suggest an early Late Pleistocene age, though there is no clear consensus (Norton and Gao, 2008a; Nagatomo et al., 2009).
Xujiayao is best known for the presence of an array of hominin fossils, including twelve parietals, two occipitals, and one temporal fragment that have been assigned to the archaic H. sapiens group. The thickness of the Xujiayao parietal fragments is within the range of H. erectus (Jia et al., 1979; Wu, 1980; Pope, 1992; Wu and Poirier, 1995). However, the mastoid angle for at least two of the parietals (nos. 6 and 10) are thinner than H. erectus, but thicker than modern humans. The mandibular fossa is as deep as H. erectus, but the glenoid process is more similar to Dali. The occipitals display a reduced torus region. A partial maxilla of a juvenile was also excavated from Xujiayao. Presence of a deciduous second molar in the maxilla suggests the individual died between 7 and 9 years of age (Jia et al. (1979). Permanent maxillary teeth measurements all fall within the range of H. erectus. Pope (1992, p 269) notes that overall, the Xujiayao materials “present a combination of characters of less developed ectocranial superstructures (a reduced occipital torus, reduced occipital angulation, reduced parietal curvature, and expanded cerebellar fossae) in conjunction with greater general robusticity in some specimens (thick cranial bones and large dentition) and probably more marked sexual dimorphism than is generally seen in modern populations.” The importance of the Xujiayao material is that we are in a better position to understand the degree of morphological variation within a population from a single spatiotemporal point.
Homo cf. erectus and archaic H. sapiens fossils have been reported from at least three cave localities in North Korea: Yokpo Daehyundong, Dokchon Soongnisan, and Ryonggok (Park, 1992; Norton, 2000). Interestingly, North Korean paleoanthropologists have assigned these fossils to H. sapiens neandertalensis (e.g., Archaeology Research Laboratory, 1978; Kim et al., 1985; Jun et al., 1986). My own reading of the admittedly sparse North Korean literature indicates that North Korean paleoanthropologists refer all hominin fossils not clearly H. erectus or modern H. sapiens to H. sapiens neandertalensis. Two primary problems with accurately assessing the North Korean materials are the current access constraints and uncertainties about the chronometric ages (Norton, 2000). Currently, all non-North Korean scholars are restricted to the published literature and secondary or tertiary casts of the North Korean hominin fossils. The North Korean fossils date to the Middle to Late Pleistocene based usually on biostratigraphy, but occasionally a U-series or TL date has been reported (Norton, 2000). I present as much dental metric data as available from the Korean literature for all reported Pleistocene hominin fossils, including generally accepted modern H. sapiens (Table 2). Although other metric data is available for the North Korean hominin fossils, how the measurements match western terminology is still being ascertained. I describe the Yokpo Daehyundong, Dokchon Soongnisan, and Ryonggok fossils below (see also Norton, 2000).
Table 2. Published raw data for Korean hominin fossil teetha
|Middle/Late Pleistocene||Dokchon Soongnisan||Archaic H. sapiens||Lower right M1||11.6||10.5|
|Middle/Late Pleistocene||Dokchon Soongnisan||Archaic H. sapiens||Upper left M2||9.4||11.7|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 1 left M1||10.5||11.8|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 1 left M2||11.2||11.5|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 1 left M3||9.8||10.6|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 1 right M1||11.8||12.2|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 1 right M2||11.7||12.2|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 1 right M3||9.8||10.4|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 2 left P3||6.8||8.7|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 2 left P4||7.0||9.2|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 2 left M1||12.2||11.6|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 2 left M2||11.2||11.8|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 2 left M3||11.8||11.4|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 2 right M1||11.2||11.6|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 2 right M2||12.6||11.8|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 2 right