Interplay of dietary diversity and mobility of the Late Neolithic people in the Central Plains of China: multi-isotope analysis at the Haojiatai site

Abstract

The origins of agriculture in East Asia can be distinguished between rice agriculture in southern China and millet agriculture (foxtail and broomcorn millet) in the north. In the Longshan period (c. 4300–3800 cal BP), with the northward expansion of rice farming, a mixed rice–millet agricultural area was established in the Central Plains of China, which was also in the core area of the establishment of the early state. The interaction between these two agricultural traditions is significant as a backdrop for the emergence of the Erlitou civilization, although the specifics of this cultural contact remain unclear. In this study, we conducted a multiple-isotope analysis to investigate the relationship between dietary diversity and human mobility at the Haojiatai site in Henan Province, where previous studies suggest that C₄ millet farmers and C₃ hunter-gatherer-fishers coexisted in this place. In addition to assessing carbon and nitrogen isotope ratios of bone collagen, the results of nitrogen isotope ratios of individual amino acids suggested that the C₃-diet group primarily relied on terrestrial resources such as rice and wild herbivores but also showed a slight influence of C₄ food signature. Meanwhile, the C₄-diet group was also influenced by C₃ foods (wild herbivores and plants) and aquatic animals. Oxygen and strontium isotope ratios indicated that although there is no evidence of long-distance migrations, it does suggest an interaction with the core area of the Central Plains. C₃-diet individuals likely originated from non-local regions. A considerable number of C₄ millet farmers also appeared to have migrated from the surrounding area as well. The high mobility observed among populations of the Longshan culture implies a connection between the emergence of civilization and increased human mobility.

Introduction

The establishment of agriculture and the spread of farming played a crucial role in the evolution of human societies, providing a stable economic foundation for complex societies (Childe, 1936). Agriculture is widely recognized to have had multiple origins worldwide, with domesticated crops having spread outward from their original centers (Price and Bar-Yosef, 2011). The Farming/Language Dispersal Hypothesis (Bellwood and Renfrew, 2002; Bellwood, 2011, 2005) suggests that the diffusion of language families was catalyzed by the migration of farmers and the introduction of agriculture. As agriculture facilitated population growth, agrarian communities and their languages expanded over larger geographical areas (Renfrew, 1990, 2002; Bellwood, 2001, 2005; Diamond and Bellwood, 2003). These opinions have been supported by the genomic approach applied in several regions, such as Europe (Dupanloup et al., 2004; Lazaridis et al., 2014, 2016), North Africa (Simões et al., 2023), and Oceania (Oliveira et al., 2022).

In East Asia, however, the situation is more complicated due to the presence of two distinct agricultural origins and their mutual influence. Archaeological evidence indicates that millet (Panicum miliaceum and Setaria italica) and rice (Oryza sativa) farming originated independently in the Yellow River basin and the middle and lower reaches of the Yangtze River in China dating back at least 10000 years (Zhang and Yuan, 1988; Zhao, 1998; Jiang and Liu, 2006; Liu et al., 2007). As rice farming expanded northward, mixed rice–millet farming emerged in the Huai River valley (Luo et al., 2019) and also appeared in the Central Plain area during the Peiligang culture period (9000–7000 cal BP) (Zhang et al., 2012; Wang et al., 2017; Bestel et al., 2018). Recent ancient DNA studies of Southeast Asia suggest that the spread of agriculture involved a mixture of local hunter-gatherers and incoming farmers associated with the Neolithic expansion from South China (Kutanan et al., 2018; Lipson et al., 2018; McColl et al., 2018). Other studies propose that Yellow River farmers could be the common ancestor of both Tibetans and Han Chinese, with their westward migration potentially contributing to the spread of the Sino-Tibetan language family (Zhang et al., 2019b; Wang et al., 2021). Moreover, although controversial, some studies explore the influence of northern Chinese farmers on the larger spread of Transeurasian languages (Robbeets et al., 2021).

Genetic evidence suggests that agricultural demic expansions played a significant role in the spread of agriculture in East Asia (Yang et al., 2020; Stoneking et al., 2023). However, the genetic background of Neolithic rice farmers in the Yangtze River Basin remains unknown (Liu et al., 2021), due to the poor preservation of ancient DNA in South China, making it challenging to ascertain the role of rice farmers in this historical context. Additionally, the genetic differences between millet farmers from the north and rice farmers from the south make it difficult to determine how these two societies interacted genetically and culturally during the spread of agriculture.

The Central Plains region, situated at the confluence of Northern and Southern China, exhibits unique agricultural development characteristics and is recognized as a significant birthplace of Chinese civilization (Zhang et al., 2019a). Millet and rice agriculture had already begun to develop separately during the Peiligang period (c. 8000 BP) in Henan, and the earliest evidence of domesticated foxtail millet (Setaria italica) was found at the Zhuzai site (Bestel et al., 2018), located in the core area of the Central Plains. During the same period, a large number of domesticated rice remains were found in Jiahu, located in southern Henan. (Zhao and Zhang, 2009; Zhang et al., 2018). Isotopic analysis results also indicated that the inhabitants of Jiahu primarily consumed C₃ foods (i.e. plants and animals in ecosystems based on C₃ plants, including rice), suggesting a dietary transition from hunting and gathering to fishing, rice farming, and livestock management (Hu et al., 2006; Wu et al., 2015). During the subsequent Yangshao period (7000–5000 cal BP), the isotope signature of human bones and plant flotation from the sites within this timeframe suggests the development of livelihood based on millet agriculture in the Central Plains region (Zhao, 2017, 2019; Chen, 2021). Furthermore, during the Longshan period (c. 4300–3800 cal BP), dietary differentiation was observed at the sites of Wadian, Meishan, and Haojiatai. Humans are segregated into two distinct dietary groups (hereafter referred to as the C₃-diet group and the C₄-diet group) coexisting at the same archaeological sites (Chen et al., 2016; Zhou, 2017; Li et al., 2021). The coexistence of different cultures, as suggested by the presence of two dietary traditions in a population, may have contributed to the emergence of the Erlitou culture (3800–3500 cal BP), which is considered the beginning of Chinese civilization.

The Longshan culture introduced domesticated sheep and cattle from West Asia through close interaction with neighboring regions (Yuan, 2010). Although millets were the primary staple in this region, the significant presence of rice at the Wadian site (Liu and Fang, 2010; Liu et al., 2018), along with the observed divergence in dietary habits potentially linked to it, as well as distinctive burial practices, suggest the interaction between the Central Plains and the Yangtze River basin (Zhang, 2015). Consequently, dietary variation has been interpreted in relation to social stratification (Zhou, 2017). However, comparable dietary diversity may also result from active human mobility, such as migration from rice-farming areas, although no isotope studies of human provenance have been conducted at Haojiatai. Hence, this study focuses on the Haojiatai site, to explore the correlation between food habits and the geographical origin where an individual spent their childhood. By combining multi-isotope analysis of human bones and tooth enamel, we aim to interpret in the context of mixed rice–millet farming to improve understanding of the interaction between the two Chinese agricultural systems.

Materials and Methods

The carbon and nitrogen isotope ratios of 13 humans and 52 animals from the Haojiatai site have been studied previously (Zhou, 2017; Li et al., 2021; Li, 2022). In this study, we supplemented 6 humans, 31 animals, and 4 charred rice to enhance understanding of the dietary complexity during this period. Furthermore, we reanalyzed the collagen carbon and nitrogen isotopes of 13 previously published human samples to align the isotope data of collagen with individual amino acids with newly analyzed six samples (in sum, 19 humans). Additionally, we conducted oxygen and strontium isotope analyses to explore mobility patterns. In total, 48 samples were analyzed from the Haojiatai site (Table 1).

Table 1.

Data for the samples measured in this study

Sample ID	Species		Sex	Age (years)		Bone	Tooth	Period
M101	Human	Homo sapiens	Unknown	4	± 1	Limb	—	Late Longshan
M102	Human	Homo sapiens	Female	12	± 2.5	Limb	URM1	Late Longshan
M103	Human	Homo sapiens	Unknown	8	± 2	Limb	LLM1	Late Longshan
M104	Human	Homo sapiens	Male	55	±	Limb	ULI1	Late Longshan
M105	Human	Homo sapiens	Unknown	15	± 2	Limb	—	Late Longshan
M106	Human	Homo sapiens	Male	15	– 17	Limb	—	Early and Middle Longshan
M107	Human	Homo sapiens	Male	9	± 2	Limb	LLC	Early and Middle Longshan
M108	Human	Homo sapiens	Male	40	±	Limb	—	Early and Middle Longshan
M109	Human	Homo sapiens	Male	15	± 2	Limb	ULC	Early and Middle Longshan
M112	Human	Homo sapiens	Unknown	adult		Limb	—	Early and Middle Longshan
M113	Human	Homo sapiens	Male	60+		Limb	LLM3	Early and Middle Longshan
M114	Human	Homo sapiens	—	15	± 2	Limb	ULC	Early and Middle Longshan
M116	Human	Homo sapiens	Female	40	±	Limb	LLM1	Early and Middle Longshan
M117	Human	Homo sapiens	Female	60	±	Limb	URP2	Early and Middle Longshan
M118	Human	Homo sapiens	Female	40	±	Limb	LLM1	Early and Middle Longshan
M119	Human	Homo sapiens	Male	50	– 55	Limb	URI1	Early and Middle Longshan
M120	Human	Homo sapiens	Female	50	– 55	Limb	URP2	Early and Middle Longshan
H554	Human	Homo sapiens	Male	40	– 45	Limb	ULC	Late Longshan
H608	Human	Homo sapiens	Female	30	±	Limb	URM3	Early and Middle Longshan
TG1_9	Pig	Sus domestica	—	—		Limb	—	Early and Middle Longshan
H671_1	Pig	Sus domestica	—	—		Limb	—	Early and Middle Longshan
T8339_11	Pig	Sus domestica	—	—		Limb	—	Early and Middle Longshan
T8339_1	Pig	Sus domestica	—	—		Limb	—	Early and Middle Longshan
H506_1	Cattle	Bos taurus	—	—		Limb	—	Late Longshan
H609_1	Cattle	Bos taurus	—	—		Limb	—	Late Longshan
H619_1	Cattle	Bos taurus	—	—		Limb	—	Early and Middle Longshan
H527_1	Cattle	Bos taurus	—	—		Limb	—	Longshan
H501_1	Dog	Canis familiars	—	—		Limb	—	Late Longshan
H502_2	Dog	Canis familiars	—	—		Limb	—	Late Longshan
H527_1	Dog	Canis familiars	—	—		Limb	—	Longshan
H627_1	Dog	Canis familiars	—	—		Limb	—	Early and Middle Longshan
H556_1	Pere David’s deer	Elaphurus davidianus	—	—		Limb	—	Early and Middle Longshan
H521_1	Pere David’s deer	Elaphurus davidianus	—	—		Limb	—	Late Longshan
TG1_3	Pere David’s deer	Elaphurus davidianus	—	—		Limb	—	Early and Middle Longshan
H619_1	Sika deer	Cervus nippon	—	—		Limb	—	Early and Middle Longshan
T8339_11	Sika deer	Cervus nippon	—	—		Limb	—	Early and Middle Longshan
H538_1	Sika deer	Cervus nippon	—	—		Limb	—	Late Longshan
H622_1	River deer	Hydropotes inermis	—	—		Limb	—	Early and Middle Longshan
H523_1	River deer	Hydropotes inermis	—	—		Limb	—	Longshan
H671_2	River deer	Hydropotes inermis	—	—		Limb	—	Early and Middle Longshan
H714_1	Snakehead fish	Channa argus	—	—		Vertebrae	—	Early and Middle Longshan
H540_1_1	Snakehead fish	Channa argus	—	—		Vertebrae	—	Longshan
H671_2	Catfish	Clarias fuscus	—	—		Vertebrae	—	Early and Middle Longshan
H540_1_2	Yellow catfish	Pelteobagrus fulvidraco	—	—		Vertebrae	—	Longshan
H538_1	Tortoise	Testudinidae	—	—		Plastron	—	Late Longshan
F58_1	Tortoise	Testudinidae	—	—		Plastron	—	Early and Middle Longshan
H619_1	Soft-shelled turtle	Trionychidae	—	—		Plastron	—	Early and Middle Longshan
T8339_12	Soft-shelled turtle	Trionychidae	—	—		Plastron	—	Early and Middle Longshan
T8339_6b	Shell	—	—	—		—	—	Early and Middle Longshan
H564_2	Shell	—	—	—		—	—	Early and Middle Longshan

Haojiatai site

The Haojiatai site (114° 01' 56.9" E, 33° 35' 57.4" N) is situated in Luohe City, Henan Province, approximately 1.5 km south of the Sha River and 8 km north of the Yin River, with an elevation of 60 m above sea level. The geological environment of the site is relatively uniform, predominantly consisting of Quaternary deposits (Figure 1). Excavations were conducted on two occasions: initially by the Henan Provincial Institute of Cultural Relics and Archaeology (HPICRA) during the construction of the Jing–Guang railway in 1986–1987, and subsequently in 2016–2017 by a collaborative effort involving the School of Archaeology and Museology at Peking University (SAM at PKU), HPIRA, and Luohe Municipal Institute of Cultural Relics and Archaeology (HPICRA, 2012; SAM at PKU et al., 2017).

