Journal of Evolution 106 (2017) 54e65

Contents lists available at ScienceDirect

Journal of Human Evolution

journal homepage: www.elsevier.com/locate/jhevol

An updated age for the Xujiayao hominin from the Nihewan Basin, North China: Implications for Middle Pleistocene human evolution in East Asia

* Hong Ao a, b, , Chun-Ru Liu c, Andrew P. Roberts d, Peng Zhang a, Xinwen Xu e a State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences, Xi'an 710061, China b Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY 10964, USA c State Key Laboratory of Earthquake Dynamics, Institute of Geology, China Earthquake Administration, Beijing 100029, China d Research School of Earth Sciences, The Australian National University, Canberra 2601, Australia e School of Urban and Environmental Science, Northwest University, Xi'an 710127, China article info abstract

Article history: The Xujiayao site in the Nihewan Basin (North China) is one of the most important Paleolithic sites in Received 19 June 2016 East Asia. Twenty Homo fossils, which were previously assigned to an archaic Homo sapiens group, have Accepted 26 January 2017 been excavated along with more than 30,000 lithic artifacts and ~5000 mammalian fossil specimens. Available online 17 March 2017 Dating of the Xujiayao hominin has been pursued since its excavation in the 1970s, but its age has remained controversial because of limitations of the dating techniques that have been applied to Keywords: available materials. Here, we report new ages for the Xujiayao hominin based on combined electron spin Magnestostratigraphy resonance (ESR) dating of quartz in the sediments and high-resolution magnetostratigraphy of the Electron spin resonance dating fl China uvio-lacustrine sequence. The magnetostratigraphy suggests that the upper Matuyama and Brunhes Archaic Homo sapiens polarity chrons are recorded at Xujiayao. The ESR dating results indicate a pooled average age of 260 Neanderthal e370 ka for the Homo-bearing layer, which is consistent with its position within the middle Brunhes normal polarity chron indicated by magnetostratigraphy. This age estimate makes the Xujiayao hominin among the oldest mid-Pleistocene hominins with derived Neanderthal traits in East Asia. This age is consistent with the time when early , a sister group of Neanderthals, appeared and colonized eastern Eurasia. Our updated age and the Neanderthal-like traits of the Xujiayao Homo fossils, particu- larly the Denisovan-like molar teeth, make it possible that the Xujiayao hominin could represent an early Denisovan. © 2017 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

1. Introduction 2014; Wu and Trinkaus, 2014). Multidisciplinary studies by several Chinese institutions using a variety of techniques have sought to With an inventory of >60 early Paleolithic sites, >100,000 in date the Xujiayao hominin since the 1970s. Based on initial ana- situ stone tools, and tons of mammal fossils, the Nihewan Basin in lyses of its faunal composition and stratigraphy, the site was sug- North China is one of the most important Paleolithic areas in East gested to date to at least >100 ka and possibly to the late Asia (Yuan et al., 2011). Xujiayao (also named Houjiayao, 40060 N, mid-Pleistocene (Jia et al., 1979). Uranium (U)-series dating of six 113 590 E, 980 m above sea level) is an open-air Paleolithic site that mammalian fossil teeth provided an estimated age of ~90e125 ka was discovered on the northwestern margin of the Nihewan Basin (Chen et al., 1982, 1984), which was possibly underestimated due in 1974 (Fig. 1). This site not only contains Oldowan-like stone to an open system in bones and teeth (Grün et al., 2014; Shen et al., tools (i.e., Mode 1 core and flake technologies) and mammalian 2014). Conventional 14C dating of organic material and mammalian fossils, but more importantly has yielded archaic Homo fossils (Jia bones later indicated an age older than 40 ka (Institute of and Wei, 1976; Jia et al., 1979; Wu, 1980; Bae, 2010; Wu et al., 2013, Archaeology of Chinese Academy of Social Sciences, 1991; Hayase, 2012). Based on magnetostratigraphy and correlation be- tween the Xujiayao environmental magnetic data and marine d18O

* Corresponding author. record, an age as old as ~500 ka has been suggested (Løvlie et al., 26 10 E-mail address: [email protected] (H. Ao). 2001; Wang et al., 2008). Recent Al/ Be burial dating of two http://dx.doi.org/10.1016/j.jhevol.2017.01.014 0047-2484/© 2017 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65 55

Figure 1. Location maps of the Xujiayao Paleolithic site. (a) Topographic map of the eastern Tibetan Plateau margin, the Nihewan Basin, and other hominin sites mentioned in the text. The Yellow and Yangtze rivers are the major river systems in north and south China, respectively. (b) Topographic map of the Nihewan Basin, the surrounding mountains, and two main rivers (Sanggan and Huliu rivers), with locations of the Xujiayao, Majuangou, and Shangshazui Paleolithic sites. The Nihewan Basin map is modified with permission from Tu et al. (2015). Triangles, solid circles, and five-pointed stars represent the sites occupied by modern H. sapiens, archaic H. sapiens,andH. erectus, respectively, with ages for Majuangou from Zhu et al. (2004), Shangshazui from Ao et al. (2013c), Yuanmou from Zhu et al. (2008), Gongwangling from Zhu et al. (2015), Yunxian from Dennell (2015), Zhoukoudian from Shen et al. (2001, 2004a, 2009), Chenjiawo from An and Ho (1989), Nanjing from Zhao et al. (2001), Hexian from Grün et al. (1998), Chaoxian from Shen et al. (2010), Jinniushan from Rosenberg et al. (2006), Dali from Xiao et al. (2002), Daoxian from Liu et al. (2015), Liujiang from Shen et al. (2002), Luna Cave from Bae et al. (2014),ZhirenCavefromLiu et al. (2010b), and Huanglong Cave from Shen et al. (2013).

quartz samples yielded a weighted mean age of 240 ± 50 ka (Tu The exposed horizontally stratified fluvio-lacustrine succession et al., 2015). The various age estimates have resulted in intense is ~17 m thick at Xujiayao (Fig. 2). To further extend the magneto- debate about the age of the Xujiayao hominin (Tu et al., 2015 and stratigraphic record, we dug a well down to a stratigraphic level of references therein), which complicates in-depth investigation of 24.2 m (the 0 m stratigraphic level indicates the top of the section). its evolutionary significance. The sequence is dominated by silt, clay, silty clay, and sand (Fig. 2). With recent improvement of electron spin resonance (ESR) About 200 m to the northwest of the studied section, the fluvio- technology, ESR dating of fluvio-lacustrine sediments provides an lacustrine sediments are patchily capped by last interglacial soil opportunity to date Pleistocene archeological sites (e.g., Voinchet (S1, not sampled; Wang et al., 2008). The artifact layer is located in et al., 2010). Electron spin resonance dating has been applied suc- the ~8e12 m interval that consists of brown clay, grayish-yellow cessfully in the Nihewan Basin (Liu et al., 2010a, 2013, 2014a). Here, silty clay, and grayish-dark silty clay (Fig. 2). we present new age estimates for the Xujiayao hominin using ESR Xujiayao is best known for the excavation of 20 archaic Homo dating of quartz-bearing sediment and high-resolution magneto- fossils (Jia and Wei, 1976; Jia et al., 1979; Wu, 1980; Wu et al., 2013, stratigraphy of the fluvio-lacustrine sequence and address the 2014; Wu and Trinkaus, 2014). The site was discovered during field paleoanthropological significance of the Xujiayao hominin in the surveys in 1974, which were conducted by the Institute of Vertebrate light of this updated dating evidence. Paleontology and , Chinese Academy of Sciences. Subsequent extensive excavation between 1976 and 1979 yielded 2. Geological and archeological settings two fairly complete parietals, a number of parietal fragments, a nearly complete temporal, two relatively complete occipitals, the The Nihewan Basin is an intermontane basin situated about ramus and posterior body of a lower jaw, a mandible, a juvenile 150 km northwest of the well-known Zhoukoudian maxilla, two upper molars, and a lower molar (Jia and Wei, 1976; Jia sites (Fig. 1) and has a relatively small area of roughly 150e200 km2 et al., 1979; Wu, 1980; Wu and Poirier, 1995). Some Homo fossils are (Yuan et al., 2011). Mid-Pliocene to Late Pleistocene fluvio- shown in Figure 3, including a maxilla, 3 molars, three parietal lacustrine sediments, which are known as the Nihewan Forma- fragments, an occipital, a mandible, and a temporal. Furthermore, tion (sensu lato; Barbour, 1924), were widely deposited in the Basin more than 30,000 pieces of stone artifacts (examples in Fig. 4), fl (Deng et al., 2008; Ao et al., 2013a). Stratigraphically continuous including cores, akes, scrapers, points, burins, borers, stone balls or and thick sequences of the Nihewan Formation are mainly exposed bolas, and chopper-chopping tools, have been reported from the site > along the southwest-northeast trending Sanggan River and along (Jia and Wei, 1976; Jia et al., 1979). Additionally, 10,000 stone tools the southeast-northwest trending Huliu River (Fig. 1). The eastern have been unearthed in recent years but have not yet been published margin of the Basin is a rich source of Pleistocene Paleolithic sites, formally. The artifacts (Fig. 5) from Xujiayao are typical Oldowan- with the presently oldest known Paleolithic evidence being docu- like technology and are made primarily of quartz (Ma, 2009). mented as early as ~1.7 Ma from the Majuangou and Shangshazui Direct hard-hammer percussion is the principal stone-knapping and sites (Zhu et al., 2004; Ao et al., 2013b). The Basin has a mean tool-touching technique (Ma, 2009). Also, ~5000 mammalian fossil annual temperature of 7.7C and precipitation of 365 mm, with specimens were unearthed from the Homo-bearing layer. At least 21 fi over 60% of the precipitation falling in the summer. taxa have been identi ed, including Struthio sp., Ochotona sp., 56 H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65