M3||11.4||11.6|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 6 M1||11.2||12.2|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 6 M2||11.4||11.8|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Mandible no. 6 M3||12.1||12.2|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Skull no. 3 Upper left M1||10.4||10.5|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Skull no. 3 Upper left M2||9.2||11.2|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Skull no. 3 Upper left M3||8.2||10.2|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Skull no. 3 Upper right M1||10.2||11.0|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Skull no. 3 Upper right M2||9.8||11.2|
|Late Pleistocene||Ryonggok||Modern H. sapiens||Skull no. 3 Upper right M3||8.2||11.2|
|∼30 ka||Kumchon||Modern H. sapiens||Mandible canine||7.2||8.0|
|∼30 ka||Kumchon||Modern H. sapiens||Mandible P3||7.3||8.3|
|∼30 ka||Kumchon||Modern H. sapiens||Mandible P4||7.2||8.2|
|∼30 ka||Kumchon||Modern H. sapiens||Mandible M1||12.9||11.8|
|∼30 ka||Kumchon||Modern H. sapiens||Mandible M2||12.3||11.7|
|Late Pleistocene||Mandalli||Modern H. sapiens||Lower left M2||12.4||11.4|
|Late Pleistocene||Mandalli||Modern H. sapiens||Lower right M2||12.3||11.4|
The Yokpo Daehyundong hominin fossils consist of frontal, occipital, and parietal fragments of a juvenile estimated to be 7–8 years of age at time of death (Fig. 2; Kim et al., 1985). However, because of a well-developed occipital torus and a pronounced supraorbital torus, characteristics similar to ZKD Locality 1 H. erectus, it may be possible the Yokpo hominin might be better allocated to the H. erectus taxon (Norton, 2000). If this holds up to further scientific scrutiny, Yokpo Daehyundong would be the most eastern locality yielding H. erectus yet known. The Dokchon Soongnisan hominin fossils were excavated from two different stratigraphic levels. In the lower stratigraphic level, two teeth (lower M1 and upper M2) and a scapula fragment were recovered, whereas in the upper layer a partial mandible retaining a P4 and M1 were excavated (Fig. 3). Bronze Age artifacts were present 50 cm above the stratigraphic level containing the partial mandible, (Archaeology Research Laboratory, 1978; Norton, 2000). The partial mandible displays clear evidence of a mental eminence, indicating clear affiliation with modern H. sapiens. Based on biostratigraphy, Yokpo Daehyundong and Dokchon Soongnisan were both dated to the Middle-Late Pleistocene (Norton, 2000).
Figure 2. Yokpo Daehyundong hominin fossils. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Download figure to PowerPoint
Figure 3. Dokchon Soongnisan hominin fossils. Note the presence of a distinct mental eminence on the mandible. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Download figure to PowerPoint
Perhaps the most important Pleistocene hominin fossil site in North Korea is Ryonggok Cave, located near the present day capital city, Pyongyang (Jun et al., 1986; Norton, 2000). An assortment of hominin fossils were recovered (see Norton, 2000: his Table 3) representing at least five individuals from at least four separate stratigraphic levels. Jun et al. (1986) assigned Ryonggok to archaic H. sapiens, primarily based on the initial chronometric dating analysis. The chronometric dates are very disparate: initial TL = ∼500–400 ka; later U-series = ∼48–46 ka. Based on biostratigraphy, the associated archaeology, and the morphology of the hominin fossils, I suggest a Late Pleistocene age is a more reasonable one for the Ryonggok hominins (see also Chung, 1996; Norton, 2000). In particular, it should be noted that the Ryonggok hominin morphology includes “a rounded cranial vault, weak supraorbital tori, short face, steeply inclined forehead, absence of an occipital torus…, and presence of a chin” (Norton, 2000, p 814). Additionally, Skull no. 3 has an estimated cranial capacity of 1,650 cm3, and Skull no. 7 is estimated to be 1,450 cm3. Both are well within the range of modern humans and demonstrably outside the range of archaic H. sapiens (Fig. 4). Metric analysis of the mesiodistal and buccal-lingual measurements of the Ryonggok upper molars also generally fall within the range of modern humans (Fig. 5). Thus, it seems fairly clear that the Ryonggok hominins represent modern humans rather than archaic H. sapiens.
Figure 4. Ryonggok hominin fossils. Fairly intact crania of modern Homo sapiens. Note the rounded vault, reduced supraorbital tori, and flattened occipital tori.