Figure 1.

Simplified geological map of the study area, showing the location of the Haojiatai site and other related sites mentioned in this paper. The geological map was modified from Ma (2020). Maps throughout this paper were created using ArcGIS Pro software by Esri.

Excavations have revealed that the Haojiatai site is primarily attributed to the Longshan period, including its early and middle stage to late stage, also known as the Xinzhai period. The central area of the site is considered an ancient city site of the Central Longshan culture (SAM at PKU et al., 2017). Analysis of excavated plant remains at Haojiatai indicated that 90% of the plant remains from the Longshan period at the Haojiatai site were crops, with foxtail millet being the principal agricultural crop, followed by broomcorn millet, and rice and soybeans present in smaller quantities (Deng et al., 2021). In addition, the domestic animals found include the domestic pig, which originated indigenously, as well as sheep and cattle originated in West Asia. On the other hand, the abundance of wild mammal bones, fish bones, and shellfish also suggests that not only domestic animal management but also hunting and fishing played an important role in the livelihood of the inhabitants (HPICRA, 2012).

Carbon and nitrogen isotope ratios in bone collagen

Dietary reconstruction through carbon and nitrogen isotope analyses of bone collagen has been widely utilized (Ambrose and DeNiro, 1986). Different photosynthetic pathways result in distinct δ¹³C values, lower in C₃ and higher in C₄ plants (O’Leary, 1988). Moreover, marine and terrestrial ecosystems exhibit different carbon sources, enabling differentiation between marine and terrestrial food sources (Schwarcz and Schoeninger, 1991). Nitrogen isotope ratios in organisms reflect their trophic levels (i.e. their position within the food chain). δ¹⁵N values tend to be higher by 3–4‰ per trophic level up the food chain (Deniro and Epstein, 1981; Minagawa and Wada, 1984). The δ¹³C and δ¹⁵N values in bone collagen reflect the components, averaging the isotope signatures of consumed foods (Ambrose and Norr, 1993). Furthermore, due to the slow turnover rate of bone collagen, especially in limb bones, this signature retains a long-term dietary record over the decade preceding an individual’s death (Hedges et al., 2007).

Collagens were extracted from bones through gelatinization, following an improved method established in previous studies (Longin, 1971; Yoneda et al., 2002). Approximately 0.5 g of bone samples were initially cleaned using a sandblaster and ultrasonication. Subsequently, bone fragments were placed in small cellulose tubes and immersed in 0.4 M HCl for two nights until complete decalcification. Following decalcification, samples were neutralized with Milli-Q water and then treated with 0.1 M NaOH for up to 1 hour to remove contaminants such as humic and fulvic acids. After rinsing with Milli-Q water, the remnants were heated in 0.0001 M HCl (pH 4) at 90°C for two nights for gelatinization. The resultant dissolved gelatin was filtered and freeze-dried to obtain gelatine samples.

Carbon (%C) and nitrogen (%N) contents and atomic C:N ratios were computed to assess collagen preservation. Collagen failing to meet the conditions of a C:N ratio from 2.9% to 3.6%, a %C of 13% or higher, and a %N of 4.8% or higher may indicate negligible influence of diagenesis or contamination (DeNiro, 1985; Van Klinken, 1999). Approximately 0.4 mg of each extracted gelatin was weighed and placed in precleaned tin capsules. Stable isotope measurements were conducted with a Thermo Scientific EA Flash 2000 coupled to a Delta V Advantage isotope mass spectrometer (EA-IRMS) at The University Museum, The University of Tokyo (UMUT). The isotope values of all samples were measured relative to the laboratory standards of l-alanine, glycine, and l-histidine. Carbon and nitrogen isotope values are reported relative to the Vienna Pee Dee Belemnite (VPDB) and AIR, respectively, and the analytical precision is ±0.1‰ (1σ). The pretreatment of charred rice samples followed the method described by Brinkkemper et al. (2018). The samples were measured by an EA Flash coupled to a Thermo Scientific 253 plus isotope ratio mass spectrometer in the Stable Isotope Analysis Lab of Fudan University, China. Carbon and nitrogen isotope values are reported relative to Vienna Pee Dee Belemnite (VPDB) and AIR, respectively, and the analytical precision is lower than ±0.2‰ (1σ).

Nitrogen isotope ratios in individual amino acids

Amino acid nitrogen isotope analysis quantifies the nitrogen isotope ratio of individual amino acids from collagen, offering a more precise estimation of an individual’s trophic level (Chikaraishi et al., 2009; Chikaraishi, 2021). Calculating trophic levels directly from collagen δ¹⁵N, especially in archaeological samples, is challenging due to the difficulty in determining the isotope composition of primary producers. The method focuses on source amino acids (phenylalanine, Phe), which exhibit minimal δ¹⁵N enrichment between prey and predators (approximately 0.4‰), and trophic amino acids (glutamic acid, Glu), which become enriched by 3–8‰ with each trophic level increase (Chikaraishi et al., 2010; Ohkouchi et al., 2017). This method was originally used to identify the dietary intake and trophic level of animals within an ecosystem (McClelland and Montoya, 2002; Chikaraishi et al., 2007, 2009; Potapov et al., 2019) and is now applied in archaeological research to evaluate the relative contribution of difference ecosystems (Naito et al., 2010, 2016; Itahashi et al., 2017;). The trophic position (TP) for each sample was estimated by the following equations (Chikaraishi et al., 2010):

  

$$\begin{gathered} {\text{TP}}_{\mathit{\text{aqua}}}=\left[\left({\delta }^{15}{N}_{\mathit{\text{Glu}}}-{\delta }^{15}{N}_{\mathit{\text{Phe}}}-3.4\right)/7.6\right]+1 \\ \left(\text{for aquatic food webs}\right) \end{gathered}$$

  

$$\begin{gathered} {\text{TP}}_{\mathit{\text{ter}}}=\left[\left({\delta }^{15}{N}_{\mathit{\text{Glu}}}-{\delta }^{15}{N}_{\mathit{\text{Phe}}}+8.4\right)/7.6\right]+1 \\ \left(\text{for terrestrial }{\text{C}}_{\text{3}}\text{ food webs}\right) \end{gathered}$$

  

$$\begin{gathered} {\text{TP}}_{C4}=\left[\left({\delta }^{15}{N}_{\mathit{\text{Glu}}}-{\delta }^{15}{N}_{\mathit{\text{Phe}}}-0.4\right)/7.6\right]+1 \\ \left(\text{for terrestrial }{\text{C}}_{\text{4}}\text{ food webs}\right) \end{gathered}$$

A previous study by Itahashi et al. (2020) observed that the Δ¹⁵N_Glu-Phe values of C₄ plant seeds (–5.8 ± 1.6‰) were lower than the theoretical Δ¹⁵N_Glu-Phe values of C₄ plants measured by leaves (0.4 ± 1.7‰). Therefore, we applied the mean Δ¹⁵N_Glu-Phe value of foxtail millet (–7.3 ± 0.8‰) reported by Itahashi et al. (2020) as the baseline for the primary producers in the diet of millet farmers and applied the new TP_{C4_modified} equation to estimate the TP of humans and animals within the C₄ food web:

  

$$\begin{gathered} {\text{TP}}_{C4\_\mathit{\text{modified}}}=\left[\left({\delta }^{15}{N}_{\mathit{\text{Glu}}}-{\delta }^{15}{N}_{\mathit{\text{Phe}}}+7.3\right)/7.6\right]+1 \\ \left(\text{for terrestrial }{\text{C}}_{\text{4}}\text{ food webs}\right) \end{gathered}$$

Collagen samples utilized for carbon and nitrogen isotope analysis were used to extract individual amino acids. The pretreatment methods are described in Chikaraishi et al. (2007). Approximately 2 mg of the collagen samples were hydrolyzed in 12 M HCl at 110°C for 12 hours. The hydrolyzed samples underwent derivatization using thionyl chloride/2-propanol (1:4, v/v) at 110°C for 2 hours followed by pivaloyl chloride/dichloromethane (1:4, v/v) at 110°C for 2 hours. After derivatization, the amino acid derivatives were extracted via liquid–liquid extraction with n-hexane/dichloromethane (3:2, v/v) and distilled water.

The amino acid samples were introduced into a gas chromatograph-hydrogen flame ionization detector (GC-FID, Thermo Scientific TRACE 1310) to determine the amounts of glutamic acid and phenylalanine present. Dichloromethane/pentane (9:1, v/v) solution was used to adjust amino acid samples to a concentration of approximately 300 μg phenylalanine per sample. The δ¹⁵N values of the amino acids were determined using a gas chromatography-combustion-isotope ratio mass spectrometer (GC-C-IRMS, Agilent 6890GC coupled to Thermo Finnigan Delta plus XP IRMS via a GC-C/TC III interface to a combustion furnace at 950°C and a reduction furnace at 550°C) at UMUT.

Modern horse and cow collagen were employed as internal standards for quality control of the pretreatment process and pretreated alongside unknown samples. The measurement uncertainty was monitored using standard mixtures of amino acids (SI Science Co., Ltd) inserted after every five unknown samples in analytical runs. The δ¹⁵N values of amino acids values were calibrated with multipoint calibration using values of laboratory standards (l-alanine, glycine, l-methionine, l-aspartic acid, and l-hydroxyproline). The analytical precision was ±0.5‰ (1σ) for more than 2 nmol N of amino acids. The errors (1σ) were 0.2 for the estimated TP_ter (Chikaraishi et al., 2010) and 0.1 for TP_aqua (Chikaraishi et al., 2009).

Oxygen isotope ratios in bone/enamel

Hydroxyapatite, the main inorganic component of bones and teeth, is a compound of calcium phosphate and carbonate (Dorozhkin, 2009). The oxygen isotope composition in mammalian skeletal tissue is determined by the oxygen isotope composition of water in vivo, which primarily originates from drinking water and food (Kohn, 1996; Daux et al., 2008; Podlesak et al., 2008). Thus, the oxygen isotope ratios in hydroxyapatite during its formation can reflect those of consumed water (δ¹⁸O_w) (Longinelli, 1984). The primary source of drinking water is mainly from rainwater. In turn, the oxygen isotopes of rainwater are influenced by latitude, elevation, and distance from the coast. Proximity to the ocean and higher temperatures generally result in higher oxygen isotope ratios (Dansgaard, 1964).

Additionally, oxygen isotope ratios in hydroxyapatite can analyze oxygen atoms in both phosphate and carbonate groups. The ratio of carbonate is more widely applied than phosphate due to simpler measurement conditions and the ability to measure carbon isotope ratios simultaneously (Lanehart et al., 2011). However, due to hydroxyapatite’s crystal structure, phosphate groups are highly stable and less susceptible to diagenesis than carbonate (Zazzo et al., 2004; Passey et al., 2007). This study measured oxygen isotope ratios in phosphate (δ¹⁸O_p) using silver phosphate (Bryant et al., 1996; Vennemann et al., 2002). The δ¹⁸O_w values of humans and mammals were calculated with the formula proposed by Daux et al. (2008) and Kohn and Cerling (2002) from δ¹⁸O_p, respectively.

After cleaning the surface by removing a thin layer of bone/tooth surface with a tungsten carbide drill, approximately 30 mg of bone/tooth enamel powder was collected. To remove organic matter, 1 ml of 4% (v/v) sodium hypochlorite solution was added overnight at room temperature. Next, 1 ml of 0.1 M acetate buffer (pH 4.4) was added to remove secondary carbonate deposition for 4 hours. After being washed with Milli-Q water, the samples were dried overnight at 60°C. Approximately 10 mg of the prewashed sample was weighed and purified with phosphoric acid for measurement of the oxygen isotope ratio. To remove calcium, which inhibits the precipitation of silver phosphate, 0.2 ml of 68% ultrapure nitric acid (TAMAPURE-AA-100, Tama Chemical Industry Co., Ltd) and 1.8 ml of 3 M hydrofluoric acid were added and allowed to react overnight at room temperature. Calcium was precipitated as CaF₂ and centrifuged to remove this from the supernatant liquid. Subsequently, 2 ml of 25% ammonia and 1 M silver nitrate solution were added to react for 20 hours at 70°C to produce silver phosphate. Once silver phosphate crystals were precipitated, the supernatant was discarded after centrifugation. The crystals washed with Milli-Q water were dried overnight in a 60°C oven and were finely powdered in a mortar. Consequently, 1.2 mg of the sample was weighed and wrapped in silver foil for analysis. Measurements were performed using a pyrolysis elemental analyzer and stable isotope ratio mass spectrometer (TC/EA-IRMS, Thermo Fisher Scientific, Finnigan TC/EA, and DELTA V Plus) at UMUT. Samples were calibrated to δ¹⁸O values relative to V-SMOW using internal standards, including NBS127 (9.4‰), Merck Cellulose (28.7‰), and NBS120c (21.7‰). The uncertainties of the measurements were less than 0.5‰ (1σ).