Figure 2. Lithostratigraphic column and field photographs of the exposed part of the studied Xujiayao section. The position of the well that was dug to sample the lower part of the sequence is evident in the lower photograph. Upper photo courtesy of Guanjun Shen.

Myospalax fontanieri, Microtus brandtioides, Canis lupus, Equus demagnetization experiments. We also took six sediment samples przewalskii, Coelodonta antiquitatis, Cervus elaphus, Megaloceros from the artifact layer for detailed ESR dating. ordosianus, Bos primigenius, Panthera cf. tigris, Palaeoloxodon cf. Progressive thermal demagnetization of the natural remanent naumanni, Equus hemionus, Cervus nippon grayi, Spirocerus hsu- magnetization (NRM) was carried out using a TD-48 thermal chiayaocus, Spirocerus peii, Procapra picticaudata przewalskii, Gazella demagnetizer. Oriented samples were stepwise heated at 15e50C subgutturosa, Gazella sp., and Sus sp. (Jia and Wei, 1976; Jia et al., temperature increments to a maximum temperature of 585Cor 1979). The taxonomic composition of the fossil mammals in the 640C, at which point >90% of the initial NRM was demagnetized. Nihewan Basin implies a cool and open savannah environment with After each demagnetization step, the remaining NRM was patches of forest (Norton and Gao, 2008; Ao et al., 2013c). measured using a 2-G Enterprises Model 755-R cryogenic magne- tometer housed in a magnetically shielded room (Institute of Earth Environment, Chinese Academy of Sciences, Xi'an). The NRM in- 3. Methods tensity of the samples was usually in the 10 4e10 2 A/m range, In order to obtain samples that were as fresh as possible, while the magnetometer background (or noise) magnetization 7e 6 weathered outcrop surfaces were removed. Compass-oriented level is 10 10 A/m. Demagnetization results were evaluated block samples were collected from 336 layers in the 24.2 m thick using orthogonal vector component diagrams (Zijderveld, 1967); fluvio-lacustrine sequence at 5e10 cm stratigraphic intervals. From the principal component direction for each sample was computed fi each layer, one large block sample was collected in the field, which using a least-squares tting technique (Kirschvink, 1980). The was later cut in the laboratory into two cubic (2 cm 2cm 2 cm) principal component analysis was carried out using the PaleoMag specimens (with identical stratigraphic position) for thermal software (Jones, 2002). H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65 57

Figure 3. Photographs of typical Homo remains from Xujiayao (Liu et al., 2014b; Wu et al., 2014; Xing et al., 2015). (a) Maxilla, (b) 3 molars, (c) three parietal fragments, (d) an occipital, (e) a mandible, and (f) a temporal.

Figure 4. Typical stone tools from Xujiayao (Ma, 2009). (a) Burin, (b) denticulate, (c) scraper, (d) notch, (e) core, (f) flake, (g) borer, (h) pointed, (i) stone ball, and (j) chopper. 58 H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65 3

2

1 g=1.979

0 Ti-Li g=1.913 -1

-2 Depth: 9 m ESR intensity (arbitrary unit) Nature (without additional irradiation dose) -3 2.0795 2.0480 2.0174 1.9878 1.9590 1.9310 1.9038 G-Factor

Figure 5. ESR spectrum and illustration of the method used to determine the intensity of the quartz TieLi center.

The 105e200 mm size fraction for each ESR sample was extracted (Fig. 6). After removal of the secondary overprint, the characteristic by wet sieving. The extracted samples were treated with 30% H2O2 remanent magnetization (ChRM) component is isolated at higher for a day to remove organic material and with 40% HCl for another temperatures (Fig. 6). For samples from the 336 demagnetized day to remove carbonates, followed by washing and cleaning with levels, 205 yielded stable ChRM components based on strict se- distilled water. The samples were then etched with 40% HF for lection criteria: (1) data from at least four (but typically 6e11) 40 min to remove feldspars followed by rinsing to neutral pH with consecutive demagnetization steps were used for linear fitting, distilled water, drying at low temperature, removing magnetic starting at least at 250C (with upper temperatures of 450C), and minerals with a rare earth magnet, and a heavy liquid separation (2) a maximum angular deviation (MAD) < 15 was required for the using sodium polytungstate. After rinsing with distilled water, line fit. The virtual geomagnetic pole (VGP) latitudes calculated drying, and checking the mineral separate under a microscope, pure from all 205 ChRM directions are used to establish the Xujiayao quartz grains were obtained. Each sample was divided into 200 mg magnetostratigraphy, which allows us to recognize two main po- subsamples. These aliquots received additional gamma doses larity intervals (Fig. 7). The upper 18.3 m thick normal polarity increasing from 0 to 10 kGy using a panoramic 60Co source (Peking interval correlates to the Brunhes chron. The underlying 5.9 m thick University, China), with a dose rate of 20e40 Gy/min. reversed polarity interval (only one normal polarity sample) is thick The sampled site is characterized by typical fluvio-lacustrine enough to represent a sufficiently long period of time, which makes sediments, thus, we did not perform in situ measurements of the it highly unlikely to represent a short-duration geomagnetic dose rate (D; Guerin and Mercier, 2011), which was calculated from excursion within the Brunhes chron (Roberts, 2008). Thus, we can the concentrations of uranium, thorium, and potassium of each readily correlate it to the upper Matuyama reversed polarity chron. sample (Aitken, 1998). 238U and 232Th contents were obtained using Our magnetostratigraphic assignment is consistent with the pre- a thick source Daybreak 530 Model alpha counter. The potassium viously reported Xujiayao magnetostratigraphy (Løvlie et al., 2001). oxide content was determined by atomic absorption. Cosmic dose rates for deep (>2 m) burial were ignored (Prescott and Hutton, 4.2. ESR dating 1988). The water content was assumed to be 10% with an error of 5% based on previous studies of Nihewan fluvio-lacustrine sedi- The resetting rate of ESR centers in quartz extracted from fluvial ments (Liu et al., 2010a, 2013, 2014a). sediments plays a key role in ESR dating. Recent improvements in Electron spin resonance measurements were carried out with a understanding the bleaching mechanism on quartz ESR signal BRUKER ER041XG X-band spectrometer in a finger dewar cooled to resetting confirms the potential of dating bleached quartz by ESR. 77 K with liquid nitrogen in the ESR laboratory of the Institute of Sunlight exposure was considered the first driver for quartz Geology, China Earthquake Administration, Beijing. The TieLi cen- bleaching in fluvial sediments. In contrast to the Al center ESR ter was used because it is more suitable for dating mid-Pleistocene signals that are known to have somewhat slow bleaching kinetics fluvial-lacustrine sediments than the classic Al center (Beerten and an unbleachable (residual) component, TieLi center signals et al., 2006; Tissoux et al., 2007). The experimental parameters have been shown to be completely bleached after the equivalent of were: microwave power 5 mW, modulation amplitude 0.16 mT, a few days of sunlight exposure (e.g., Toyoda et al., 2000; Tissoux modulation frequency 100 kHz, conversion time 40.96 ms, center et al., 2007). Furthermore, ESR dating of the Dongpo Paleolithic field 3396 G, and sweep width 350 G. The TieLi center intensity was site (~60 km east of Xujiayao) in the northeastern Nihewan Basin measured from the top of the peak at g ¼ 1.979 to the bottom at margin (Liu et al., 2010a) demonstrated that Pleistocene Nihewan g ¼ 1.913 as shown in Figure 5 (Rink et al., 2007). Each sample was fluvio-lacustrine sediments were bleached almost to zero before measured six times to obtain an average intensity, while rotating their last deposition. Similarly, fluvial samples collected along the the sample tube within the cavity. Creuse River, France, from its spring to 170 km downstream indi- cated that all modern sediments had completely bleached Ti cen- 4. Results ters (Voinchet et al., 2007). In addition, recent experimental results suggest that fluvio-mechanical action could be another non- 4.1. Magnetostratigraphy negligible bleaching driver, leading to obvious partial resetting with short tumbling times (Liu and Grün, 2011). Therefore, equiv- Generally, thermal demagnetization to 250 C removes second- alent dose values (DE) from the quartz TieLi center in our six ary NRM components in fluvio-lacustrine sediments from Xujiayao studied samples from Xujiayao were obtained using a complete H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65 59