Download figure to PowerPoint
Figure 5. Mesial-distal/buccal-lingual measurements for upper M1 (UM1) and upper M2 (UM2) Ryonggok human fossils (see Table 2 for raw data) compared with Holocene Chinese and different hominin taxa [MD along X axis; BL along Y axis; data for Holocene Chinese from Brace et al. (1984); data for various hominin taxa from Bailey and Liu (2010); all data are population averages]. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Download figure to PowerPoint
Table 3. Seventeen cranial morphological traits compared and contrasted from archaic Homo sapiens fossils from China and Africa and 10 characters distinguishing China and Europe (after Wu, 1988b; Wu and Brauer, 1993)
|1. Midsagittal elevation on the frontal squama||1. Midsagittal elevation|
|2. Bregmatic eminence||2. Flatness of the nasal saddle|
|3. Superior view of glabella region||3. Orientation of the anterolateral surface of the frontal process of the zygomatic|
|4. Midsagittal contour of the frontal squama||4. Contour of the lower border of the zygomatic process of the maxilla|
|5. Position of minimum distance between the superior temporal lines||5. Lower upper face|
|6. Interparietal sulcus||6. Shovel-shaped incisor|
|7. Angular torus||7. Position of the maximum breadth of the skull|
|8. Location of the maximum cranial breadth||8. Orbital margins|
|9. Lambdoidal ossicle||9. Sutures between the frontal and the nasal and maxillary bones|
|10. Protruding of the occipital region||10. Lambdoidal ossicle|
|11. Occipital torus|| |
|12. The posterior part of the nuchal plane in front of the occipital torus|| |
|13. The orientation of the anterolateral surface of the frontal process of the zygomatic bone|| |
|14. Bulging lateral to the upper part of pyriform aperture|| |
|15. The lower border of the zygomatic process of the maxillary bone|| |
|16. Alveolar prognathism|| |
|17. Upper facial height|| |
Interestingly, most of the Ryonggok hominin fossils appear to represent old individuals. For example, Mandible no. 1 is from a ∼50-year-old female, Mandible no. 2 is thought to be from a ∼45-year-old male, and Mandible no. 4 represents a ∼40-year-old female (Jun et al., 1986). Although determining age of individual clearly past the juvenile stage to exact year should be treated cautiously, the human mandibles clearly represent fully mature or aged individuals. As such, the possibility that Ryonggok Cave may have been a Late Pleistocene burial site should at least be considered. Intriguingly, a bone artifact with perforations reminiscent of a human face was found in stratigraphic level 11, which is also the most abundant human fossil layer (Jun et al., 1986; Norton, 2000). More detailed studies of the site and materials are clearly needed. In particular, comparative study ofthe fossils and archaeology from Zhoukoudian Upper Cave and Xiaogushan in China (Norton and Gao, 2008b; Norton and Jin, 2009) and Ryonggok Cave are warranted.
In 2006, a partial hominin calvarium was excavated from a pit 6-m deep during gold mining at the Salkhit site in northeastern Mongolia (Coppens et al., 2008). The hominin fossils are represented by a complete frontal bone and two partial parietal fragments. Although no chronometric dates are yet available, the associated fauna (e.g., Coelodonta antiquitatis) suggests a Late Pleistocene age (Coppens et al., 2008). Morphological analysis of the hominin cranial fragments revealed the presence of a mosaic of modern and archaic traits. Plesiomorphic characters include a frontal keel and relatively developed brow ridges, though clearly not as pronounced as H. erectus. Apomorphies include the absence of a sagittal keel on the parietals. According to Coppens et al. (2008), comparative metric analysis more closely aligns the Salkhit hominin with Neandertals, H. erectus sensu stricto, and archaic H. sapiens and away from modern humans of East Asian affinity. A number of Neandertal traits were identified, including the absence of the sphenoparietal sinus and superciliary arches that are thick at the frontal squama area and that taper off to the sides (Coppens et al., 2008). Given the recent argument that Neandertals have been identified as far east as Okladnikov Cave in southern Siberia (Krause etal., 2007), assigning the Salkhit hominin to Neandertals cannot be entirely discounted. This would be particularly interesting because recent morphometric studies of some modern human fossils from China (e.g., Tianyuandong and ZKD Upper Cave) suggest certain affinities with western Asian and European modern humans (Kamminga and Wright, 1988; Kamminga, 1992; Cunningham and Wescott, 2002; Cunningham and Jantz, 2003; Shang et al., 2007; Harvati, 2009). Although it has recently been suggested that evidence of modern human behavior moved west to east (Norton and Jin, 2009), it should be remembered that morphological and behavioral evolution do not always go hand in hand.