Strontium isotope ratios in bone/enamel

Studies utilize strontium isotope ratios as an indicator to incorporate geographic information into the analysis of animals inhabiting a given area and to reconstruct their migrations (Evans et al., 2006; Kusaka et al., 2011; Laffoon et al., 2017; Wang et al., 2020). Strontium consists of four isotopes (⁸⁴Sr, ⁸⁶Sr, ⁸⁷Sr, ⁸⁸Sr), with ⁸⁷Sr being produced through the beta decay of the radioactive ⁸⁷Rb (Steiger and Jäger, 1977). Hence, strontium isotope ratios (⁸⁷Sr/⁸⁶Sr) vary with geologic age, rubidium and strontium contents, and closure temperature within the rock. The concentration and isotope composition of strontium also differ in soils and are initially absorbed by ecosystems through plants (Bentley, 2006). It is fixed in animals by replacing calcium after the ingestion of strontium through food. (Faure and Mensing, 2005). Due to the similar mass numbers and abundance ratios of ⁸⁶Sr and ⁸⁷Sr, strontium isotope ratios are minimally influenced by dynamic isotope effects in chemical reactions, resulting in negligible in vivo isotope fractionation (Bentley, 2006). Therefore, understanding the variation and geographical distribution of strontium isotope ratios across different regions is crucial for reconstructing migration history. Obtaining local strontium isotope values of plants and animals inhabiting a research area is essential (Durrant and Ward, 2005). An increasing number of studies contribute to the discussion on the diversity of strontium available to native organisms (Bataille et al., 2012; Ryan et al., 2018; Wang and Tang, 2020). In this study, we analyzed a series of animal samples to estimate the local range of strontium isotope ratios.

About 3–5 mg of the prewashed sample was weighed, and 0.5 ml of 3.5 M HNO₃ (Tama Chemicals Co., Ltd, TAMAPURE AA-100) was added to dissolve the sample completely at room temperature. After centrifugation, the supernatant solution underwent purification to remove heavy elements other than strontium, according to Kusaka et al. (2022), with slight modifications, using a handmade column (Penn & Nitto Co., PENTUBE TFE4X, 3/8EX, with PTEE filter, 100 μm) packed with strontium-specific resin (Eichrom Technologies Inc., 100–150 μm particle size). These pretreatments were performed on a clean draft at UMUT, to avoid contamination. Column-purified sample solutions were analyzed by inductively coupled plasma-mass spectrometry (ICP-MS, Agilent 7500a, Agilent Technologies Co.) to determine the strontium concentration at the Micro Analysis Laboratory, Tandem accelerator (MALT), The University of Tokyo. The ⁸⁷Sr/⁸⁶Sr ratios were analyzed by multi-collector ICP mass spectrometer (MC-ICP-MS, NEPTUNE plus, Thermo Fisher Scientific) at the Laboratory of Advanced Instrumentation and Chemistry, The University of Tokyo. To calibrate the ⁸⁷Sr/⁸⁶Sr values, we applied a calculation method advanced based on the exponential law (Russell et al., 1978), as described by Ohno and Hirata (2007), to eliminate as much as possible any isobaric interference by ⁸⁷Rb. A correction factor, k, was applied according to the following equation:

  

$\begin{array}{ll}{{}^{87}Rb}_{\mathit{\text{corr}}}=& {{}^{85}Rb}_{\mathit{\text{meas}}}\times {\left(\frac{{}_{ }{}^{87}Rb}{{}_{ }{}^{85}Rb}\right)}_{\mathit{\text{ref}}}\\ & \times {\left[{\left(\frac{{}_{ }{}^{88}Rb}{{}_{ }{}^{86}Rb}\right)}_{meas}/{\left(\frac{{}_{ }{}^{88}Sr}{{}_{ }{}^{86}Sr}\right)}_{ref}\right]}^{ln\left(m87/m85\right)/ln\left(m88/m86\right)\times k}\end{array}$

In this study, a correction factor (k = 1.030) was applied to correct mass bias between rubidium and strontium. The ⁸⁸Sr/⁸⁶Sr value, obtained from the measurements was normalized to 8.37861. The long-term reproducibility of the NIST SRM 987 standard conducted throughout this study is 0.71037 ± 0.0003 (2σ; n = 30), well within the certified range of 0.71034 ± 0.00026 (Moore et al., 1982). The strontium isotope ratio of samples was normalized to the recommended values (⁸⁷Sr/⁸⁶Sr = 0.71025) of NIST SRM 987 (Faure and Mensing, 2005). The long-term reproducibility of the ⁸⁷Sr/⁸⁶Sr measurement for NIST SRM 1400 (bone ash) is 0.71312 ± 0.0003 (2σ; n = 30).

Statistical analyses

Given the sample size limitations, the non-parametric Wilcoxon rank-sum test was employed to assess significant differences between the two datasets. Spearman’s rank correlation coefficient was used to evaluate linearity. P-values less than 0.05 were deemed statistically significant. All analyses were conducted using R statistical software v. 4.3.3 (R Core Team, 2021).

Results

Carbon and nitrogen isotope ratios in collagen

The results of the carbon and nitrogen isotope data for the Haojiatai site are presented in Table 2 and Figure 2. Among 48 samples (19 human and 29 animal bones), 47 samples (excluding one fish sample) exhibited C:N ratios ranging from 2.9 to 3.6 for the gelatin components (DeNiro, 1985). The yields of extracted gelatin were over 1%, demonstrating excellent preservation of collagen (Van Klinken, 1999).

Table 2.

The results of bulk carbon and nitrogen isotopes of collagen, and the nitrogen isotope results of individual amino acids

Sample ID	Species	Group	Yield of bone collagen (%)	C%	N%	C/N	δ13C (‰VPDB)	δ15N (‰AIR)	δ15NPro	δ15NAsp + Thr	δ15NSer	δ15NGlu	δ15NPhe	δ15NHyp	TPaqua	TPter	TPC4	TPC4_modified	Δ15NGlu-Phe
M101	Human	C4	9.3	45.2	16.4	3.2	–8.5	9.1	13.8	7.2	7.5	13.7	6.7	14.4	1.5	3.0	1.9	2.9	7.1
M102	Human	C4	1.0	9.2	3.1	3.5	–9.0	9.9	15.9	7.5	8.2	13.5	8.4	14.7	1.2	2.8	1.6	2.6	5.2
M103	Human	C4	1.0	19.3	6.7	3.4	–8.9	8.7	14.7	11.0	5.9	13.2	8.0	—	1.2	2.8	1.6	2.6	5.1
M104	Human	C4	5.0	44.3	16.3	3.2	–8.7	9.2	16.1	7.1	7.6	13.2	7.5	13.5	1.3	2.9	1.7	2.7	5.7
M105	Human	C4	5.9	41.9	15.2	3.2	–11.0	8.8	—	8.8	9.2	13.4	10.0	14.7	1.0	2.6	1.4	2.4	3.5
M106	Human	C4	13.1	44.0	16.0	3.2	–8.4	6.6	15.5	5.4	4.0	15.5	4.6	11.4	2.0	3.5	2.4	3.4	11.0
M107	Human	C4	11.6	45.7	16.6	3.2	–9.5	7.6	15.5	8.1	8.4	13.4	7.9	14.7	1.3	2.8	1.7	2.7	5.5
M108	Human	C3	7.9	41.8	15.2	3.2	–18.3	11.0	20.4	12.1	12.6	16.9	13.1	21.2	1.1	2.6	1.5	2.5	3.9
M109	Human	C4	7.7	43.5	16.0	3.2	–8.6	8.3	14.1	10.6	7.8	13.7	8.3	14.8	1.3	2.8	1.7	2.7	5.4
M112	Human	C3	10.6	43.6	15.8	3.2	–16.5	10.0	18.1	11.6	9.0	15.1	11.7	17.8	1.0	2.6	1.4	2.4	3.4
M113	Human	C4	3.4	42.4	15.1	3.3	–9.6	9.0	16	8.1	8.5	14.4	9.8	15.2	1.2	2.7	1.5	2.6	4.6
M114	Human	C4	14.1	42.9	15.9	3.2	–8.4	8.4	14.1	7.9	8.2	12.9	9.6	14.3	1.0	2.5	1.4	2.4	3.2
M116	Human	C3	7.2	42.2	15.2	3.2	–20.1	9.8	15.5	11.2	9.6	15.1	9.7	17.3	1.3	2.8	1.7	2.7	5.4
M117	Human	C4	6.3	42.5	15.6	3.2	–9.2	8.8	14.9	9.5	9.1	13.2	9.2	15.0	1.1	2.6	1.5	2.5	4.0
M118	Human	C3	2.2	42.5	15.2	3.3	–20.2	9.9	16.7	10.7	9.7	15.9	9.3	17.4	1.4	3.0	1.8	2.8	6.6
M119	Human	C4	8.4	42.4	15.3	3.2	–9.0	8.9	13.8	8.2	9.3	14.3	9.2	15.8	1.2	2.8	1.6	2.6	5.1
M120	Human	C4	8.8	44.9	16.4	3.2	–10.0	7.7	11.8	9.8	7.0	13.0	8.5	13.4	1.1	2.7	1.5	2.5	4.4
H554	Human	C3	5.0	43.5	15.8	3.2	–18.8	9.3	15.5	9.0	8.0	14.2	9.8	16.2	1.1	2.7	1.5	2.5	4.4
H608	Human	C4	14.5	46.4	16.8	3.2	–9.1	7.2	15.2	8.6	7.7	12.5	8.9	13.6	1.0	2.6	1.4	2.4	3.6
TG1_9	Pig	Domestic animal	4.2	45.3	16.3	3.2	–9.7	6.6	12.7	9.6	5.9	11.7	9.5	11.4	0.8	2.4	1.2	2.2	2.2
H671_1	Pig	Domestic animal	2.2	42.7	15.3	3.3	–9.6	7.9	14.2	8.6	8.6	13.2	10.3	12.9	0.9	2.5	1.3	2.3	2.8
T8339_11	Pig	Domestic animal	9.1	44.5	15.9	3.3	–10.2	8.1	13.2	9.7	8.0	13.6	8.7	12.7	1.2	2.8	1.6	2.6	4.9
T8339_1	Pig	Domestic animal	1.7	36.4	12.9	3.3	–7.2	6.7	13.1	8.7	6.9	11.8	9.6	11.7	0.8	2.4	1.2	2.2	2.2
H506_1	Cattle	Domestic animal	9.1	45.3	16.4	3.2	–9.6	5.3	11.3	8.0	4.4	9.8	10.6	9.9	0.4	2.0	0.8	1.8	–0.9
H609_1	Cattle	Domestic animal	2.8	43.2	15.7	3.2	–10.6	7.2	13.4	8.9	6.3	12.0	11.4	11.5	0.6	2.2	1.0	2.0	0.6
H619_1	Cattle	Domestic animal	2.1	45.0	16.3	3.2	–9.2	6.3	12.1	7.0	6.1	11.2	10.6	11.4	0.6	2.2	1.0	2.0	0.6
H527_1	Cattle	Domestic animal	5.6	42.0	15.2	3.2	–9.1	6.7	11.0	8.3	6.0	10.9	11.4	10.0	0.5	2.0	0.9	1.9	–0.5
H501_1	Dog	Domestic animal	13.3	41.1	14.9	3.2	–10.8	6.8	13.1	9.5	8.4	15.2	8.3	—	1.5	3.0	0.8	1.8	6.8
H502_2	Dog	Domestic animal	2.7	43.2	15.5	3.3	–9.7	6.7	11.6	9.5	9.3	14.6	10.2	—	1.1	2.7	0.9	1.9	4.4
H527_1	Dog	Domestic animal	11.6	44.1	15.8	3.3	–9.2	7.5	14.0	8.5	8.0	13.9	9.3	13.8	1.2	2.7	0.7	1.8	4.6
H627_1	Dog	Domestic animal	5.5	44.7	15.9	3.3	–21.7	3.6	6.4	3.8	3.5	7.8	10.4	8.5	0.2	1.8	0.7	1.7	–2.6
H556_1	Pere David’s deer	Wild herbivore	9.0	42.8	15.3	3.3	–8.5	5.2	6.5	4.7	3.7	9.3	10.3	7.5	0.4	2.0	0.6	1.6	–1.0
H521_1	Pere David’s deer	Wild herbivore	8.9	45.2	16.3	3.2	–15.7	6.4	8.8	5.8	7.8	10.8	11.5	10.2	0.5	2.0	0.6	1.6	–0.7
TG1_3	Pere David’s deer	Wild herbivore	5.9	42.7	15.4	3.2	–13.6	5.2	8.0	4.6	9.4	9.8	11.3	8.8	0.4	1.9	0.7	1.7	–1.5
H619_1	Sika deer	Wild herbivore	3.6	43.5	15.7	3.2	–20.2	3.7	8.4	3.7	1.8	8.4	10.8	—	0.2	1.8	0.6	1.6	–2.4
T8339_11	Sika deer	Wild herbivore	8.8	45.5	16.5	3.2	–16.3	5.7	11.3	4.9	10.6	10.4	12.0	10.6	0.3	1.9	0.6	1.6	–1.6
H538_1	Sika deer	Wild herbivore	5.8	43.9	15.8	3.2	–17.9	4.8	9.7	4.6	4.5	9.6	12.1	9.1	0.2	1.8	0.6	1.6	–2.5
H622_1	River deer	Wild herbivore	4.4	44.0	16.0	3.2	–20.9	4.2	8.9	4.4	4.5	9.0	11.2	9.6	0.3	1.8	1.8	2.9	–2.2
H523_1	River deer	Wild herbivore	6.9	40.5	14.3	3.3	–22.4	3.0	11.5	6.7	6.4	10.4	13.1	9.7	0.2	1.8	1.5	2.5	–2.7
H671_2	River deer	Wild herbivore	1.6	42.2	15.3	3.2	–11.5	5.4	9.3	5.8	5.4	10.4	12.9	9.7	0.2	1.8	1.6	2.6	–2.6
H714_1	Snakehead fish	Aquatic animal	1.8	39.9	14.4	3.2	–23.6	6.9	—	—	—	—	—	—	—	—	3.1	4.1	—
H540_1_1	Snakehead fish	Aquatic animal	2.1	41.3	14.7	3.3	–21.9	7.9	14.5	6.5	6.3	18.9	2.4	15.1	2.7	4.3	2.8	3.8	16.5
*H671_2	Catfish	Aquatic animal	0.1	2.7	1.1	3.0	–16.6	—	—	—	—	—	—	—	—	—	—	—	—
H540_1_2	Yellow catfish	Aquatic animal	3.2	42.1	15.2	3.2	–24.6	7.2	14.7	5.5	9.1	17.8	3.6	13.9	2.4	4.0	1.6	2.6	14.2
H538_1	Tortoise	Aquatic animal	1.0	40.5	14.5	3.3	–21.4	6.6	9.2	4.5	5.7	11.9	4.1	9.8	1.6	3.1	1.8	2.8	7.8
F58_1	Tortoise	Aquatic animal	7.4	44.8	16.0	3.3	–20.5	7.3	10.4	5.3	5.5	13.6	8.9	11.5	1.2	2.7	1.8	2.8	4.7
H619_1	Soft-shelled turtle	Aquatic animal	4.0	44	15.7	3.3	–24.4	8.2	11.7	6.4	6.0	12.8	6.1	11.9	1.4	3.0	2.2	3.2	6.7
T8339_12	Soft-shelled turtle	Aquatic animal	1.8	41.8	15.1	3.2	–24.5	8.0	12.0	5.6	6.1	13.0	6.6	—	1.4	2.9	2.1	3.1	6.4
T8339_6b	Shell	Aquatic animal	—	—	—	—	—	—	11.4	9.0	—	12.1	2.9	—	1.8	3.3	1.9	2.9	9.2
H564_2	Shell	Aquatic animal	—	—	—	—	—	—	12.5	16.4	9.6	13.5	5.0	—	1.7	3.2	1.6	2.6	8.6
211206	Rice	Plant	—	55.4	2.9	21.9	–23.0	4.2	—	—	—	—	—	—	—	—	1.6	2.6	—
211207	Rice	Plant	—	44.2	2.3	22.7	–26.0	3.5	—	—	—	—	—	—	—	—	1.7	2.7	—
211208	Rice	Plant	—	44.1	3.2	16.2	–26.0	6.8	—	—	—	—	—	—	—	—	1.4	2.4	—
211210	Rice	Plant	—	46.7	2.2	24.9	–24.4	6.1	—	—	—	—	—	—	—	—	2.4	3.4	—