(a) 7 m: yellow silt (b) 16.3 m: grayish-green clay (c) 19.35 m: grayish-yellow silty clay (d) 23.36 m: grayish-dark silty clay W / Up W / Up Scale: 4 x 10 A/m Scale: 1 x 10 A/m 200 Scale: 4 x 10 A/m Scale: 4 x 10 A/m W / Up W / Up 150 300 400 450 640 N N 100 NRM 200 250 640 500 100 350 NRM 50 300 400 585 300 300 640 350 250 N 250 400 200 200 585 150 150 N 100 100 NRM NRM Horizontal plane Vertical plane 160 100 160 200

) 80 160 120 120 A/m 60 120 10 80 80 40 80 40 40

Intensity 20 40

0 0 0 0 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 Temperature ( C) Temperature ( C) Temperature ( C) Temperature ( C)

0 0 0 0

640 640 NRM NRM 270 90 270 90 270 90 270 90

NRM NRM

Lower hemisphere 640 585 180 Upper hemisphere 180 180 180

Figure 6. Results from thermal demagnetization of the natural remanent magnetization (NRM), with vector component demagnetization diagrams, thermal decay curves, and corresponding equal-area stereographic projections for demagnetization data from representative samples from Xujiayao. W/Up ¼ west and upward directions; N ¼ north di- rection; A/m ¼ ampere per meter. bleaching assumption. Even if there was a small residual TieLi 5. Discussion center signal, it could only result in slightly younger ESR ages than our calculated ages based on complete bleaching (Liu and Grün, 5.1. Age estimation of the Xujiayao Paleolithic site 2011). For all samples, DE values and their individual errors were determined from dose response data fitted with a single saturating Like previous magnetostratigraphic results from Xujiayao exponential function using the protocol (software) of Yokoyama (Løvlie et al., 2001), our results indicate that the artifact layer is et al. (1985). Thus, errors in established ESR ages result mainly located in the middle Brunhes normal polarity chron (Fig. 7), which from fitting errors of DE, while errors for conversion of radioactive corresponds to a mid-Pleistocene age. The ESR dating provided isotope concentrations to DE values (Guerin et al., 2011) are not consistent error-weighted mean ages of 374 ± 27 ka and 260 ± 30 significant. Dose responses of quartz ESR signals for the TieLi ka for the lower (12e9.6 m) and upper (9.6e8 m) parts of the center for two samples from Xujiayao are shown in Figure 8. artifact layer, respectively. This indicates rapid deposition of the Quartz TieLi center ESR results and associated data for six lower and upper parts of the Xujiayao artifact layer, which is samples from the Xujiayao artifact layer are shown in Table 1. Two consistent with rapid burial of the abundant Homo and mammalian brown clay samples (at 8 and 9 m) and a grayish-yellow silty clay fossils and stone tools. These dates also support the possible sample (9.4 m) from the upper artifact layer yielded ages of presence of a sedimentary hiatus between the two parts of the 267 ± 29 ka, 232 ± 34 ka, and 282 ± 27 ka, respectively (Fig. 7). artifact layer, as suggested by Tu et al. (2015) based on sharp lith- Although the middle sample has a slightly younger age than the ological change (Fig. 2). The whole artifact layer spanning from 12 upper sample, this minor difference (~30 ka) is within the error to 8 m thus ranges in age from ~370 to 260 ka. This mid-Pleistocene range of ESR dating (Liu et al., 2014a). Three grayish-dark silty clay age assignment is supported by the Xujiayao hominin teeth (ex- samples from the lower artifact layer (10e12 m) yielded consistent amples in Fig. 3b), which are consistently larger and have ages of 374 ± 32 ka, 373 ± 22 ka, and 376 ± 25 ka, respectively morphological features (e.g., metric dimensions, trapezoidal shape (Fig. 7). These consistent ESR ages support a complete, or nearly of the M1 crown outline, and shape, robusticity, and divergence of complete, bleaching of the ESR signal prior to deposition and, thus, molar roots) that are more typical of Early and Middle Pleistocene are expected to yield a well-established ESR chronology. Asian populations (Xing et al., 2015). This age estimate also agrees 60 H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65

Figure 7. Age constraints for the Xujiayao hominin with magnetostratigraphy and ESR dating. (a) Lithology, (b) paleomagnetic declination (Dec.), (c) inclination (Inc.), (d) maximum angular deviation (MAD), (e) virtual geomagnetic pole (VGP) latitude, (f) paleomagnetic polarity sequence for the Xujiayao section, (g) ESR dates for the artifact layer, and (h) geomagnetic polarity timescale (GPTS; Hilgen et al., 2012).

90 180 80 150 70 60 120 Depth: 9.4 m 50 Depth: 9 m 90 40

30 60 20 ESR intensity (arbitrary unit) ESR intensity (arbitrary unit) DE = 1131 ± 166 30 DE = 1282 ± 121 10

-2500 02500 5000 7500 10000 12500 -2500 0 2500 5000 7500 10000 12500 Dose (Gy) Dose (Gy)

Figure 8. Dose response of the quartz ESR signal for the TieLi center for typical samples from Xujiayao. DE ¼ equivalent dose. with those for associated mammalian fossils from Xujiayao, which depositional system and the presence of erosional unconformities contain many of the same species as the mid-Pleistocene Zhou- at Xujiayao (Jia and Wei, 1976; Jia et al., 1979; Ma, 2009; Li et al., koudian H. erectus sites (Wu et al., 1985). 2014; Tu et al., 2015) may have resulted in an older age from Our estimated age for the Xujiayao hominin is much younger magneto-cyclochronology, which assumed no stratigraphic dis- than the ~500 ka age suggested by magneto-cyclochronology continuities (Løvlie et al., 2001; Wang et al., 2008). Uranium-series (Løvlie et al., 2001; Wang et al., 2008) and is considerably older dates from pure calcites are generally considered to be accurate and than the ~90e125 ka age obtained from conventional U-series precise, but datable calcite is not present at Xujiayao. Unlike calcite, dating of mammalian teeth (Chen et al., 1982, 1984). The complex bones are susceptible to post-burial U migration, thus, U-series H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65 61

Table 1 ESR ages and associated data (mean ± standard deviation) from the Xujiayao artifact layers.a