Important SE Asian hominin fossil localities
A number of important Middle Pleistocene H. erectus localities are present in China, including Tangshan Huludong, Hexian, Chenjiawo, and Yunxian (Etler and Li, 1994; Wu and Poirier, 1995). Other SE Asian H. erectus fossils include Tham Kuyen (Vietnam), Tam Hang (Laos), Had Pu Dai (Thailand), and Ngandong and Sambungmacan (Indonesia) (Olsen and Ciochon, 1990 Ciochon et al., 1996; Schepartz et al., 2000; Anton, 2003; Marwick, 2009), though the phylogeny of the hominin fossils from mainland SE Asia is still debated. The primary Middle Pleistocene archaic H. sapiens localities are Chaoxian (China), Maba (China), and Ma U'Oi (Vietnam). Besides Chaoxian and Maba, a number of archaic H. sapiens localities from southern China are described in Wu and Poirier (1995), including Changyang and Tongzi. Thum Wiman Nakin in Thailand is another potentially important locality that may eventually be considered an archaic H. sapiens site (Tougard et al., 1998). Reviews of the Paleolithic record of SE Asia published over the past two decades (e.g., Olsen and Ciochon, 1990; Pope and Keates, 1994; Schepartz et al., 2000; Marwick, 2009) indicate the great potential and need for more detailed multidisciplinary field and laboratory research programs in the region.
Chaoxian is an archaic H. sapiens locality situated in Anhui Province, eastern China and excavated in 1982 and 1983 (Wu and Poirier, 1995; Bailey and Liu, 2010). The Chaoxian hominin fossils are represented by a partial occipital, a partial maxilla, and three isolated maxillary teeth. Chaoxian is actually comprised of two distinct localities (A and B), separated by a 2–4 m limestone ridge. Locality A is represented by a typical Early/Middle Pleistocene fauna (e.g., Hyaena brevirostris licenti and Meganterion sp.), but no hominin fossils. Locality B has Middle Pleistocene fauna (e.g., Hyaena brevirostris sinensis, Stegodon, and Sus xiaozhu) and the archaic H. sapiens fossils (Shen et al., 2010). The initial U-series dates on associated animal teeth suggested an age range of 200–160 ka (Chen et al., 1987). However, a more recent study using TIMS U-series on associated speleothems indicated the hominin fossils could be bracketed between 360 and 310 ka or even older (Shen et al., 2010). Although the association between the dated speleothem section and the hominin fossils cannot be made with full degree of confidence, minimally, the Chaoxian hominin fossils should date to the Middle Pleistocene. No hominin trace fossils (manuports, lithics, or hominin-modified bones) were recovered from the site during either the initial surveys or the ensuing excavations (Wu and Poirier, 1995).
The Chaoxian occipital and maxilla are described in detail by Wu and Poirier (1995) and the teeth by Bailey and Liu (2010). The occipital fragment includes a large section of the occipital squama. Although an estimate, the occipital angle falls between ZKD H. erectus and modern humans. The thickness of the center of the occipital torus (7 mm) falls below the range of ZKD H. erectus, except ZKD no. VIII (Wu and Poirier, 1995). Based on tooth eruption and wear, the estimated age of the individual is about 30 years (Wu and Poirier, 1995) or more conservatively, a young- to middle-aged adult (Bailey and Liu, 2010). Morphometric analysis indicates that the Chaoxian teeth are “unremarkable” in that they fall within range of other Middle Pleistocene hominins (Bailey and Liu, 2010). In general, the teeth are considered to be large, with an occlusal morphology retaining primitive features (e.g., developed P4 accessory ridges, accessory fissures, and crests on the molars) of other Middle Pleistocene archaic hominins (Bailey and Liu, 2010). The anterior teeth are heavily worn and attributed to paramasticatory use, a trait found in many Neandertal fossils and other archaic H. sapiens (Brace, 1964; Brose and Wolpoff, 1971). The Chaoxian hominin teeth do not display any character traits that might align them with Neandertals (Bailey and Liu, 2010).
In 1958, local farmers excavated an archaic H. sapiens cranium from Shizishan near Maba village, Shaoguan Municipality, Guangdong Province, southeastern China (Wu and Poirier, 1995). Shizishan is represented by two low-lying limestone hills that has many cave entrances and a complicated network of naturally formed tunnels that stretch for several hundred meters at least. Separated by less than 100 m, it is thought that the two hills once formed a single hill. In the same cave that contained the hominin cranium, a partial hominin mandible and seven teeth were also discovered during fieldwork conducted in 1960 and 1984, respectively. The mandible and teeth are currently under study (Bae et al., n.d.). The age of the hominin fossils is unclear. The initial U-series analysis indicated an age between 135 and 129 ka (Yuan et al., 1986). However, a more recent TIMS U-series study of the same deposits suggested the capping flowstone should more readily date to 237 ka (Gao et al., 2007), thus pushing the minimum age of the Maba hominin back by at least 100,000 years. It should be noted, however, that the relationship between the capping flowstone and the hominin fossils is unclear; thus, caution is warranted when discussing the chronology of the Maba hominin (Bae et al., n.d.). Interestingly, Palaeoloxodon namadicus, a taxon usually found in higher latitudes was identified in the Maba faunal assemblage. No hominin trace fossils were found during subsequent survey or excavation (Wu and Poirier, 1995).