* Samples marked as not having good-quality collagen were excluded from the discussion.

Figure 2.

Carbon and nitrogen isotope results of samples measured in this study.

A previously published study reported that the δ¹³C and δ¹⁵N values of 13 humans from Haojiatai could divided into two groups, with mean δ¹³C values of –19.1 ± 0.59‰ and –9.6 ± 0.9‰ (Zhou, 2017; Li, 2022). In this study, all measurements were well reproduced, and humans were also categorized into two groups based on their δ¹³C values, with means of –18.8 ± 1.5‰ and –9.1 ± 0.7‰, respectively. Significant differences in δ¹³C were observed between the C₃- and C₄-diet groups (Wilcoxon rank-sum test, P = 0.0014), suggesting distinct dietary patterns between the groups consuming C₃ and C₄ foods at this site. The statistical difference in δ¹⁵N was also significant between the C₃- and C₄-diet groups (Wilcoxon rank-sum test, P = 0.003). The C₃-diet group exhibited higher δ¹⁵N values than the C₄-diet group, with mean δ¹⁵N values of 10.0 ± 0.6‰ and 8.4 ± 0.9‰, respectively. This result is consistent with previously reported results.

The animal samples from Haojiatai comprised terrestrial domestic animals, terrestrial wild herbivores, and freshwater animals. Four domestic pigs exhibited high δ¹³C and δ¹⁵N values of –9.2 ± 1.3‰ and 7.3 ± 0.8‰, respectively. These values indicate a diet primarily consisting of C₄ plants, aligning with the expectation that humans raised these animal as livestock which shared similar dietary patterns with humans. Cattle and dogs also demonstrated comparable δ¹³C and δ¹⁵N values to those of pigs and C₄-diet humans. An exception was observed in one dog, which exhibited low δ¹³C (–21.7‰) and δ¹⁵N (3.6‰) values. In terrestrial wild animals, the δ¹³C values exhibit a wide distribution. Three large deer displayed average δ¹³C and δ¹⁵N values of –12.6 ± 3.7‰ and 5.6 ± 0.7‰, respectively. A medium-sized sika deer (n = 3) demonstrated average δ¹³C and δ¹⁵N values of –18.1 ± 2.0‰ and 4.7 ± 1.0‰, respectively. Three small-sized river deer showed average δ¹³C and δ¹⁵N values of –18.3 ± 5.9‰ and 8.1 ± 1.0‰, respectively. This variation in isotope values reflects the diverse natural environment surrounding the Haojiatai site, including C₄ plants in wetlands or floodplains. Fish (n = 3), comprising two snakehead fish and one yellow catfish, displayed average δ¹³C and δ¹⁵N values of –23.4 ± 1.4‰ and 7.3 ± 0.5‰, respectively. Two tortoises exhibited average δ¹³C and δ¹⁵N values of –20.9 ± 0.7‰ and 7.0 ± 0.5‰, respectively. Two soft-shelled turtles displayed average δ¹³C and δ¹⁵N values of –24.4 ± 0.07‰ and 8.1 ± 0.1‰, indicating a range of freshwater ecosystems with higher δ¹⁵N values than terrestrial systems.

The rice samples (n = 4) showed average δ¹³C values of –24.9 ± 1.4‰ and higher δ¹⁵N values of 5.2 ± 1.6‰, relative to wild deer consuming wild C₃ plants. Their higher δ¹⁵N values were probably caused by the anaerobic soil conditions at irrigated paddy fields and the physiological processes in rice with their preference for ammonia as a primary nitrogen source (Yoneda et al., 2019). There is also a strong possibility that this finding was influenced by fertilizer activities, given the higher δ¹⁵N value of the millets excavated from Haojiatai (Lin, 2022), and fertilization was found at the Wadian and Wangchenggang sites of the same period in Henan (Wang, 2018).

Nitrogen isotope ratios of amino acids

The δ¹⁵N values of amino acids in both humans and animals are presented in Table 2 and Figure 3. Despite the influence of C₄ plants on wild deer at Haojiatai, the trophic position calculated from the TP_ter equation ranged from 1.8 to 2.0 (n = 9, including Pere David’s deer, sika deer, and river deer). This suggests that these animals were terrestrial herbivores that predominantly consumed plants. The trophic positions of domestic animals, which clearly relied on a C₄ diet, were calculated using the TP_{C4_modified} equation. Those ranged from 2.0 ± 0.1 for herbivore cattle (n = 4) to 2.4 ± 0.2 for omnivore pigs (n = 4). These values are nearly consistent with those obtained using the TP_ter equation, suggesting that the TP_ter equation is equally applicable and effective within the C₄ food web. Two snakehead fish and two shellfish exhibited reasonable values of TP_aqua. While we initially assumed soft-shell turtles and tortoises to be aquatic carnivores, their TP_aqua values lower than 2.0 probably originated from intake from both aquatic and terrestrial ecosystems in their diet. Consequently, we have included only fish and shellfish as representative aquatic animals in the following discussion.

Figure 3.

Nitrogen isotope values of amino acids samples at the Haojiatai site. Horizontal solid lines represent expected Δ¹⁵N_Glu-Phe for TP_ter = 2 (red), TP_aqua = 2 (blue), TP_{C4_modified} = 2 (yellow). The dashed lines represent the expected Δ¹⁵N_Glu-Phe for TP_ter = 3 (red), TP_aqua = 3 (blue), TP_{C4_modified} = 3 (yellow). Light blue and pink lines represent the linearity of the C₄- and C₃-diet groups, respectively.

The average of δ¹⁵N_Glu and δ¹⁵N_Phe values in humans were 13.7 ± 1.5‰ and 9.0 ± 1.8‰, respectively. The C₃-diet group exhibited higher δ¹⁵N_Glu and δ¹⁵N_Phe values (15.4 ± 1.0‰ and 10.7 ± 1.6‰, respectively, n = 5) compared with those in the C₄-diet group (13.1 ± 1.2‰ and 8.3 ± 1.4‰, respectively, n = 14). Significant differences were observed between the C₃- and C₄-diet groups for both δ¹⁵N_Glu (Wilcoxon rank-sum test, P = 0.0062) and δ¹⁵N_Phe (Wilcoxon rank-sum test, P = 0.0095). The trophic position of the C₃-diet group calculated from the TP_ter equation ranged from 2.6 to 3.0, whereas TP_aqua ranged from 1.0 to 1.4. This suggests that the C₃-diet group preferred terrestrial sources for their diet. On the other hand, the trophic position of all individuals within the C₄-diet group, which was calculated from the TP_{C4_modified} equation, had a trophic position range from 2.4 to 2.9. A significant correlation was observed between δ¹⁵N_Phe and Δ¹⁵N_Glu-Phe values among the C₄-diet group (Spearman’s rank correlation ρ = –0.76, P = 0.0017). A negative correlation generally indicates the mixing of aquatic and terrestrial food sources. The tropic position of the C₄-diet group, however, as calculated from TP_aqua, ranged from 1.0 to 2.0, suggesting that the contribution of aquatic resources to their diet is minimal.

Oxygen isotope ratios in bone and enamel

Due to the poor preservation of five individuals from Haojiatai who were without teeth, in order to facilitate comparison of data for all, we obtained oxygen isotopes from bone for 19 individuals, and for 14 individuals also analyzed teeth. The results of 19 humans (n = 33) along with animal bones (n = 21) are presented in Table 3 and Figure 4. The oxygen isotope ratio of phosphate (δ¹⁸O_p) for human bone samples ranged from 14.3‰ to 17.0‰, with an average value of 15.6 ± 0.9‰ (1σ). No significant differences were observed between the C₃- and C₄-diet groups (Wilcoxon rank-sum test, P = 0.926). The δ¹⁸O_p value of human teeth ranged from 14.3‰ to 19.4‰, with an average value of 16.2 ± 1.4‰ (1σ).

Table 3.

Oxygen and strontium isotope results at the Haojiatai site

Sample ID	Species	Group	Tooth	δ18Op(‰VSMOW*)tooth	δ18Ow(‰)tooth	δ18Op(‰VSMOW)bone	δ18Ow(‰)bone	87Sr/86Sr	2σ
M101	Human	C4	—	—	—	16.8	–7.8	0.71307	0.00098
M102	Human	C4	URM1	15.6	–9.7	17.0	–7.5	0.71253	0.00019
M103	Human	C4	LLM1	19.4	–3.97	16.5	–8.2	0.71286	0.00019
M104	Human	C4	ULI1	15.8	–9.47	16.3	–8.6	0.71231	0.00090
M105	Human	C4	—	—	—	15.6	–9.7	0.71311	0.00081
M106	Human	C4	—	—	—	15.3	–10.1	0.71298	0.00091
M107	Human	C4	LLC	15.6	–9.87	16.3	–8.6	0.71260	0.00082
M108	Human	C3	—	—	—	14.6	–11.3	0.71298	0.00093
M109	Human	C4	ULC	17.3	–7.2	16.2	–8.8	0.71301	0.00019
M112	Human	C3	—	—	—	16.4	–8.4	0.71300	0.00088
M113	Human	C4	LLM3	16.1	–9.1	14.6	–11.3	0.71267	0.00018
M114	Human	C4	ULC	14.4	–11.6	14.5	–11.5	0.71169	0.00017
M116	Human	C3	LLM1	17.9	–6.2	15.9	–9.2	0.71240	0.00017
M117	Human	C4	URP2	15.9	–9.3	15.1	–10.5	0.71291	0.00017
M118	Human	C3	LLM1	17.7	–6.5	16.0	–9.0	0.71146	0.00018
M119	Human	C4	URI1	15.8	–9.4	14.3	–11.6	0.71263	0.00016
M120	Human	C4	URP2	14.7	–11.1	15.5	–9.8	0.71231	0.00014
H554	Human	C3	ULC	17.2	–7.3	15.4	–10.0	0.71272	0.00015
H608	Human	C4	URM3	14.6	–11.2	14.3	–11.8	0.71233	0.00019
TG1_9	Pig	Domestic animal	—	—	—	14.0	–10.0	0.71307	0.00014
H671_1	Pig	Domestic animal	—	—	—	13.3	–10.7	0.71295	0.00014
T8339_11	Pig	Domestic animal	—	—	—	13.2	–10.9	0.71298	0.00015
T8339_1	Pig	Domestic animal	—	—	—	13.9	–10.1	0.71301	0.00016
H506_1	Cattle	Domestic animal	—	—	—	17.0	–6.7	—	—
H609_1	Cattle	Domestic animal	—	—	—	15.2	–8.6	—	—
H619_1	Cattle	Domestic animal	—	—	—	15.7	–8.1	—	—
H527_1	Cattle	Domestic animal	—	—	—	16.4	–7.3	—	—
H501_1	Dog	Domestic animal	—	—	—	14.8	–9.1	0.71299	0.00014
H502_2	Dog	Domestic animal	—	—	—	15.4	–8.5	0.71305	0.00013
H527_1	Dog	Domestic animal	—	—	—	15.3	–8.5	0.71304	0.00016
H627_1	Dog	Domestic animal	—	—	—	15.7	–8.1	0.71290	0.00015
H556_1	Pere David’s deer	Wild herbivore	—	—	—	16.7	–6.9	—	—
H521_1	Pere David’s deer	Wild herbivore	—	—	—	16.8	–6.9	—	—
TG1_3	Pere David’s deer	Wild herbivore	—	—	—	15.2	–8.6	—	—
H619_1	Sika deer	Wild herbivore	—	—	—	16.1	–7.6	—	—
T8339_11	Sika deer	Wild herbivore	—	—	—	15.3	–8.5	—	—
H538_1	Sika deer	Wild herbivore	—	—	—	15.2	–8.6	—	—
H622_1	River deer	Wild herbivore	—	—	—	18.4	–5.1	—	—
H523_1	River deer	Wild herbivore	—	—	—	15.3	–8.5	—	—
H671_2	River deer	Wild herbivore	—	—	—	17.8	–5.8	—	—

The δ¹⁸O_w was calculated from the equation for humans as δ¹⁸O_w = 1.54 × δ¹⁸O_p – 33.72 (Daux et al., 2008), and for other mammals as δ¹⁸O_p = 0.9 × δ¹⁸O_w – 23 (Kohn and Cerling, 2002).