Sample number Burial Water U Th K2O % D (Gy/ka) DE ESR age depth % (ppm) (ppm) (Gy) (ka) (m)

11385 8.0 10 ± 5 4.11 ± 0.21 12.39 ± 0.62 2.65 ± 0.13 4.21 ± 0.23 1124 ± 122 267 ± 29 12130 9.0 10 ± 5 5.36 ± 0.27 13.38 ± 0.67 2.66 ± 0.14 4.88 ± 0.28 1131 ± 166 232 ± 34 12131 9.4 10 ± 5 5.10 ± 0.26 12.05 ± 0.61 2.51 ± 0.13 4.54 ± 0.25 1282 ± 121 282 ± 27 12133 10.0 10 ± 5 5.04 ± 0.25 12.93 ± 0.65 2.62 ± 0.13 4.74 ± 0.26 1772 ± 152 374 ± 32 11382 11.0 10 ± 5 3.10 ± 0.16 12.11 ± 0.61 2.45 ± 0.13 3.34 ± 0.19 1245 ± 72 373 ± 22 11381 12.0 10 ± 5 4.50 ± 0.23 13.10 ± 0.66 2.35 ± 0.12 3.73 ± 0.21 1403 ± 94 376 ± 25

a D ¼ dose rate, DE ¼ equivalent dose; ppm ¼ parts per million. dating of bones and teeth is prone to producing substantially young Chaoxian (310e360 ka) in eastern China (Shen et al., 2010), plus H. ages as demonstrated by U-series dating of intercalated calcite erectus sites at Zhoukoudian Locality 1 (400e500 ka; Shen et al., formations and excavated fossil bones in archeological sites 2001) and Nanjing (580e620 ka) in eastern China (Zhao et al., (Bischoff et al., 2003; Shen et al., 2004a, b, 2010; 2014; Grün et al., 2001). These new or updated ages have contributed to funda- 2014). Technical limitations and the low resolution of measure- mental revision of our understanding of human evolution in China. ments made more than 30 years ago therefore mean that initial U- For example, modern H. sapiens remains dated to >100 ka from series dating of teeth from Xujiayao may have underestimated their China, although debated (Dennell, 2010; Michel et al., 2016), true ages. Also, about 200 m to the northwest of the studied section, significantly challenge the traditional view that modern the fluvio-lacustrine sediments are patchily capped by a ~0.5 m were restricted to portions of Africa before 100 ka and that they thick paleosol from the last interglacial (S1) of the Chinese loess dispersed rapidly throughout Eurasia after 60 ka (Qiu, 2016). sequence (Wang et al., 2008) that ranges in age from ~80 to 130 ka A complete and updated chronological framework for the (Sun et al., 2006). Accordingly, the top of the Xujiayao section must abundant Pleistocene sites in China remains far from established be older than 80 ka. Consistently, optically stimulated lumines- because many sites have poorly constrained ages due to the lack of cence (OSL) dating of quartz grains suggests that two layers (at volcanic products and pure calcite for 40Ar/39Ar and U-series dating, depths of 1 and 3 m) from the uppermost Xujiayao section are older respectively. Along with improved ESR technology over the past than the conventional quartz OSL dating range (100e150 ka) for decade, ESR dating of bleached quartz extracted from sediments fluvial-lacustrine sediments (Zander and Hilgers, 2013) and have has become a useful dating tool for many Pleistocene sites, which identical OSL ages of ~100 ka (Li et al., 2014). are otherwise often devoid of suitable material for radiometric In agreement with recent 26Al/10Be burial dating of the lower dating. In particular, ESR dating of quartz sediments combined with artifact layer at ~12 m (Tu et al., 2015), our ESR dating of quartz high-resolution magnetostratigraphy is useful for dating Pleisto- sediments, together with high-resolution magnetostratigraphy, cene sites and has been applied successfully to the mid-Pleistocene suggests a mid-Pleistocene age for the Xujiayao hominin, in Paleolithic site of Dongpo (304e333 ka; Liu et al., 2010a) and to the contrast to the previous Late Pleistocene age based on U-series Early Pleistocene sites of Dongguotuo (~1.1 Ma; Liu et al., 2013), dating of fossil remains. Given an age of 100 ka at a depth of 1 m at Banshan (~1.35 Ma), and Majuangou (1.4e1.7 Ma; Liu et al., 2014a) the Xujiayao section according to the OSL dating (Li et al., 2014), in the Nihewan Basin. Our combined ESR and magnetostratigraphic dates of 260 ka at 8.7 m and 370 ka at 11 m according to our ESR dating of Xujiayao represents a further step in establishing a dating, and an age of 780 ka at 18.3 m according to our magneto- complete chronological framework for Pleistocene sites in China stratigraphy, we establish a relationship between stratigraphic and makes possible the comparative study of contemporaneous depth and age (Fig. 9). Consistent with lithological variations, the Asian and non-Asian hominins. yellow fluvial silts in the upper section had higher sedimentation rates than the underlying lacustrine grayish-green clays and 5.2. Implications for hominin evolution grayish-dark silty clays. Until now, ages for Middle and Late Pleistocene hominin or The Xujiayao Homo fossils have mixed characteristics associated Paleolithic sites in China were mainly determined through U-series with European Neanderthals, Asian H. erectus, and modern H. sa- dating of fossil bones one or two decades ago, which contributed piens (Jia and Wei, 1976; Jia et al., 1979; Wu, 1980; Bae, 2010; Wu significantly to our understanding of human evolution. However, et al., 2013, 2014; Wu and Trinkaus, 2014; Xing et al., 2015), the considerably underestimated dates due to open systems in which makes it difficult to affiliate the Xujiayao hominins to bones and teeth (Bischoff et al., 2003; Shen et al., 2004a, b, 2010, “classic” H. erectus, modern humans, or Neanderthals. Thus, the 2014; Grün et al., 2014) has complicated in-depth investigation of Xujiayao hominins were assigned to archaic H. sapiens (Jia and Wei, the evolution of Middle and Late Pleistocene hominins in East Asia 1976; Jia et al., 1979; Wu, 1980; Wu and Poirier, 1995), although this and comprehensive understanding of their links to well-dated unique term in China is controversial as noted by later studies (e.g., contemporaries in Africa and Europe (Dennell and Petraglia, Rightmire, 1998; Dennell and Petraglia, 2012). Our updated chro- 2012). With improved technology in the past 10 years, U-series nology makes the Xujiayao Homo fossils among the oldest archaic dating of intercalated calcite formations rather than bones, which H. sapiens in China. They are contemporaneous with the earliest were used previously, has established relatively precise ages for archaic H. sapiens remains in eastern China from Chaoxian several key Late and Middle Pleistocene sites, such as the modern (310e360 ka; Shen et al., 2010). Combining the archaic H. sapiens Homo sapiens sites of Huanglong Cave (81e101 ka) in central China remains from New Cave (248e269 ka; Shen et al., 2004a)at (Shen et al., 2013), Luna Cave (70e127 ka; Bae et al., 2014), Daoxian Zhoukoudian, Dali (~270 ka; Xiao et al., 2002) and Jinniushan (~260 (80e120 ka; Liu et al., 2015; but see Michel et al., 2016), Liujiang ka; Rosenberg et al., 2006), it appears that archaic H. sapiens (111e139 ka; Shen et al., 2002), and Zhiren Cave (>100 ka; Liu et al., occupied a vast area across China during the mid-Pleistocene. Un- 2010b) in South China, as well as the archaic H. sapiens sites of New like some African mid-Pleistocene Homo individuals that were Cave (248e269 ka) at Zhoukoudian (Shen et al., 2004a) and associated with Acheulian stone tools (Rightmire, 2008), the 62 H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65