I restrict the following description to the Maba partial cranium, with more detailed description of the partial mandible and teeth presented later (Bae et al., n.d.). The Maba fossil is composed of a partial frontal, parietals, right orbit, and nasal region, thought to represent an adult male (Pope, 1992; Wu and Poirier, 1995). The Maba cranium is relatively flat, but higher than H. erectus (Wu and Wu, 1985; Wu and Poirier, 1995). The degree of postorbital constriction is pronounced. The arch-cord index of the parietal falls close to H. erectus and is much higher than in modern humans (Wu and Poirier, 1995). However, the Maba cranial walls are thinner than H. erectus. Perhaps the two most distinguishing characteristics of the cranium are the pronounced supraorbital tori, which are reminiscent of Neandertals and the rounded orbits, the latter feature which is not found in any other Early or Middle Pleistocene hominin from China (Wu and Peng, 1959; Wu and Wu, 1985). Although there have been suggestions that Maba could represent an eastern Asian Neandertal (e.g., Wu and Wu, 1985), there is general agreement that Maba is more similar to other archaic H. sapiens from China. Interestingly, the Maba partial cranium displays similarities with the Hathnora calotte from India, including rounded eye orbits, relatively robust supraorbital tori, and a flattened occipital region.
Ma U'Oi is a cave located in northern Vietnam just south of Hanoi, which was discovered during survey work at the nearby Chieng Xen cave site in 1999 (Demeter et al., 2004, 2005). Subsequent fieldwork at Ma U'Oi revealed the presence of two teeth (upper M2 and lower M1) and a partial occipital bone that were assigned to the archaic H. sapiens taxon. Ma U'Oi is situated in a tower karst region of the Annamitic Mountain Chain, with many additional caves located in the region (Demeter et al., 2005). The hominin fossils were found in the same stratigraphic level approximately 25–30 cm below the surface and in association with typical SE Asian Ailuropoda-Stegodon faunas (Bacon et al., 2004). Based on comparative analysis of the biostratigraphy from Thum Wiman Nakin (∼169 ka)4 and Lang Trang (∼80–60 ka), Ma U'Oi is thought to date to the late Middle Pleistocene to Late Pleistocene (Demeter et al., 2005). U-series dating of associated speleothems indicated an age range between 193 and 49 ka, which largely substantiates the earlier biostratigraphic study (Bacon et al., 2006). No hominin trace fossils have been reported from the cave.
The Ma U'Oi teeth were assigned to archaic H. sapiens for the following reasons (Demeter et al., 2004, 2005). In general, heavily worn molars of Pongo and Homo can be difficult to distinguish (Ciochon et al., 1996). Although heavily worn, the Ma U'Oi lower M1 displays “asymmetrical distribution of the enamel thickness on the occlusal surface” (Demeter et al., 2004, p 538), a characteristic more indicative of Homo rather than Pongo. The mesiodistal and buccal-lingual measurements more readily fall within the range of H. erectus rather than H. sapiens. However, the lower M1 lacks taurodontism, a trait considered more characteristic of Neandertals and H. erectus. Primarily because of this mosaic of morphological characteristics and partly because of its assumed geological age, the Ma U'Oi lower M1 is considered to be from an archaic H. sapiens (Demeter et al., 2004). The upper M2 also displays an “irregular distribution of enamel thickness and the absence of a doubled crista transversa” (Demeter et al., 2005, p 397), thus the Homo rather than Pongo assignment. The upper M2 lacks any evidence of taurodontism and occlusal wrinkles, suggestive of H. sapiens rather than H. erectus and Pongo. In addition, the mesiodistal and buccal-lingual measurements of the upper M2 fall within range of archaic H. sapiens (Demeter et al., 2005). The occipital fragment (upper left squamous region) is morphologically indistinguishable from archaic and modern H. sapiens (Demeter et al., 2005).