* VSMOW, Vienna Standard Mean Ocean Water.

Figure 4.

A boxplot of the oxygen isotope ratios of drinking water calculated from the phosphoric oxygen isotope ratios at the Haojiatai site. The black dashed line represents the overall mean of the samples; the red dashed line indicates the modern surface water average.

The δ¹⁸O_p values of the 21 animals ranged widely from 13.2‰ to 18.4‰, with an average value of 15.6 ± 0.9‰ (1σ). However, it could be noticed that values vary considerably between species, especially with a variation of around 3.5‰ for river deer compared to 0.9‰ for pigs, reflecting differences in body water intake between species. Various factors affect the δ¹⁸O value in the body, including drinking behavior, diet, and physiology (Kohn, 1996; Kohn et al., 1996). Large mammals primarily have two primary sources of body water: meteoric water and food (Longinelli, 1984; Luz et al., 1984). It has been shown that the δ¹⁸O values of tooth enamel in obligate drinkers tend to be lower than those in non-obligate drinkers (Moritz et al., 2012; Reid et al., 2019). Due to the loss of the lighter isotope water molecule (H₂O¹⁶) during evapotranspiration, leaf water is more enriched in ¹⁸O relative to meteoric water (Dongmann et al., 1974). Therefore, herbivores that preferentially obtain body water from plants will have higher δ¹⁸O_w values. In contrast, domestic pigs mainly obtain oxygen from drinking water, which is more homogeneous.

To identify outliers using estimated δ¹⁸O_w values for humans, it is imperative to compare them with the oxygen isotope ratio in surface water in the area around the archaeological site. The δ¹⁸O value of surface water in Luohe City, Henan Province (33° 30' 00" N, 114° 12' 36" E) was acquired from the Waterisotope Database (2017), and ranged from –10.4‰ to –8.9‰, with an average of –10.0 ± 0.5‰. Additionally, to investigate the seasonal variation in δ¹⁸O_w, δ¹⁸O of precipitation at Zhengzhou City, Henan Province (34° 43' 11.99" N, 113° 39' 00" E) was obtained from the International Atomic Energy Agency (IAEA) WISER Database (IAEA, 2022). The annual mean δ¹⁸O of precipitation was –6.5 ± 1.0‰. The monthly variation in the δ¹⁸O ranged from –9.3‰ to –3.2‰. Oxygen isotope values can exhibit significant variation across different seasons within the same region.

The δ¹⁸O_w of humans derived from δ¹⁸O_p ranged from –11.8‰ to –7.5‰, with an average of –9.7 ± 1.4‰ (Table 3, Figure 4), similar to the value of surface water in Luohe City. In Figure 4, the variability of human δ¹⁸O_w is considered smaller than the seasonal fluctuations in surface water. The δ¹⁸O_w value is not only affected by environments, but physiological processes could also be one of the factors (Pederzani and Britton, 2019).

The isotope signatures in teeth are fixed after crystallization, while teeth formed during infancy are affected by breastfeeding and would have higher δ¹⁸O values (Britton et al., 2015). Comparing the individuals with both tooth and bone δ¹⁸O_w values shows that the offset between bone and enamel becomes larger in individuals (M103, M116, M118) analyzed using the first molar (M1), ranging from 2.5‰ to 4.3‰. M1 begins to calcify after birth, and therefore is mostly affected and elevated by breast milk, making δ¹⁸O enrichment around 0.5–1.2‰ (Wright and Schwarcz, 1998). Therefore, the oxygen isotope results imply that individuals with significant long-distance mobility were not detectable among the Haojiatai human population.

Strontium isotope ratios in enamel

Strontium isotope ratio analysis was conducted on 19 humans (14 teeth and 5 bones) and 8 domesticated animal bones (4 pigs and 4 dogs) from the Haojiatai site, as shown in Table 3 and Figure 5. The ⁸⁷Sr/⁸⁶Sr values for human teeth ranged from 0.71146 to 0.71301, exhibiting a greater range of variability, with an average value of 0.71246 ± 0.0004 (1σ). The bioavailable strontium isotope values reported by Wang and Tang (2020) in the central region of Henan—where Haojiatai is located—range from 0.7110 to 0.7125, compared to the higher values ranging from 0.7125 to 0.7150 in western Henan. Apparently, the values of the Haojiatai humans cover both of these ranges. In contrast, the animal and human bones displayed narrower variability in strontium isotope ratios of 0.71301 ± 0.00006 (1σ) compared with human teeth. A similarity between animal and human bones is evidence that they reflect the value at the site soil through diagenetic effects. Hence, in this study, we utilized the average and 2σ range of ⁸⁷Sr/⁸⁶Sr in bone samples, 0.71301 ± 0.00012 (2σ), as the local baseline for comparing with ⁸⁷Sr/⁸⁶Sr values in human teeth (Figure 5).

Figure 5.

Strontium isotope ratios in animals and humans at the Haojiatai site. The solid dark gray lines and the light gray banded ranges represent the ‘local’ isotopic ranges, defined as means ± 2σ, derived from the ⁸⁷Sr/⁸⁶Sr values of both human and animal bone samples.

Non-local individuals were identified within both the C₃- and C₄-diet groups, irrespective of dietary distinctions. Among the C3-diet group, two females and one male were designated as outliers. Notably, one adult female labeled as M118 exhibited an exceptionally low ⁸⁷Sr/⁸⁶Sr value of 0.71146. In the C₄-diet group, eight individuals—three females, four males, and one of unspecified sex—displayed values exceeding the local range. The individual labeled M114 (age 15 ± 2 years, sex unknown) also demonstrated a notably low value of 0.71168. No discernible sexual bias in migration patterns was observed in humans. Considering the local baseline in this study was established using livestock animals with limited migratory ranges, and the significantly broader migratory pattern observed in humans, it is crucial to assess strontium isotope ratios in wild animals to enhance our understanding of human migration behaviors.

Discussion

Detailed reconstruction of diet

The carbon and nitrogen isotope analyses of bone collagen conducted in this study delineated two distinct human dietary traditions: the C₃- and C₄-diet groups. Domestic pigs and cattle predominantly consumed C₄ foods, whereas wild deer exhibited a broader range of δ¹³C values, suggesting the ingestion of both C₃ and C₄ plants (Figure 6a, b). These results are consistent with previous research (Zhou, 2017; Li et al., 2021; Li, 2022). The Central Plains region was characterized by millet agriculture from the Yangshao period to the Longshan period, with a notable prevalence of foxtail millet (Deng and Qin, 2017). Nevertheless, sites such as Wadian exhibited a high proportion of rice remains (Liu and Fang, 2010; Liu et al., 2018;), and carbon and nitrogen isotope analyses on human bones indicated the coexistence of C₃ and C₄ signals, similar to the findings at Haojiatai (Chen et al., 2016; Li et al., 2021). This suggests the diffusion of rice farming to the Central Plains during this period, with rice farmers contributing to the establishment of mixed agriculture. Previous studies proposed that the C₃-diet group at Haojiatai comprised hunter-gatherers rather than rice farmers (Li et al., 2021).

Figure 6.

(a) Comparison of δ¹³C and δ¹⁵N values in humans between the Haojiatai and other related sites. Data from published studies include 13 Haojiatai humans from Zhou (2017) and Li (2022), 22 Jiahu humans from Hu et al. (2006) and Wu et al. (2015), 20 Wadian samples from Chen et al. (2016), and 4 Meishan humans from Zhou (2017). (b) Summary of the published δ¹³C and δ¹⁵N values of Haojiatai, including 11 samples from Zhou (2017), 38 samples from Li et al. (2021), 13 samples from Li (2022), 5 millet samples from Lin (2022), and 51 samples from this study.

In the preceding investigation, the higher δ¹⁵N values in the C₃-diet group compared to the C₄-diet group were attributed to increased consumption of aquatic animals, including freshwater fish, soft-shelled turtles, and tortoises. However, amino acid analysis revealed that the trophic positions of Haojiatai humans lay within terrestrial ecosystems. The δ¹⁵N_Phe values of Haojiatai humans exhibit a positive correlation with δ¹⁵N in collagen (Figure 7a). Based on the δ¹⁵N_Phe values, the C₃-diet group can be divided into two subgroups. One subgroup, including individuals M116, M118, and H554, had lower δ¹⁵N_Phe and higher Δ¹⁵N_Glu-Phe, indicating a diet influenced by wild herbivores or aquatic resources. The other, consisting of individual M108 and M112, showed higher δ¹⁵N values of bone collagen with higher δ¹⁵N_Phe values, while their trophic positions were comparable to or lower than others in the group. Therefore, we concluded that the higher δ¹⁵N values in M108 and M112 within the C₃-diet group were mainly due to the consumption of food resources with high δ¹⁵N_Phe values such as paddy rice (Itahashi et al., 2020), rather than as a result of the extensive consumption of aquatic animals. The presence of charred rice with higher δ¹⁵N values (ranging from 3.5‰ to 6.8‰) from Haojiatai further supports the consumption of rice by the C₃-diet group. Accounting for a 3.5‰ enrichment from the diet in consumers, the δ¹⁵N values in bone collagen would span from 7.0‰ to 10.3‰. Even without fish consumption, if individuals consumed animals raised with rice, the δ¹⁵N values in human bones are anticipated to fall within the range of 10.5–13.7‰.

Figure 7.

(a) Comparison of δ¹⁵N in bone collagen and δ¹⁵N_Phe of humans. The solid black line indicates the regression line for humans. (b) Comparison of δ¹³C in bone collagen and Δ¹⁵N_Glu-Phe values of humans and animals of the Haojiatai site. The solid blue-pink line and the dashed black line indicate the regression line for C₄-diet group, the regression line for C₃-diet group, and the regression line for wild herbivores, respectively.

Additionally, the lack of correlation between δ¹⁵N in collagen and the TP_ter range from 2.5 to 3 in the C₃-diet group indicates the involvement of both wild plants and animals with lower δ¹⁵N values. Comparison of δ¹³C in collagen with Δ¹⁵N_Glu-Phe (Figure 7b) revealed no significant correlation within the C₄-diet group (P = 0.098). Conversely, a significant correlation between δ¹³C in bone collagen and Δ¹⁵N_Glu-Phe was observed in the C₃-diet group (P = 0.0037). This correlation is likely due to the consumption of a small quantity of millets, which exhibit higher δ¹³C but lower Δ¹⁵N_Glu-Phe values.

The 5% enrichment of the carbon isotope ratio from diet to collagen (Lee-Thorp, 2008) implies that the δ¹³C values in collagen for an individual exclusively consuming millet would be –3.0‰, based on the δ¹³C values of millet from Haojiatai (–8.0 ± 0.5‰, n = 5). Notably, the average δ¹³C values in bone collagen of the C₄-diet group are lower at –9.1 ± 0.7‰. This suggested that the C₄-diet group likely consumed food with lower δ¹³C values, such as wild herbivores and aquatic animals, as indicated by the amino acid results. The inverse relationship observed between δ¹⁵N_Phe and Δ¹⁵N_Glu-Phe (Figure 3) suggests a small amount of consumption of aquatic components by the C₄-diet group.