0

2 artifact layer 4

6 yellow silt 8

10 Artifact layer bronw clay 12

Depth (m) 14 fine-grained sand 16

18 grayish-dark silty clay 20

grayish-green clay 22

24 Brunhes Matuyama grayish-yellow silty clay 0 100 200 300 400 500 600 700 800 900 1000 Age (ka)

Figure 9. Stratigraphic depth versus age relationship for the Xujiayao section. archaic H. sapiens from Xujiayao and other East Asian sites (e.g., heidelbergensis is documented in East Asia. Based on more precise Jinjiushan and New Cave) were associated with a relatively simple recent ages for various Homo fossils established in recent years and Oldowan-like technology (Bae, 2010; cf. Fig. 4). Despite having a without regard to previous imprecise ages that were under- simple technology, the Xujiayao hominins were able to successfully estimated by U-series dating of bones, it is possible that archaic H. obtain regular sources of animal fat and protein that probably sapiens (370e250 ka) may have not interacted with the older H. helped them to survive harsh mid-latitude northeast Asian winters. erectus (1700e400 ka) or younger modern H. sapiens (<150 ka) in Surface modifications on long bone midshafts indicate that the East Asia, as indicated by our updated Chinese Homo chro- Xujiayao hominins were skilled large mammal (e.g., horse) hunters nostratigraphy in Figure 10. and had access to high utility (meat-bearing, marrow-rich) long Recent studies of the Xujiayao Homo fossils (Wu et al., 2012, bones (Norton and Gao, 2008), which was important for over- 2014; Wu and Trinkaus, 2014; Xing et al., 2015) indicate that they wintering in the >40N temperate zone. are more similar to Neanderthals than to H. heidelbergensis or H. Homo erectus occupation of East Asia started at 1.7e1.6 Ma and erectus. For example, the bi-level nasal floor in the maxilla of persisted to ~400 ka as suggested by fossils from Yuanmou Basin Xujiayao individual I (PA1480) and the presence of a well- (~1.7 Ma; Zhu et al., 2008), Nanjing (580e620 ka; Zhao et al., 2001), developed medial pterygoid tubercle, a retromolar space, and an Hexian (400e420 ka; Grün et al., 1998), and Yunxian (0.936 Ma; asymmetrical mandibular notch in the Xujiayao 14 mandibular Dennell, 2015) in South China and Gongwangling (1.62e1.63 Ma; ramus are typical characteristics of Neanderthals (Wu et al., 2012; Zhu et al., 2015), Chenjiawo (0.65 Ma; An and Ho, 1989), and Wu and Trinkaus, 2014). The morphology of the temporal laby- Zhoukoudian (0.4e0.77 Ma; Shen et al., 2001, 2009) in North China rinth of Xujiayao 15 and the relative proportions of their anterior, (Fig. 10). In Africa, H. erectus was giving way to Homo heidelbergensis posterior, and lateral canals are consistent with Neanderthal traits during the terminal Early Pleistocene to the earliest mid- (Wu et al., 2014). Likewise, the Xujiayao teeth have some Nean- Pleistocene (ca 600e800 ka; Rightmire, 1998, 2008, 2009, 2013). derthal features (e.g., high degrees of shovel shape and labial Until now, Homo fossils with unambiguous affinities to H. hei- convexity for anterior dentition and continuous distal trigonid delbergensis have not been reported from East Asia (Bae, 2010), crests in the molar teeth; Xing et al., 2015). However, several other although some divergences of Yunxian crania from the standard H. features that are common to H. erectus and modern H. sapiens (Jia erectus pattern imply links to H. heidelbergensis (Rightmire, 1998; and Wei, 1976; Jia et al., 1979; Wu, 1980; Bae, 2010; Wu et al., Stringer, 2002). Whether H. heidelbergensis dispersed to East Asia 2012, 2013, 2014; Wu and Trinkaus, 2014; Xing et al., 2015) pre- remains enigmatic. Further in-depth study of Homo fossils and vented studies from affiliating the Xujiayao hominin to the classic more material are needed to assess the history of H. heidelbergensis European Neanderthals. in Asia. However, persistence of H. erectus in East Asia to at least Recent discovery of Denisovan hominins in southern Siberia 400 ka, when H. heidelbergensis was giving way to Homo nean- (Reich et al., 2010; Meyer et al., 2012; Sawyer et al., 2015) and the derthalensis in Europe (Rightmire, 1998), does not support the Sima de los Huesos hominins in Spain (Arsuaga et al., 2014; Meyer replacement of H. erectus by H. heidelbergensis in East Asia (Groves et al., 2016) prompts us to reevaluate the Xujiayao hominins from a and Lahr, 1994; Etler, 2004). Coexistence of H. heidelbergensis and H. global prospective. What the Denisovan hominins looked like re- erectus is possible if the presence of early mid-Pleistocene H. mains enigmatic due to a lack of crania, but DNA studies of their H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65 63

0 Luna Cave Xujiayao and other archaic Asian hominins with Denisovans is an Zhiren Cave (Bae et al., 2014) Modern (Liu et al., 2010b) Liujiang important and contentious new topic (Smith et al., 2017), which (Shen et al., 2002) H. sapiens 100 Huanglong Cave Daoxian must be explored intensively to develop an in-depth understanding (Shen et al., 2013) (Liu et al., 2015) of the mid-Pleistocene human evolution in eastern Eurasia. DNA 200 Jinniushan Archaic Dali Xujiayao (Xiao et al., 2002) (Rosenberg et al., 2006) (This study) study of Chinese mid-Pleistocene Homo fossils, particularly the H. sapiens Chaoxian 300 New Cave (Shen et al., 2010) Neanderthal-like Xujiayao hominin, and comparison with Deniso- or (Shen et al., 2004a) Denisovans? van fossils is crucial to testing whether Denisovans were distrib- 400 uted broadly across China in the mid-Pleistocene. Such studies will H. erectus Hexian (Grün et al., 1998) be significant in establishing the taxonomic affinity of Chinese 500 Homo fossils with transitional characteristics and for a compre- 600 hensive understanding of Asian and global mid-Pleistocene homi- Nanjing nin evolution. Until now, East Asian Homo fossils with mixed H. (Zhao et al., 2001) 700 Chenjiawo fi (An and Ho, 1989) erectus and modern human features, which cannot be af liated to classic H. erectus and modern H. sapiens, have been labelled as 800 Zhoukoudian (Shen et al., 2001, 2009) archaic H. sapiens. This makes in-depth assessment of their global 900 evolutionary position difficult and is likely to produce confusion Age (ka)

Yunxian because the same term is sometimes used to describe the earliest 1000 (Dennell, 2015) examples of H. sapiens in Africa (Dennell and Petraglia, 2012).

