The Jomon — the forager societies that held the Japanese archipelago for over ten thousand years, firing some of the world's oldest pottery while everyone else was inventing agriculture — have been one of ancient genomics' most undersampled famous populations. A study published this month in Science Advances (Watanabe et al.) changes that at a stroke: 42 Jomon genomes, 25 of them newly sequenced from petrous bones across Japan, analysed for origins, structure and — the headline — natural selection. The Jomon, it turns out, carry a legible genetic toolkit for surviving the Ice Age's northeast.
A deep lineage, structured by islands#
The expanded dataset sharpens the Jomon's place in the East Eurasian family tree: a deeply divergent lineage that separated from the ancestors of other East Asians in the Upper Palaeolithic — on the order of twenty-seven thousand years ago by the study's reconstruction — and then evolved in relative isolation at the continent's edge. Within the archipelago, the genomes resolve internal structure between Hokkaido and the main-island (Hondo) Jomon populations, diverging thousands of years before farming reached Japan: even inside their island world, the Jomon were never one undifferentiated people.
Their legacy persists: modern Japanese populations carry a minority share of Jomon ancestry — layered under the dominant continental stream that arrived with rice agriculture in the Yayoi period — with the share famously higher among the Ainu and in Okinawa. That structure is exactly what era-aware ancestry methods exist to read, and why a Japanese genome's "closest populations" list changes character between ancient and modern reference panels.
The cold toolkit#
The study's centrepiece is selection: scanning the Jomon genomes for the marks of adaptation, the team reports signals in genes tied to cold response — the physiological package (lipid metabolism and related pathways) that let Upper Palaeolithic foragers winter at the northeast edge of Eurasia. It is among the first genetic frameworks for how ancient East Eurasians adapted to glacial environments — and some of those variants still circulate in living populations, an Ice Age inheritance carried by people who will never see a mammoth steppe.
Two disciplined cautions travel with the finding, and they are the same ones any selection claim deserves: selection scans identify candidate loci whose histories fit adaptation, not certified survival genes; and adaptation stories are about populations, never a warrant for reading any individual's genome as "built for cold". The methods that decompose ancestry — f-statistics, qpAdm, coordinate systems — answer where from; selection scans answer what happened along the way. The Jomon study is a model of keeping the two questions separate and answering both.
The details worth keeping#
- 42 Jomon individuals — the largest Jomon genomic dataset yet, 25 newly sequenced — spanning the archipelago.
- A deep Upper Palaeolithic divergence from other East Asian ancestors, with internal Hokkaido–Hondo structure resolved millennia before agriculture.
- Selection signals in cold-response pathways: a genetic record of Ice Age survival at Eurasia's eastern edge.
- Jomon ancestry persists as a minority stream in modern Japan — highest among Ainu and Okinawans — under the Yayoi-era continental majority.
Terms used here are defined in the glossary.
Source#
- Watanabe, Y. et al. (2026). Jomon genomics reveal cold adaptation in Upper Paleolithic hunter-gatherers of eastern Eurasia. Science Advances, eaea7469 (published August 2026).