Dietary diversity and human mobility

Although only 14 individuals with enamel suitable for strontium isotope analysis were available, the investigation of population mobility reveals intriguing possibilities. Notably, all three individuals within the C₃-diet group were identified as outliers, aligning with several pieces of evidence. Firstly, collating collagen isotope results from previous studies with the present study indicates that 11 pigs (7 from previous studies and 4 from this study) and 10 cattle (6 from previous studies and 4 from this study) primarily consumed C₄ plants (Figure 6b). This dietary pattern contrasts with the expectations based on the early Neolithic site of Jiahu, known for pig domestication (Luo and Zhang, 2018) and rice cultivation (Zhang et al., 2018). It suggests an absence of pigs raised on C₃ diets at the Haojiatai site, contrasting with those associated with the C₃-diet group believed to follow the Jiahu tradition. Furthermore, even among the dogs believed to be involved in hunting, seven out of eight dogs (five from previous studies and three from this study) consumed mainly C₄ plants. This result is more consistent with the propsal that the Haojiatai site primarily focused on C₄ millet agriculture and pastoralism, with the migration of C₃ rice farmers and their companion dogs, rather than being a settlement of indigenous C₃-diet hunter-gatherer-fishers coexisting with C₄ millet farmers.

In this study, the strontium isotope ratios revealed diversity not only within the C₃-diet group but also in the C₄-diet group, suggesting the presence of a substantial number of non-locals without gender bias. This outcome contrasts with the observed female-dominated long-distance migrations during the same period (Wu et al., 2023). It implies that factors beyond marital networks also played a significant role in driving population movements and interactions among ancient societies. In comparison with previously reported strontium isotope ratios in Henan, the ⁸⁷Sr/⁸⁶Sr diversity at Haojiatai is significantly higher than at other sites in this region, such as the Jiahu site from the Early Neolithic period and the Wadian site from the Longshan period (Figure 8). Since the ⁸⁷Sr/⁸⁶Sr values in the bones of domestic animals (pigs and dogs) and humans exhibit minimal variation and are essentially identical, we infer that these values reflect the local strontium signature at the Haojiatai site. However, it is acknowledged that this inference may underestimate the diversity of strontium isotope ratios from the diet. Nevertheless, the diversity of ⁸⁷Sr/⁸⁶Sr is significantly higher than that at the Jiahu and Wadian sites, potentially highlighting a unique characteristic of the Haojiatai site.

Figure 8.

Comparison of ⁸⁷Sr/⁸⁶Sr values of the Haojiatai samples and those from other sites in the surrounding region. Published data include 59 Jiahu samples from Yin (2008) and Yin et al. (2008), and 16 Wadian samples from Zhao et al. (2012) and Zhao and Fang (2014).

Within the C₃- and C₄-diet groups, individuals M118 (0.71146, female, age 40 ± years) and M114 (0.71169, unknown, age 15 ± 2 years), respectively, exhibited notably lower ⁸⁷Sr/⁸⁶Sr values compared with other individuals. Previous research had identified strontium isotope ratios below 0.7120 predominantly distributed in northwestern Henan (Wu et al., 2023), which constituted a core region of the Central Plains in this period. This pattern suggests potential population movements and exchanges between the Haojiatai and the cultural core area, underscoring the interconnectedness of communities within the Central Plains.

The Haojiatai site, identified as a walled city from the Longshan period in the Central Plains, revealed a significant number of weapon relics, including stone arrowheads (50.88%) and bone arrowheads (14.19%), along with fortifications such as walls and trenches, reflecting the strong military characteristic of this settlement (HPICRA, 2012; Jia, 2021). It is also worth noting that the Haojiatai site exhibits an absence of stone tools and agricultural processing waste, suggesting the essential resources within the city, such as weapons and grains, were likely manufactured and transported from surrounding areas (Deng et al., 2021; Jia, 2021). This observation may contribute to explaining the diversity in ⁸⁷Sr/⁸⁶Sr values and the complex dietary patterns at Haojiatai. Furthermore, the dates of burials at Haojiatai span almost the entire chronology of settlement (SAM at PKU et al., 2017), indicating continuous human movements and interactions with neighboring areas from the beginning of this settlement to its abandonment. The proliferation of walled-town sites at varying levels in the Central Plains during the Longshan period signifies widespread expansion, interaction, and the merging of diverse cultures, contributing to the development of social complexity and urbanization. This study employed strontium isotope ratios, confirming that as populations migrated to the central settlements, the accumulation of information, technology, and materials played an important role in the development of civilization and the formation of the state of the following the Erlitou civilization. The coexistence of people with diverse cultures was vital for urbanization (Hou et al., 2021; Li et al., 2021), while social changes enabling large-scale immigration from various regions were equally pivotal.

Conclusion

In this study, we present the first application of amino acid nitrogen isotope analysis in Henan, providing a new insight into the utilization of aquatic and terrestrial resources by the Haojiatai humans. This study revealed that the C₃-diet group primarily depended on terrestrial resources, including rice and wild herbivores, with a small amount of C₄ foods such as millet and domestic animals. The C₄-diet group exhibited influence from C₃ foods (wild plants and herbivores) and aquatic animals, alongside the predominant consumption of a millet-based agricultural diet. The utilization of aquatic resources at the Haojiatai site appeared to be limited.

The presence of migrants may contribute to dietary shifts, and the coexistence of two distinct dietary groups at the Haojiatai site offers a unique opportunity to investigate the relationship between dietary diversification and migration patterns. This case study enhances our understanding of the development of social complexity and the urbanization process during the Longshan period in the Central Plains, China. A comprehensive investigation into the dietary habitat at Haojiatai must be coupled with an exploration of individual travel histories. In the case of the Haojiatai site, strontium and oxygen isotope analyses indicate that the C₃-diet group is likely composed entirely of foreign migrants, and a substantial proportion of the C₄-diet group was also composed of migrants. Although there is no evidence of long-distance population migration, it appears that migration between neighboring areas occurred frequently. This finding aligns with the defensive character of the Haojiatai site. The migration of populations with different cultures to urban settlements and the acceptance of foreign migrants during the Longshan period could be crucial for the development of social complexity and urbanization in the Central Plains, as well as for the formation of the early state in the succeeding Erlitou culture.

Acknowledgments

We would like to express our gratitude to all the participants who contributed to this study. This study was supported by JSPS KAKENHI grant numbers JP20H05821, JP15H05969, JP23KJ0617.

Declaration of Conflict of Interest

The authors declare no competing interests.

Author Contributions

Y.S. wrote the manuscript, performed the isotope experiments and measurements, and analyzed data. W.L. provided samples and archeological information, performed the isotope analysis of rice samples in China. Y.I. supported the amino acid isotope measurements and data analysis. Y.K. and T.H. supported the strontium isotope measurements and data analysis. L.Q., F.L., and H.Z. provided samples. M.Y. conceived the research concept and edited the manuscript.