1100 6. Conclusions 1200 Magnetostratigraphic results indicate that the Xujiayao section 1300 records the upper Matuyama and Brunhes chrons. In agreement with its position in the middle Brunhes normal polarity chron 1400 indicated by magnetostratigraphy, the Xujiayao hominin has a 1500 pooled average quartz ESR age of 260e370 ka. Motivated by sig- Gongwangling nificant recent progress in Neanderthal lineage studies in Europe 1600 (Zhu et al., 2015) and northeastern Asia, our results shed new light on the Xujiayao Yuanmou (Zhu et al., 2008) hominin from a global perspective. Our updated age combined with 1700 Neanderthal-like traits for the Xujiayao Homo fossils, particularly Figure 10. Chronostratigraphy of well-dated modern H. sapiens (orange), archaic H. the Denisovan-like molar teeth, imply a possible affinity of the sapiens (blue), and H. erectus (green) sites from China. Whether some archaic H. sapiens Xujiayao hominin to early Denisovans, a sister group of European fossils can be assigned to Denisovans is an open question. (For interpretation of the Neanderthals that diverged from Neanderthals. Our study appears references to color in this figure legend, the reader is referred to the web version of this to support that Denisovans may have become widely distributed article.) across eastern Eurasia during the Middle Pleistocene, although clearly more Denisovan fossils and detailed DNA studies are teeth and bones indicate that they were a sister group of Nean- needed. derthals and were distributed broadly across Siberia (Reich et al., 2010; Meyer et al., 2012; Sawyer et al., 2015). Cranial and DNA Acknowledgments evidence indicates that the ~430 ka Sima de los Huesos hominins were early Neanderthals (Arsuaga et al., 2014; Meyer et al., 2016). We thank Dennis V. Kent, Robin W. Dennell and Christopher J. After divergence from a common ancestral population at ~430 ka, Lepre for suggestions that improved the paper, and to Qi Wei and Neanderthals and Denisovans possibly colonized Europe and Fei Xie for assistance during fieldwork. We also thank the anony- eastern Eurasia, respectively (Arsuaga et al., 2014; Sawyer et al., mous reviewers for their constructive comments that helped to 2015; Meyer et al., 2016). The distance from southern Siberia to improve the paper and the Associate Editor and Editor for editorial the Nihewan Basin is just over 2000 km, thus it is reasonable that handling. This study was supported financially by the Chinese early Denisovans could have expanded from high-latitude Siberia Academy of Sciences Key Research Program of Frontier Sciences southeastward to the mid-latitude Nihewan Basin that had better (QYZDB-SSW-DQC021) and Youth Innovation Promotion Program, ecological and climatic conditions. Based on the Neanderthal traits the National Natural Science Foundation of China (Grants 41174057 in the Xujiayao Homo fossils and the presence of a sister group of and 41290253), and the State Key Laboratory of Loess and Quater- Neanderthals (i.e., Denisovans) in northeastern Eurasia, it is nary Geology at the Institute of Earth Environment, Chinese possible that the Xujiayao hominins were early Denisovans, just as Academy of Sciences. the Sima de los Huesos hominins were ancestral to Neanderthals in Europe. Furthermore, molar teeth of Xujiayao hominins (Xing et al., References 2015) have massively flaring roots and relatively large and complex crowns (Fig. 3b), which are reminiscent of Denisovans (Sawyer Aitken, M.J., 1998. An Introduction to Optical Dating. Oxford University Press, et al., 2015). Our updated age of 370e260 ka for the Xujiayao Oxford. An, Z.S., Ho, C.K., 1989. New magnetostratigraphic dates of Lantian Homo erectus. hominins is also consistent with the time (~430 ka) when early Quatern. Res. 32, 213e221. Denisovans appeared and colonized eastern Eurasia (Arsuaga et al., Ao, H., Dekkers, M.J., An, Z.S., Xiao, G.Q., Li, Y.X., Zhao, H., Qiang, X.K., Chang, H., 2014; Sawyer et al., 2015; Meyer et al., 2016). After divergence of Chang, Q.H., Wu, D.C., 2013a. Magnetostratigraphic evidence of a mid-Pliocene onset of the Nihewan Formationeimplications for early fauna and hominid Neanderthals and Denisovans at ~430 ka (Meyer et al., 2016), occupations in the Nihewan Basin, North China. Quatern. Sci. Rev. 59, 30e42. eastern Eurasian Denisovans had not only Neanderthal traces but Ao, H., Dekkers, M.J., Wei, Q., Qiang, X.K., Xiao, G.Q., 2013b. New evidence for early also developmentally derived features. This could explain why the presence of hominids in North China. Sci. Rep. 3, 2403. http://dx.doi.org/ 10.1038/srep02403. Xujiayao hominins had mixed Neanderthal and derived modern Ao, H., An, Z.S., Dekkers, M.J., Li, Y.X., Xiao, G.Q., Zhao, H., Qiang, X.K., 2013c. human characteristics. Understanding potential relationships of Pleistocene magnetochronology of the fauna and Paleolithic sites in the 64 H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65

