Threads of Early China

From caves on the Guangxi karst to high valleys of Xinjiang, the ancient Chinese archaeological record unfolds like layered silk. Archaeological data indicates settled farming communities along the Yellow River and its tributaries by the Middle Neolithic, while more marginal regions in the northeast and the Tarim Basin show distinct material cultures and funerary practices. Sites in this dataset such as Wuzhuangguoliang (Shaanxi), Longtoushan (Hexigten, Inner Mongolia), Baojianshan Cave A (Guangxi) and multiple Xinjiang loci (Xiaohe, Jierzankale, Liushui, Jirentaigoukou) span 6690 BCE to the late Han period (248 CE), offering a rare long-term view.

Material change—ceramics, house plans, metallurgy and burial complexity—marks the slow emergence of regional hierarchies during the Late Neolithic and Bronze Age phases (e.g., Shimao-related complexity in northern Shaanxi and the West Liao River cultural horizon). Archaeobotanical remains and village architecture show intensifying agriculture in central China, while Xinjiang and Inner Mongolia record mixed subsistence strategies that include pastoralism and long-distance exchange. Limited evidence suggests some sites functioned as ritual centers, others as seasonal camps.

Taken together, the archaeological picture is of strong local continuity punctuated by episodes of mobility and contact. Genetic sampling across these sites allows us to test whether material links reflect people moving, ideas spreading, or both.

Genetic data from 316 individuals spanning 6690 BCE to 248 CE reveals a nuanced story of continuity and contact. Y-chromosome haplogroup O, the dominant East Asian male lineage in our samples (36 counts), is concentrated in central and eastern sites and fits archaeological evidence for long-term local continuity in Yellow River and southern regions. In contrast, elevated counts of R (26), Q (19) and C (21) are more frequent in Xinjiang and some northern sites, consistent with western Eurasian and northern Eurasian male-line influences—particularly during Bronze and Iron Age horizons when steppe mobility intensified.

Mitochondrial haplogroups also record mixture across maternal lines. East Asian‑linked mtDNA D (50) and B (16) are common and widespread, while northern/central Asian markers such as C4 (18) appear in Inner Mongolian and Xinjiang contexts. Notably, haplogroups typical of western Eurasia (U, 25; H, 18) are present, especially in Xinjiang Bronze Age assemblages like Xiaohe and later Iron Age burials, indicating female-mediated gene flow or complex ancestry mixtures.

Archaeogenetic patterns align with archaeology: central China shows genetic continuity through the Neolithic into historic times, while Xinjiang and the northern frontiers are genetic mosaics reflecting repeated contacts across the Inner Asian corridor. Statistical analyses point to admixture pulses rather than wholesale population replacement. While the sample size (316) is robust for detecting regional trends, fine-scale demographic inferences for specific sites should be treated with caution and refined as more genomes become available.