Reference

•  Ambrose  S.H. and  DeNiro  M.J. (1986) The isotopic ecology of East African mammals. Oecologia, 69: 395–406. DOI: 10.1007/BF00377062
•  Ambrose  S.H. and  Norr  L. (1993) Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In: Lambert J.B. and Grupe G. (eds), Prehistoric Human Bone: Archaeology at the Molecular Level. Springer, Berlin, pp. 1–37. DOI: 10.1007/978-3-662-02894-0_1
•  Bataille  C.P.,  Laffoon  J., and  Bowen  G.J. (2012) Mapping multiple source effects on the strontium isotopic signatures of ecosystems from the circum‐Caribbean region. Ecosphere, 3, 1–24. DOI: 10.1890/ES12-00155.1
•  Bellwood  P. (2001) Early agriculturalist population diasporas? Farming, languages, and genes. Annual Review of Anthropology, 30: 181–207. DOI: 10.1146/annurev.anthro.30.1.181
•  Bellwood  P. (2005) First Farmers: The Origins of Agricultural Societies. Blackwell, Malden, MA.
•  Bellwood  P. (2011) The checkered prehistory of rice movement southwards as a domesticated cereal—from the Yangzi to the equator. Rice, 4: 93–103. DOI: 10.1007/s12284-011-9068-9
•  Bellwood  P. and  Renfrew  C. (2002) Examining the Farming/Language Dispersal Hypothesis. McDonald Institute for Archaeological Research, University of Cambridge.
•  Bentley  R.A. (2006) Strontium isotopes from the earth to the archaeological skeleton: a review. Journal of Archaeological Method and Theory, 13: 135–187. DOI: 10.1007/s10816-006-9009-x
•  Bestel  S.,  Bao  Y.J.,  Zhong  H.,  Chen  X.C., and  Liu  L. (2018) Wild plant use and multi-cropping at the early Neolithic Zhuzhai site in the middle Yellow River region, China. The Holocene, 28: 195–207. DOI: 10.1177/0959683617721328
•  Brinkkemper  O.,  Braadbaart  F.,  van Os  B.,  van Hoesel  A.,  van Brussel  A.A.N., and  Fernandes  R. (2018) Effectiveness of different pre‐treatments in recovering pre‐burial isotopic ratios of charred plants. Rapid Communications in Mass Spectrometry, 32: 251–261. DOI: 10.1002/rcm.8033
•  Britton  K.,  Fuller  B.T.,  Tütken  T.,  Mays  S., and  Richards  M.P. (2015) Oxygen isotope analysis of human bone phosphate evidences weaning age in archaeological populations. American Journal of Physical Anthropology, 157: 226–241. DOI: 10.1002/ajpa.22704
•  Bryant  J.D.,  Koch  P.L.,  Froelich  P.N.,  Showers  W.J., and  Genna  B.J. (1996) Oxygen isotope partitioning between phosphate and carbonate in mammalian apatite. Geochimica et Cosmochimica Acta, 60: 5145–5148. DOI: 10.1016/S0016-7037(96)00308-0
•  Chen  X.L. (2021) The development of subsistence economy and social changes in the Central Plains region during the Neolithic age: reflections based on the study of carbon and nitrogen stable isotope analysis in Henan Province. Cultural Relics in Southern China, 1: 179–190 (in Chinese with English summary).
•  Chen  X.L.,  Fang  Y.M.,  Hu  Y.W.,  Hou  Y.F.,  Lü  P., et al (2016) Isotopic reconstruction of the late Longshan period (ca. 4200–3900 BP) dietary complexity before the onset of state-Level societies at the Wadian site in the Ying River valley, Central Plains, China. International Journal of Osteoarchaeology, 26: 808–817. DOI: 10.1002/oa.2482
•  Chikaraishi  Y. (2021) Food webs viewed via stable nitrogen isotopic composition of amino acids. Low Temperature Science, 79: 13–21. DOI: 10.14943/lowtemsci.79.13 (in Japanese with English summary).
•  Chikaraishi  Y.,  Kashiyama  Y.,  Ogawa  N.O.,  Kitazato  H., and  Ohkouchi  N. (2007) Metabolic control of nitrogen isotope composition of amino acids in macroalgae and gastropods: implications for aquatic food web studies. Marine Ecology Progress Series, 342: 85–90. DOI: 10.3354/meps342085
•  Chikaraishi  Y.,  Ogawa  N.O.,  Kashiyama  Y.,  Takano  Y.,  Suga  H., et al (2009) Determination of aquatic food-web structure based on compound-specific nitrogen isotopic composition of amino acids. Limnology and Oceanography: Methods, 7: 740–750. DOI: 10.4319/lom.2009.7.740.
•  Chikaraishi  Y.,  Ogawa  N.O., and  Ohkouchi  N. (2010) Further evaluation of the trophic level estimation based on nitrogen isotopic composition of amino acids. Earth, Life, and Isotopes, 415: 36–51.
•  Childe  V.G. (1936) Man makes himself. Watts & Company, London.
•  Dansgaard  W. (1964) Stable isotopes in precipitation. Tellus, 16: 436–468. DOI: 10.3402/tellusa.v16i4.8993
•  Daux  V.,  Lécuyer  C.,  Héran  M.-A.,  Amiot  R.,  Simon  L., et al (2008) Oxygen isotope fractionation between human phosphate and water revisited. Journal of Human Evolution, 55: 1138–1147. DOI: 10.1016/j.jhevol.2008.06.006
•  Deng  Z.H. and  Qin  L. (2017) A Comparative study of the agricultural structure of the Longshan period in the Central Plains. Huaxia Archaeology, 3: 98–108. DOI: 10.16143/j.cnki.1001-9928.2017.03.008 (in Chinese with English summary).
•  Deng  Z.H.,  Zhang  H.,  Li  W.,  Liang  F.W., and  Cao  Y.P. (2021) A preliminary study of early agriculture practices at the Haojiatai site in Luohe city, Henan Province. Scientia China Earth Science, 64: 307–317. DOI: 10.1007/s11430-020-9694-3 (in Chinese with English title).
•  DeNiro  M.J. (1985) Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature, 317: 806–809. DOI: 10.1038/317806a0
•  Deniro  M.J. and  Epstein  S. (1981) Influence of diet on the distribution of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta, 45: 341–351. DOI: 10.1016/0016-7037(81)90244-1
•  Diamond  J. and  Bellwood  P. (2003) Farmers and their languages: the first expansions. Science, 300 (5619): 597–603. DOI: 10.1126/science.1078208
•  Dongmann  G.,  Nürnberg  H.W.,  Förstel  H., and  Wagener  K. (1974) On the enrichment of H₂¹⁸O in the leaves of transpiring plants. Radiation and Environmental Biophysics, 11: 41–52. DOI: 10.1007/BF01323099
•  Dorozhkin  S.V. (2009) Calcium orthophosphates in nature, biology and medicine. Materials, 2: 399–498. DOI: 10.3390/ma2020399
•  Dupanloup  I.,  Bertorelle  G.,  Chikhi  L., and  Barbujani  G. (2004) Estimating the impact of prehistoric admixture on the genome of europeans. Molecular Biology and Evolution, 21: 1361–1372. DOI: 10.1093/molbev/msh135
•  Durrant  S.F. and  Ward  N.I. (2005) Recent biological and environmental applications of laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). Journal of Analytical Atomic Spectrometry, 20: 890–919. DOI: 10.1039/b502206a
•  Evans  J.A.,  Chenery  C.A., and  Fitzpatrick  A.P. (2006) Bronze Age childhood migration of individuals near Stonehenge, revealed by strontium and oxygen isotope tooth enamel analysis. Archaeometry, 48: 309–321. DOI: 10.1111/j.1475-4754.2006.00258.x
•  Faure  G. and  Mensing  T.M. (2005) The Rb–Sr method. In: Faure G. and Mensing T.M (eds.), Isotope Principles and Application, 3rd edn. John Wiley, Hoboken, NJ, pp. 75–112.
•  Hedges  R.E.M.,  Clement  J.G.,  Thomas  C.D.L., and  O’Connell  T.C. (2007) Collagen turnover in the adult femoral mid-shaft: modeled from anthropogenic radiocarbon tracer measurements. American Journal of Physical Anthropology, 133: 808–816. DOI: 10.1002/ajpa.20598
• Henan Provincial Institute of Cultural Relics and Archaeology (2012) Yancheng Haojiatai. Elephant Press, Henan (in Chinese with English title).
•  Hou  L.L.,  Wei  D., and  Richards  P.M. (2021) Diversification origins of the populations from the Nancheng cemetery during proto-Shang culture (ca. 2000 to 1600 BC) evidenced by strontium isotope analysis. Quaternary Sciences, 41: 235–246 (in Chinese with English summary).
•  Hu  Y.W.,  Ambrose  S.H., and  Wang  C.S. (2006) Stable isotopic analysis of human bones from Jiahu site, Henan, China: implications for the transition to agriculture. Journal of Archaeological Science, 33: 1319–1330. DOI: 10.1016/j.jas.2006.01.007
• International Atomic Energy Agency (2022) Water Isotope System for Electronic Retrieval (WISER) online database. https://nucleus.iaea.org/wiser/. Accessed 18 February 2024.
•  Itahashi  Y.,  Miyake  Y.,  Maeda  O.,  Kondo  O.,  Hongo  H., et al (2017) Preference for fish in a Neolithic hunter-gatherer community of the upper Tigris, elucidated by amino acid δ¹⁵N analysis. Journal of Archaeological Science, 82: 40–49. DOI: 10.1016/j.jas.2017.05.001
•  Itahashi  Y.,  Ananyevskaya  E.,  Yoneda  M.,  Ventresca Miller  A.R.,  Nishiaki  Y., and  Matuzeviciute  G.M. (2020) Dietary diversity of Bronze–Iron Age populations of Kazakhstan quantitatively estimated through the compound-specific nitrogen analysis of amino acids. Journal of Archaeological Science: Reports, 33: 102565. DOI: 10.1016/j.jasrep.2020.102565
•  Jia  Y.Q. (2021) Haojiatai yizhi longshan wenhua shizu guzu de chubuyanjiu. Huanghe, Huangtu, Huangzhongren, 9: 23–26 (in Chinese).
•  Jiang  L.P. and  Liu  L. (2006) New evidence for the origins of sedentism and rice domestication in the Lower Yangzi River, China. Antiquity, 80: 355–361. DOI: 10.1017/S0003598X00093674
•  Kohn  M.J. (1996) Predicting animal δ¹⁸O: accounting for diet and physiological adaptation. Geochimica et Cosmochimica Acta, 60: 4811–4829. DOI: 10.1016/S0016-7037(96)00240-2
•  Kohn  M.J. and  Cerling  T.E. (2002) Stable isotope compositions of biological apatite. Reviews in Mineralogy and Geochemistry, 48: 455–488. DOI: 10.2138/rmg.2002.48.12
•  Kohn  M.J.,  Schoen1nger  M.J., and  Valley  J.W. (1996) Herbivore tooth oxygen isotope compositions: Effects of diet and physiology. Geochimica et Cosmochimica Acta, 60: 3889–3896. DOI: 10.1016/0016-7037(96)00248-7
•  Kusaka  S.,  Nakano  T.,  Yumoto  T., and  Nakatsukasa  M. (2011) Strontium isotope evidence of migration and diet in relation to ritual tooth ablation: a case study from the Inariyama Jomon site, Japan. Journal of Archaeological Science, 38: 166–174. DOI: 10.1016/j.jas.2010.09.001
•  Kusaka  S.,  Saito  T.,  Ishimaru  E., and  Yamada  Y. (2022) Strontium isotope analysis on human skeletal remains from the Hobi and Ikawazu shell-mounds in Aichi Prefecture, Japan. Anthropological Science, 130: 25–32. DOI: 10.1537/ase.2202191
•  Kutanan  W.,  Kampuansai  J.,  Brunelli  A.,  Ghirotto  S.,  Pittayaporn  P., et al (2018) New insights from Thailand into the maternal genetic history of mainland Southeast Asia. European Journal of Human Genetics, 26: 898–911. DOI: 10.1038/s41431-018-0113-7
•  Laffoon  J.E.,  Sonnemann  T.F.,  Shafie  T.,  Hofman  C.L.,  Brandes  U., and  Davies  G.R. (2017) Investigating human geographic origins using dual-isotope (⁸⁷Sr/⁸⁶Sr, δ¹⁸O) assignment approaches. PLoS One, 12: e0172562. DOI: 10.1371/journal.pone.0172562
•  Lanehart  R.E.,  Tykot  R.H.,  Underhill  A.P.,  Luan  F.S.,  Yu  H.G., et al (2011) Dietary adaptation during the Longshan period in China: stable isotope analyses at Liangchengzhen (southeastern Shandong). Journal of Archaeological Science, 38: 2171–2181. DOI: 10.1016/j.jas.2011.03.011
•  Lazaridis  I.,  Patterson  N.,  Mittnik  A.,  Renaud  G.,  Mallick  S., et al (2014) Ancient human genomes suggest three ancestral populations for present-day Europeans. Nature, 513: 409–413. DOI: 10.1038/nature13673
•  Lazaridis  I.,  Nadel  D.,  Rollefson  G.,  Merrett  D.C.,  Rohland  N., et al (2016) Genomic insights into the origin of farming in the ancient Near East. Nature, 536: 419–424. DOI: 10.1038/nature19310
•  Lee-Thorp  J.A. (2008) On isotopes and old bones. Archaeometry, 50: 925–950. DOI: 10.1111/j.1475-4754.2008.00441.x
•  Li  W. (2022) Diyuan jiehe: zhongyuan longshan shidai chengshihua jincheng xintan. Kaoguxue Yanjiu, 15: 305–329 (in Chinese).
•  Li  W.,  Zhou  L.G.,  Lin  Y.H.,  Zhang  H.,  Zhang  Y., et al (2021) Interdisciplinary study on dietary complexity in Central China during the Longshan period (4.5–3.8 kaBP): new isotopic evidence from Wadian and Haojiatai, Henan Province. The Holocene, 31: 258–270. DOI: 10.1177/0959683620970252
•  Lin  Y.X. (2022) The Diet and Subsistence Study on the Core Area of Central Plain in China from 3500 to 1500 BC based on Stable Isotope Bayesian Analysis. PhD thesis, Peking University, Beijing (in Chinese with English summary).
•  Lipson  M.,  Cheronet  O.,  Mallick  S.,  Rohland  N.,  Oxenham  M., et al (2018) Ancient genomes document multiple waves of migration in Southeast Asian prehistory. Science, 361: 92–95. DOI: 10.1126/science.aat3188
•  Liu  C. and  Fang  Y.M. (2010) Henan yuzhou wadian yizhi chutu zhiwu yicun fenxi. Nanfang Wenwu, 4: 55–64 (in Chinese).
•  Liu  C.,  Zhao  Z.J., and  Fang  Y.M. (2018) Analysis of the 2007–2009 flotation results of plant remains from the Wadian site, Yuzhou, Henan Province. Huaxia Archaeology, 1: 95–102 (in Chinese with English summary).
•  Liu  Y.C.,  Mao  X.W.,  Krause  J., and  Fu  Q.M. (2021) Insights into human history from the first decade of ancient human genomics. Science, 373 (6562): 1479–1484. DOI: 10.1126/science.abi8202
•  Longin  R. (1971) New method of collagen extraction for radiocarbon dating. Nature, 230: 241–242. DOI: 10.1038/230241a0
•  Liu  L.,  Lee  G.-A.,  Jiang  L.P., and  Zhang  J.Z. (2007) Evidence for the early beginning (c. 9000 cal. BP) of rice domestication in China: a response. The Holocene, 17: 1059–1068. DOI: 10.1177/0959683607085121
•  Longinelli  A. (1984) Oxygen isotopes in mammal bone phosphate: a new tool for paleohydrological and paleoclimatological research. Geochimica et Cosmochimica Acta, 48: 385–390. DOI: 10.1016/0016-7037(84)90259-X
•  Luo  W.H.,  Gu  C.G.,  Yang  Y.Z.,  Zhang  D.,  Liang  Z.H., et al (2019) Phytoliths reveal the earliest interplay of rice and broomcorn millet at the site of Shuangdun (ca. 7.3–6.8 ka BP) in the middle Huai River valley, China. Journal of Archaeological Science, 102: 26–34. DOI: 10.1016/j.jas.2018.12.004
•  Luo  Y.B. and  Zhang  J.Z. (2018) Restudy of pig bones excavated from Jiahu site, Wuyang, Henan Province, China. Archaeology, 1: 90–96 (in Chinese with English summary).