Nihewan Basin: significance for environmental and hominin evolution in North Liu, C.R., Yin, G.M., Deng, C.L., Han, F., Song, W.J., 2014a. ESR dating of the Majuangou China. Quatern. Geochronol. 18, 78e92. and Banshan Paleolithic sites in the Nihewan Basin, North China. J. Hum. Evol. Arsuaga, J.L., Martínez, I., Arnold, L.J., Aranburu, A., Gracia-Tellez, A., Sharp, W.D., 73, 58e63. Quam, R.M., Falgueres, C., Pantoja-Perez, A., Bischoff, J., Poza-Rey, E., Pares, J.M., Liu, W., Wu, X.J., Xing, S., Zhang, Y.Y., 2014b. Human Fossils in China. Science Press, Carretero, J.M., Demuro, M., Lorenzo, C., Sala, N., Martinon-Torres, M., García, N., Beijing (in Chinese). Alcazar de Velasco, A., Cuenca-Bescos, G., Gomez-Olivencia, A., Moreno, D., Liu, W., Martinon-Torres, M., Cai, Y.J., Xing, S., Tong, H.W., Pei, S.W., Sier, M.J., Pablos, A., Shen, C.C., Rodríguez, L., Ortega, A.I., García, R., Bonmatí, A., Ber- Wu, X.H., Edwards, R.L., Cheng, H., Li, Y.Y., Yang, X.X., Bermúdez de Castro, J.M., múdez de Castro, J.M., Carbonell, E., 2014. Neandertal roots: cranial and chro- Wu, X.J., 2015. The earliest unequivocally modern humans in southern China. nological evidence from Sima de los Huesos. Science 344, 1358e1363. Nature 526, 696e699. Bae, C.J., 2010. The late Middle Pleistocene hominin fossil record of eastern Asia: Løvlie, R., Su, P., Fan, X.Z., Zhao, Z.J., Liu, C., 2001. A revised paleomagnetic age of the synthesis and review. Yearb. Phys. Anthropol. 53, 75e93. Nihewan Group at the Xujiayao Palaeolithic site, China. Quatern. Sci. Rev. 20, Bae, C.J., Wang, W., Zhao, J.X., Huang, S.M., Tian, F., Shen, G.J., 2014. Modern human 1341e1353. teeth from Late Pleistocene Luna Cave (Guangxi, China). Quatern. Intl. 354, Ma, N., 2009. A Preliminary Study on the Stone Artifacts Excavated from the 169e183. Xujiayao Paleolithic Site in 1977, Nihewan Basin, North China. Thesis of the Barbour, G.B., 1924. Preliminary observation in Kalgan area. Bull. Geol. Soc. China 3, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy 167e168. of Sciences 67 (In Chinese with English Abstract). Beerten, K., Lomax, J., Clemer, K., Stesmans, A., Radtke, U., 2006. On the use of Ti Meyer, M., Kircher, M., Gansauge, M.T., Li, H., Racimo, F., Mallick, S., Schraiber, J.G., centres for estimating burial ages of Pleistocene sedimentary quartz: multiple- Jay, F., Prüfer, K., de Filippo, C., Sudmant, P.H., Alkan, C., Fu, Q.M., Do, R., grain data from Australia. Quatern. Geochronol. 1, 151e158. Rohland, N., Tandon, A., Siebauer, M., Green, R.E., Bryc, K., Briggs, A.W., Bischoff, J.L., Shamp, D.D., Aramburu, A., Arsuaga, J.L., Carbonell, E., Bermúdez de Stenzel, U., Dabney, J., Shendure, J., Kitzman, J., Hammer, M.F., Shunkov, M.V., Castro, J.M., 2003. The Sima de los Huesos hominids date to beyond U/Th Derevianko, A.P., Patterson, N., Andres, A.M., Eichler, E.E., Slatkin, M., Reich, D., equilibrium (>350 kyr) and perhaps to 400e500 kyr: new radiometric dates. Kelso, J., Pa€abo,€ S., 2012. A high-coverage genome sequence from an archaic J. Archaeol. Sci. 30, 275e280. Denisovan individual. Science 338, 222e226. Chen, T., Yuan, S., Gao, S., 1982. Uranium series dating of mammalian fossils at Meyer, M., Arsuaga, J.L., de Filippo, C., Nagel, S., Aximu-Petri, A., Nickel, B., Martínez, I., Xujiayao Palaeolithic site. Acta Anthropol. Sinica 1, 91e95 (in Chinese). Gracia, A., Bermúdez de Castro, J.M., Carbonell, E., Viola, B., Kelso, J., Prüfer, K., Chen, T., Yuan, S., Gao, S., 1984. The study on uranium series dating of fossil bones Pa€abo,€ S., 2016. Nuclear DNA sequences from the Middle Pleistocene Sima de los and an absolute age sequence for the main Palaeolithic sites of North China. Huesos hominins. Nature 531, 504e507. http://dx.doi.org/10.1038/nature17405. Acta Anthropol. Sinica 3, 259e269 (in Chinese). Michel, V., Valladas, H., Shen, G.J., Wang, W., Zhao, J.X., Shen, C.C., Valensi, P., Deng, C.L., Zhu, R.X., Zhang, R., Ao, H., Pan, Y.X., 2008. Timing of the Nihewan for- Bae, C.J., 2016. The earliest modern Homo sapiens in China? J. Hum. Evol. 101, mation and faunas. Quatern. Res. 69, 77e90. 101e104. Dennell, R., 2010. Early Homo sapiens in China. Nature 468, 512e513. Norton, C.J., Gao, X., 2008. Hominin-carnivore interactions during the Chinese Early Dennell, R., Petraglia, M.D., 2012. The dispersal of Homo sapiens across southern Paleolithic: taphonomic perspectives from Xujiayao. J. Hum. Evol. 55, 164e178. Asia: how early, how often, how complex? Quatern. Sci. Rev. 47, 15e22. Prescott, J.R., Hutton, J.T., 1988. Cosmic ray and gamma ray dosimetry for TL and Dennell, R.W., 2015. The Jaramillo subchron, a geochronological marker horizon in ESR. Nucl. Tracks Rad. Meas. 14, 223e227. the palaeoanthropological record of China. Quatern. Intl. 389, 241e254. Qiu, J., 2016. The forgotten continent. Nature 535, 218e220. Etler, D.A., 2004. Homo erectus in East Asia: human ancestor or evolutionary dead- Reich, D., Green, R.E., Kircher, M., Krause, J., Patterson, N., Durand, E.Y., Viola, B., end? Athena Rev. 4, 37e50. Briggs, A.W., Stenzel, U., Johnson, P.L.F., Maricic, T., Good, J.M., Marques- Grün, R., Huang, P.H., Huang, W.P., McDermott, F., Thorne, A., Stringer, C.B., Yan, G., Bonet, T., Alkan, C., Fu, Q.M., Mallick, S., Li, H., Meyer, M., Eichler, E.E., 1998. ESR and U-series analyses of teeth from the palaeoanthropological site of Stoneking, M., Richards, M., Talamo, S., Shunkov, M.V., Derevianko, A.P., Hexian, Anhui Province, China. J. Hum. Evol. 34, 555e564. Hublin, J.J., Kelso, J., Slatkin, M., Pa€abo,€ S., 2010. Genetic history of an archaic Grün, R., Eggins, S., Kinsley, L., Moseley, H., Sambridge, M., 2014. Laser ablation U- hominin group from Denisova Cave in Siberia. Nature 468, 1053e1060. series analysis of fossil bones and teeth. Palaeogeogr. Palaeoclimatol. Palae- Rightmire, G.P., 1998. Human evolution in the Middle Pleistocene: the role of Homo oecol. 416, 150e167. heidelbergensis. Evol. Anthropol. 6, 218e227. Groves, C.P., Lahr, M.M., 1994. A bush not a ladder: speciation and replacement in Rightmire, G.P., 2008. Homo in the Middle Pleistocene: hypodigms, variation, and human evolution. Persp. Hum. Biol. 4, 1e11. species recognition. Evol. Anthropol. 17, 8e21. Guerin, G., Mercier, N., 2011. Determining gamma dose rates by field gamma Rightmire, G.P., 2009. Middle and later Pleistocene hominins in Africa and South- spectroscopy in sedimentary media: results of Monte Carlo simulations. Rad. west Asia. Proc. Natl. Acad. Sci. USA 106, 16046e16050. Meas. 46, 190e195. Rightmire, G.P., 2013. Homo erectus and Middle Pleistocene hominins: brain size, Guerin, G., Mercier, N., Adamiec, G., 2011. Dose-rate conversion factors: update. skull form, and species recognition. J. Hum. Evol. 65, 223e252. Ancient TL 29, 5e8. Rink, W.J., Bartoll, J., Schwarcz, H.P., Shane, P., Bar-Yosef, O., 2007. Testing the reli- Hayase, R., 2012. 14C AMS dating at the Houjiayao Site, the Xibaimaying Site and the ability of ESR dating of optically exposed buried quartz sediments. Rad. Meas. Jijitan Peat Layers: the results and the issues. In: Matsufuji, K. (Ed.), Basic 42, 1618e1626. studies of the Paleoenvironmental changes and Paleolithic chronology in Roberts, A.P., 2008. Geomagnetic excursions: knowns and unknowns. Geophys. Res. Northeast Asia. Shinyosha Press, Tokyo, pp. 89e100. Lett. 35, L17307 doi:17310.11029/12008GL034719. Hilgen, F.J., Lourens, L.J., van Dam, J.A., 2012. The Neogene Period. In: Gradstein, F.M., Rosenberg, K.R., Zune, L., Ruff, C.B., 2006. Body size, body proportions, and Ogg, J.G., Schmitz, M., Ogg, G. (Eds.), The Geologic Time Scale 2012. Elsevier, encephalization in a Middle Pleistocene archaic human from northern China. Amsterdam, pp. 923e978. Proc. Natl. Acad. Sci. USA 103, 3552e3556. Institute of Archaeology of Chinese Academy of Social Sciences, 1991. Radiocarbon Sawyer, S., Renaud, G., Viola, B., Hublin, J.J., Gansauge, M.T., Shunkov, M.V., dates in Chinese archaeology 1965e1991. Cultural Relics Publishing House, Derevianko, A.P., Prufer, K., Kelso, J., Pa€abo,€ S., 2015. Nuclear and mitochondrial Beijing (In Chinese). DNA sequences from two Denisovan individuals. Proc. Natl. Acad. Sci. USA 112, Jia, L.P., Wei, Q., 1976. Xujiayao Palaeolithic Site, in Yanggao County, Shanxi Province. 15696e15700. Archaeol. Bull. 2, 97e114 (In Chinese). Shen, G.J., Ku, T.L., Cheng, H., Edwards, R.L., Yuan, Z.X., Wang, Q., 2001. High-pre- Jia, L.P., Wei, Q., Li, C.R., 1979. Report on the excavation of Xujiayao (Hsuchiayao) cision U-series dating of Locality 1 at Zhoukoudian, China. J. Hum. Evol. 41, Man site in 1976. Vertebrat. Palasiatic 17, 277e293 (In Chinese). 679e688. Jones, C.H., 2002. User-driven integrated software lives: “Paleomag” paleomag- Shen, G.J., Wang, W., Wang, Q., Zhao, J.X., Collerson, K., Zhou, C.L., Tobias, P.V., 2002. netics analysis on the Macintosh. Comp. Geosci. 28, 1145e1151. U-Series dating of Liujiang hominid site in Guangxi, Southern China. J. Hum. Kirschvink, J.L., 1980. The least-squares line and plane and the analysis of palae- Evol. 43, 817e829. omagnetic data. Geophys. J. Roy. Astro. Soc. 62, 699e718. Shen, G.J., Cheng, H., Edwards, R.L., 2004a. Mass spectrometric U-series dating of Li, Z.T., Xu, Q.H., Zhang, S.R., Hun, L.Y., Li, M.Y., Xie, F., Wang, F.G., Liu, L.Q., 2014. New Cave at Zhoukoudian, China. J. Archaeol. Sci. 31, 337e342. Study on stratigraphic age, climate changes and environment background of Shen, G.J., Gao, X., Zhao, J.X., Collerson, K.D., 2004b. U-series dating of Locality 15 at Houjiayao site in Nihewan Basin. Quatern. Intl. 349, 42e48. Zhoukoudian, China, and implications for hominid evolution. Quatern. Res. 62, Liu, C.R., Grün, R., 2011. Fluvio-mechanical resetting of the Al and Ti centres in 208e213. quartz. Rad. Meas. 46, 1038e1042. Shen, G.J., Gao, X., Granger, D.E., 2009. Age of Zhoukoudian Homo erectus deter- Liu, C.R., Yin, G.M., Gao, L., Bahain, J.J., Li, J.P., Lin, M., Chen, S.M., 2010a. ESR dating of mined with 26Al/10Be burial dating. Nature 458, 198e200. Pleistocene archaeological localities of the Nihewan Basin, North Chinaepreli- Shen, G.J., Fang, Y.S., Bischoff, J.L., Feng, Y.X., Zhao, J.X., 2010. Mass spectrometric U- minary results. Quatern. Geochronol. 5, 385e390. series dating of the Chaoxian hominin site at Yinshan, eastern China. Quatern. Liu, W., Jin, C.Z., Zhang, Y.Q., Cai, Y.J., Xing, S., Wu, X.J., Cheng, H., Edwards, R.L., Intl. 211, 24e28. Pan, W.S., Qin, D.G., An, Z.S., Trinkaus, E., Wu, X.Z., 2010b. Human remains from Shen, G.J., Wu, X.Z., Wang, Q., Tu, H., Feng, Y.X., Zhao, J.X., 2013. Mass spectrometric Zhirendong, South China, and modern human emergence in East Asia. Proc. U-series dating of Huanglong Cave in Hubei Province, central China: evidence Natl. Acad. Sci. USA 107, 19201e19206. for early presence of modern humans in eastern Asia. J. Hum. Evol. 65, 162e167. Liu, C.R., Yin, G.M., Fang, F., Voinchet, P., Deng, C.L., Han, F., Li, J.P., Song, W.J., Shen, G.J., Tu, H., Xiao, D.F., Qiu, L.C., Feng, Y.X., Zhao, J.X., 2014. Age of Maba Wang, D., Bahain, J.J., 2013. ESR dating of the Donggutuo Palaeolithic site in the hominin site in southern China: evidence from U-series dating of Southern Nihewan Basin, northern China. Geochronometria 40, 348e354. Branch Cave. Quatern. Geochronol. 23, 56e62. H. Ao et al. / Journal of Human Evolution 106 (2017) 54e65 65