•  Luz  B.,  Kolodny  Y., and  Horowitz  M. (1984) Fractionation of oxygen isotopes between mammalian bone-phosphate and environmental drinking water. Geochimica et Cosmochimica Acta, 48: 1689–1693. DOI: 10.1016/0016-7037(84)90338-7
•  Ma  L.F. (2002) Geological Atlas of China. Geological Publishing House, Beijing (in Chinese with English title).
•  McClelland  J.W. and  Montoya  J.P. (2002) Trophic relationships and the nitrogen isotopic composition of amino acids in plankton. Ecology, 83: 2173–2180. DOI: 10.1890/0012-9658(2002)083[2173:TRATNI]2.0.CO;2
•  McColl  H.,  Racimo  F.,  Vinner  L.,  Demeter  F.,  Gakuhari  T., et al (2018) The prehistoric peopling of Southeast Asia. Science, 361 (6397): 88–92. DOI: 10.1126/science.aat3628
•  Minagawa  M. and  Wada  E. (1984) Stepwise enrichment of ¹⁵N along food chains: further evidence and the relation between δ¹⁵N and animal age. Geochimica et Cosmochimica Acta, 48: 1135–1140. DOI: 10.1016/0016-7037(84)90204-7
•  Moore  L.J.,  Murphy  T.J.,  Barnes  I.L., and  Paulsen  P.J. (1982) Absolute isotopic abundance ratios and atomic weight of a reference sample of strontium. Journal of Research of the National Bureau of Standards, 87: 1–8. DOI: 10.6028/jres.087.001
•  Moritz  G.L.,  Fourie  N.,  Yeakel  J.D.,  Phillips-Conroy  J.E.,  Jolly  C.J., et al (2012) Baboons, water, and the ecology of oxygen stable isotopes in an arid hybrid zone. Physiological and Biochemical Zoology, 85: 421–430. DOI: 10.1086/667533
•  Naito  Y.I.,  Honch  N.V.,  Chikaraishi  Y.,  Ohkouchi  N., and  Yoneda  M. (2010) Quantitative evaluation of marine protein contribution in ancient diets based on nitrogen isotope ratios of individual amino acids in bone collagen: an investigation at the Kitakogane Jomon site. American Journal of Physical Anthropology, 143: 31–40. DOI: 10.1002/ajpa.21287
•  Naito  Y.I.,  Bocherens  H.,  Chikaraishi  Y.,  Drucker  D.G.,  Wißing  C., et al (2016) An overview of methods used for the detection of aquatic resource consumption by humans: compound-specific delta N-15 analysis of amino acids in archaeological materials. Journal of Archaeological Science: Report, 6: 720–732. DOI: 10.1016/j.jasrep.2015.11.025
•  Ohkouchi  N.,  Chikaraishi  Y.,  Close  H.G.,  Fry  B.,  Larsen  T., et al (2017) Advances in the application of amino acid nitrogen isotopic analysis in ecological and biogeochemical studies. Organic Geochemistry, 113: 150–174. DOI: 10.1016/j.orggeochem.2017.07.009
•  Ohno  T. and  Hirata  T. (2007) Simultaneous determination of mass-dependent isotopic fractionation and radiogenic isotope variation of strontium in geochemical samples by multiple collector-ICP-mass spectrometry. Analytical Sciences, 23: 1275–1280. DOI: 10.2116/ANALSCI.23.1275
•  O’Leary  M.H. (1988) Carbon isotopes in photosynthesis. Bioscience, 38: 328–336. DOI: 10.2307/1310735
•  Oliveira  S.,  Nägele  K.,  Carlhoff  S.,  Pugach  I.,  Koesbardiati  T., et al (2022) Ancient genomes from the last three millennia support multiple human dispersals into Wallacea. Nature Ecology & Evolution, 6: 1024–1034. DOI: 10.1038/s41559-022-01775-2
•  Passey  B.H.,  Cerling  T.E., and  Levin  N.E. (2007) Temperature dependence of oxygen isotope acid fractionation for modern and fossil tooth enamels. Rapid Communications in Mass Spectrometry, 21: 2853–2859. DOI: 10.1002/rcm.3149
•  Pederzani  S. and  Britton  K. (2019) Oxygen isotopes in bioarchaeology: principles and applications, challenges and opportunities. Earth-Science Reviews, 188: 77–107. DOI: 10.1016/j.earscirev.2018.11.005
•  Podlesak  D.W.,  Torregrossa  A.,  Ehleringer  J.R.,  Dearing  M.D.,  Passey  B.H., and  Cerling  T.E. (2008) Turnover of oxygen and hydrogen isotopes in the body water, CO₂, hair, and enamel of a small mammal. Geochimica et Cosmochimica Acta, 72: 19–35. DOI: 10.1016/j.gca.2007.10.003
•  Potapov  A.M.,  Tiunov  A.V.,  Scheu  S.,  Larsen  T., and  Pollierer  M.M. (2019) Combining bulk and amino acid stable isotope analyses to quantify trophic level and basal resources of detritivores: a case study on earthworms. Oecologia, 189: 447–460. DOI: 10.1007/s00442-018-04335-3
•  Price  T.D. and  Bar-Yosef  O. (2011) The origins of agriculture: new data, new ideas: an introduction to supplement. Current Anthropology, 52(S4): S163–S174. DOI: 10.1086/659964
• R Core Team (2021) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. https://www.R-project.org/. Accessed 10 May 2024.
•  Reid  R.E.B.,  Jones  M.,  Brandt  S.,  Bunn  H., and  Marshall  F. (2019) Oxygen isotope analyses of ungulate tooth enamel confirm low seasonality of rainfall contributed to the African Humid Period in Somalia. Palaeogeography, Palaeoclimatolgy, Palaeoecology, 534: 109272. DOI: 10.1016/j.palaeo.2019.109272
•  Renfrew  C. (1990) Archaeology and Language: The Puzzle of Indo-European Origins. Cambridge University Press, Cambridge.
•  Renfrew  C. (2002) The emerging synthesis: the archaeogenetics of farming/language dispersals and other spread zones. In: Bellwood P. and Renfrew C. (eds), Examining the Farming/Language Dispersal Hypothesis, McDonald Institute for Archaeological Research, Cambridge, pp. 3–16.
•  Robbeets  M.,  Bouckaert  R.,  Conte  M.,  Savelyev  A.,  Li  T., et al (2021) Triangulation supports agricultural spread of the Transeurasian languages. Nature, 599: 616–621. DOI: 10.1038/s41586-021-04108-8
•  Russell  W.A.,  Papanastassiou  D.A., and  Tombrello  T.A. (1978) Ca isotope fractionation on the Earth and other solar system materials. Geochimica et Cosmochimica Acta, 42: 1075–1090. DOI: 10.1016/0016-7037(78)90105-9
•  Ryan  S.E.,  Snoeck  C.,  Crowley  Q.G., and  Babechuk  M.G. (2018) ⁸⁷Sr/⁸⁶Sr and trace element mapping of geosphere–hydrosphere–biosphere interactions: a case study in Ireland. Applied Geochemistry, 92: 209–224. DOI: 10.1016/j.apgeochem.2018.01.007
• School of Archaeology and Museology at Peking University (SAM at PKU), Henan Provincial Institute of Cultural Relics and Archaeology, Luohe Municipal Institute of Cultural Relics and Archaeology (2017) Major field archaeological achievements in 2015–2016 at Haojiatai site, Luohe, Henan, China. Huaxia Archaeology, 3: 14–49 (in Chinese with English summary).
•  Schwarcz  H.P. and  Schoeninger  M.J. (1991) Stable isotope analyses in human nutritional ecology. Yearbook of Physical Anthropology, 34: 283–321. DOI: 10.1002/ajpa.1330340613
•  Simões  L.G.,  Günther  T.,  Martínez-Sánchez  R.M.,  Vera-Rodríguez  J.C.,  Iriarte  E., et al (2023) Northwest African Neolithic initiated by migrants from Iberia and Levant. Nature, 618: 550–556. DOI: 10.1038/s41586-023-06166-6
•  Steiger  R.H. and  Jäger  E. (1977) Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36: 359–362. DOI: 10.1016/0012-821X(77)90060-7
•  Stoneking  M.,  Arias  L.,  Liu  D.,  Oliveira  S.,  Pugach  I., et al (2023) Genomic perspectives on human dispersals during the Holocene. Proceedings of the National Academy of Sciences of the USA, 120: e2209475119. DOI: 10.1073/pnas.2209475119
•  van Klinken  G.J. (1999) Bone collagen quality indicators for palaeodietary and radiocarbon measurements. Journal of Archaeological Science, 26: 687–695. DOI: 10.1006/jasc.1998.0385
•  Vennemann  T.W.,  Fricke  H.C.,  Blake  R.E.,  O’Neil  J.R., and  Colman  A. (2002) Oxygen isotope analysis of phosphates: a comparison of techniques for analysis of Ag₃PO₄. Chemical Geology, 185: 321–336. DOI: 10.1016/S0009-2541(01)00413-2
•  Wang  C.,  Lu  H.Y.,  Gu  W.F.,  Wu  N.Q.,  Zhang  J.P., et al (2017) The spatial pattern of farming and factors influencing it during the Peiligang culture period in the middle Yellow River valley, China. Science Bulletin (Beijing), 62: 1565–1568. DOI: 10.1016/j.scib.2017.10.003
•  Wang  C.C.,  Yeh  H.-Y.,  Popov  A.N.,  Zhang  H.,  Matsumura  H., et al (2021) Genomic insights into the formation of human populations in East Asia. Nature, 591: 413–419. DOI: 10.1038/s41586-021-03336-2
•  Wang  X. (2018) Stable Isotopic Study of Manuring Management in the Middle Yellow River Valley China during the Late Neolithic. PhD thesis, University of Chinese Academy of Science (in Chinese with English summary).
•  Wang  X.Y. and  Tang  Z.H. (2020) The first large-scale bioavailable Sr isotope map of China and its implication for provenance studies. Earth-Science Reviews, 210: 103353. DOI: 10.1016/j.earscirev.2020.103353
•  Wang  X.Y.,  Shen  H.,  Wei  D.,  Hu  X.J.,  Xu  B., et al (2020) Human mobility in the Lop Nur region during the Han–Jin Dynasties: a multi-approach study. Archaeological and Anthropological Science, 12: 20. DOI: 10.1007/s12520-019-00956-8
• Waterisotope Database (2017) http://waterisotopesDB.org. Accessed 18 February 2024.
•  Wright  E.L. and  Schwarcz  P.H. (1998) Stable carbon and oxygen isotopes in human tooth enamel: identifying breastfeeding and weaning in prehistory. American Journal of Physical Anthropology, 106: 1–18. DOI: 10.1002/(SICI)1096-8644(199805)106:1%3C1::AID-AJPA1%3E3.0.CO;2-W
•  Wu  X.H.,  Shu  T.,  Xiao  H.D.,  Yuan  J.,  Zhang  J.Z., and  Fang  F. (2015) The stable isotopic analysis of human and animal bones. In: Henan Provincial Institute of Cultural Relics and Archaeology (eds.), Wuyang Jiahu (II). Science Press, Beijing, pp. 378–388 (in Chinese with no English title).
•  Wu  X.T.,  Guo  Z.Y.,  He  N.,  Zhao  S.J.,  Li  Y.X., et al (2023) Intermarriage and ancient polity alliances: isotopic evidence of cross-regional female exogamy during the Longshan period (2500–1900 BC). Antiquity, 98: 48–65. DOI: 10.15184/aqy.2023.173
•  Yang  M.A.,  Fan  X.C.,  Sun  B.,  Chen  C.Y.,  Lang  J.F., et al (2020) Ancient DNA indicates human population shifts and admixture in northern and southern China. Science, 369 (6501): 282–288. DOI: 10.1126/science.aba0909
•  Yin  R.C. (2008) Prehistoric Human Migration in Jiahu Relics: Evidence from Strontium isotopes Analysis. PhD thesis, University of Science and Technology of China, Anhui (in Chinese with English summary).
•  Yin  R.C.,  Zhang  J.Z., and  Yang  X.Y. (2008) Preliminary study of prehistoric human migration based on Sr isotope analysis from Jiahu relics. Quaternary Sciences, 28: 50–57 (in Chinese with English summary).
•  Yoneda  M.,  Hirota  M.,  Uchida  M.,  Tanaka  A.,  Shibata  Y., et al (2002) Radiocarbon and stable isotope analyses on the earliest Jomon skeletons from the Tochibara rockshelter, Nagano, Japan. Radiocarbon, 44(2): 549–557.
•  Yoneda  M.,  Kikuchi  Y.,  Nasu  H., and  Yamazaki  K. (2019) Doitai bunseki ni yoru yayoi jidai no suito riyo no hyoka ni mukete: doitai seitaigaku teki na haikei to jikken suiden ni okeru kiso kenkyu. In: Shitara H. (ed.), Noko Bunka Fukugo Keisei No Kokogaku (part 2)—Noko Ga Motarashita Mono. Yuzankaku, Tokyo, pp. 209–230 (in Japanese).
•  Yuan  J. (2010) Zooarchaeological study of the domestic animals in ancient China. Quaternary Sciences, 30: 298–306. DOI: 10.3969/j.issn.1001-7410.2010.02.06 (in Chinese with English summary).
•  Zazzo  A.,  Lécuyer  C.,  Sheppard  S.M.F.,  Grandjean  P., and  Mariotti  A. (2004) Diagenesis and the reconstruction of paleoenvironments: a method to restore original δ¹⁸O values of carbonate and phosphate from fossil tooth enamel. Geochimica et Cosmochimica Acta, 68: 2245–2258. DOI: 10.1016/j.gca.2003.11.009
•  Zhang  J.F. (2015) The Wadian site in Yuzhou, Henan Province, and its related problems. Cultural relics from Central Plains. Relics from Central Plain, 4: 35–86 (in Chinese with English summary).
•  Zhang  W.X. and  Yuan  J.R. (1998) A preliminary study on the ancient rice excavated from Yuchanyan, Daoxian, Hunan Province. Acta Agronomica Sinica, 24: 416–420 (in Chinese with English summary).
•  Zhang  C.,  Pollard  A.M.,  Rawson  J.,  Huan  L.M.,  Liu  R.L., and  Tang  X.J. (2019a) China’s major Late Neolithic centers and the rise of Erlitou. Antiquity, 93: 588–603. DOI: 10.15184/aqy.2019.63
•  Zhang  J.P.,  Lu  H.Y.,  Gu  W.F.,  Wu  N.Q.,  Zhou  K.S., et al (2012) Early mixed farming of millet and rice 7800 years ago in the Middle Yellow River region, China. PLoS One, 7: e52146. DOI: 10.1371/journal.pone.0052146
•  Zhang  J.Z.,  Chen  Z.J.,  Lan  W.L.,  Yang  Y.Z.,  Luo  W.H., et al (2018) New progress in archaeobotanical research at Jiahu site, Wuyang, Henan, China. Archaeology, 4: 100–110 (in Chinese with English summary).
•  Zhang  M.H.,  Yan  S.,  Pan  W.Y., and  Jin  L. (2019b) Phylogenetic evidence for Sino-Tibetan origin in northern China in the Late Neolithic. Nature, 569: 112–115. DOI: 10.1038/s41586-019-1153-z
•  Zhao  C.Y. and  Fang  Y.M. (2014) Analysis of strontium isotope ratios of human tooth enamel from Wadian, Yuzhou. Huaxia Archaeology, 3: 123–127 (in Chinese with English summary).
•  Zhao  C.Y.,  Lu  P.,  Yuan  J., and  Fang  Y.M. (2012) Analysis of element and strontium isotope ratios of animal remains from Wadian site in Yuzhou City, Henan Province. Archaeology, 11: 89–96 (in Chinese with English summary).
•  Zhao  Z.J. (1998) The Middle Yangtze region in China is one place where rice was domesticated: phytolith evidence from the Diaotonghuan cave, Northern Jiangxi. Antiquity, 72: 885–897. DOI: 10.1017/S0003598X00087524
•  Zhao  Z.J. (2017) The development of agriculture in the time of Yangshao culture and the establishment of agricultural society: an analysis on the flotation result of Yuhuazhai site. Jianghan Archaeology, 87: 149–200 (in Chinese with English summary).
•  Zhao  Z.J. (2019) Introduction of the origin of agriculture in China. Research on Heritages and Preservation, 4: 1–7 (in Chinese with English title).
•  Zhao  Z.J. and  Zhang  J.Z. (2009) Jiahu yizhi 2001 niandu fuxuan jieguo fenxi baogao. Archaeology, 8: 84–94 (in Chinese with English summary).
•  Zhou  L.G. (2017) A study of the tombs and society of the Henan Longshan culture by means of stable carbon and nitrogen isotope analysis. Huaxia Archaeology, 3: 145–152 (in Chinese with English summary).