Smith, F.H., Ahern, J.C.M., Jankovic, I., Karavanic, I., 2017. The Assimilation Model of among Pleistocene Homo. Plos One 8, e59587. http://dx.doi.org/10.1371/ modern human origins in light of current genetic and genomic knowledge. journal.pone.0059587. Quatern. Intl. (in press). Wu, X.J., Crevecoeur, I., Liu, W., Xing, S., Trinkaus, E., 2014. Temporal labyrinths of Stringer, C., 2002. Modern human origins: progress and prospects. Phil. Trans. Roy. eastern Eurasian Pleistocene humans. Proc. Natl. Acad. Sci. USA 111, Soc. B 357, 563e579. 10509e10513. Sun, Y.B., Clemens, S.C., An, Z.S., Yu, Z.W., 2006. Astronomical timescale and Wu, X.Z., Poirier, F.E., 1995. Human Evolution in China. Oxford University Press, palaeoclimatic implication of stacked 3.6-Myr monsoon records from the Chi- Oxford. nese Loess Plateau. Quatern. Sci. Rev. 25, 33e48. Xiao, J., Jin, C.Z., Zhu, Y.Z., 2002. Age of the fossil Dali Man in north-central China Tissoux, H., Falgueres, C., Voinchet, P., Toyoda, S., Bahain, J.J., Despriee, J., 2007. deduced from chronostratigraphy of the loess-paleosol sequence. Quatern. Sci. Potential use of Ti-center in ESR dating of fluvial sediment. Quatern. Geo- Rev. 21, 2191e2198. chronol. 2, 367e372. Xing, S., Martinon-Torres, M., Bermúdez de Castro, J.M., Wu, X.J., Liu, W., 2015. Toyoda, S., Voinchet, P., Falgueres, C., Dolo, J.M., Laurent, M., 2000. Bleaching of ESR Hominin teeth from the early Late Pleistocene site of Xujiayao. Northern China. signals by the sunlight: a laboratory experiment for establishing the ESR dating Am. J. Phys. Anthropol. 156, 224e240. of sediments. Appl. Rad. Isotopes 52, 1357e1362. Yokoyama, Y., Falgueres, C., Quaegebeur, J.P., 1985. ESR dating of quartz from Qua- Tu, H., Shen, G.J., Li, H.X., Xie, F., Granger, D.E., 2015. 26Al/10Be burial dating of ternary sediments: first attempt. Nucl. Tracks 10, 921e928. Xujiayao-Houjiayao site in Nihewan Basin, northern China. Plos One 10, Yuan, B.Y., Xia, Z.K., Niu, P.S., 2011. Nihewan Rift and Early Man. Geological Pub- e0118315. http://dx.doi.org/10.1371/journal.pone.0118315. lishing House, Beijing (In Chinese). Voinchet, P., Falgueres, C., Tissoux, H., Bahain, J.J., Despriee, J., Pirouelle, F., 2007. ESR Zander, A., Hilgers, A., 2013. Potential and limits of OSL, TT-OSL, IRSL and pIRIR290 dating of fluvial quartz: estimate of the minimal distance transport required for dating methods applied on a Middle Pleistocene sediment record of Lake getting a maximum optical bleaching. Quatern. Geochronol. 2, 363e366. El'gygytgyn, Russia. Climate of the Past 9, 719e733. Voinchet, P., Despriee, J., Tissoux, H., Falgueres, C., Bahain, J.J., Gageonnet, R., Zhao, J.X., Hu, K., Collerson, K.D., Xu, H.K., 2001. Thermal ionization mass spec- Depont, J., Dolo, J.M., 2010. ESR chronology of alluvial deposits and first human trometry U-series dating of a hominid site near Nanjing, China. Geology 29, settlements of the Middle Loire Basin (Region Centre, France). Quatern. Geo- 27e30. chronol. 5, 381e384. Zhu, R.X., Potts, R., Xie, F., Hoffman, K.A., Deng, C.L., Shi, C.D., Pan, Y.X., Wang, H.Q., Wang, X.S., Løvlie, R., Su, P., Fan, X.Z., 2008. Magnetic signature of environmental Shi, R.P., Wang, Y.C., Shi, G.H., Wu, N.Q., 2004. New evidence on the earliest change reflected by Pleistocene lacustrine sediments from the Nihewan Basin, human presence at high northern latitudes in northeast Asia. Nature 431, North China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 260, 452e462. 559e562. Wu, M.L., 1980. Humans fossils discovered at Xujiayao site in 1977. Vertebrat. Pal- Zhu, R.X., Potts, R., Pan, Y.X., Yao, H.T., Lu, L.Q., Zhao, X., Gao, X., Chen, L.W., Gao, F., asiatic. 18, 227e238 (In Chinese). Deng, C.L., 2008. Early evidence of the genus Homo in East Asia. J. Hum. Evol. 55, Wu, R.K., Ren, M.E., Zhu, X.M., Yang, Z.G., Hu, C.K., Kong, S.H., Xie, Y.Y., Zhao, S.S., 1075e1085. 1985. Multi-disciplinary Study of the Site at Zhoukoudian. Science Zhu, Z.Y., Dennell, R., Huang, W.W., Wu, Y., Rao, Z.G., Qiu, S.F., Xie, J.B., Liu, W., Press, Beijing (In Chinese). Fu, S.Q., Han, J.W., Zhou, H.Y., Yang, T.P.O., Li, H.M., 2015. New dating of the Wu, X.J., Trinkaus, E., 2014. The Xujiayao 14 mandibular ramus and Pleistocene Homo erectus cranium from Lantian (Gongwangling), China. J. Hum. Evol. 78, Homo mandibular variation. CR Pale vol. 13, 333e341. 144e157. Wu, X.J., Maddux, S.D., Pan, L., Trinkaus, E., 2012. Nasal floor variation among Zijderveld, J.D.A., 1967. AC demagnetization of rocks: analysis of results. In: eastern Eurasian Pleistocene Homo. Anthropol. Sci. 120, 217e226. Collinson, D.W., Creer, K.M., Runcorn, S.K. (Eds.), Methods in Paleomagnetism. Wu, X.J., Xing, S., Trinkaus, E., 2013. An enlarged parietal foramen in the late Archaic Elsevier, New York, pp. 254e286. Xujiayao 11 neurocranium from northern China, and rare anomalies