Back

The south Congo Basin was critical to Bantu settlement of south central Africa

Choin, J.; de Luna, K.; Fleisher, J.; Sawchuk, E.; Goldstein, S.; Kaliba, P.; Katongo, M.; Morris, A. G.; Mudenda, G.; Welling, M.; Sirak, K.; Callan, K.; Iliev, L.; Lawson, A. M.; Michel, M.; Oppenheimer, J.; Qiu, L.; Workman, J. N.; Kearns, A.; Mah, M.; Soos, G.; Mallick, S.; Rohland, N.; Thompson, J.; Prendergast, M. E.; Reich, D.

2026-08-19 evolutionary biology
10.64898/2026.08.14.741591 bioRxiv
Show abstract

South central Africa, between the Congo Basin, the Great Lakes, and southern Africa, has long served as a corridor for human movement. Yet, it remains unclear whether there is genetic continuity between pre-Iron Age foragers and later Bantu-associated populations and whether the settlement of this region reflects multiple Bantu-associated migrations rather than the simpler serial founder model suggested by existing genomic data. To do so, we generated genome-wide ancient DNA from 71 Iron Age and historical individuals from present-day Zambia and Malawi, and analyzed these genomes alongside published data from present-day Africans. One late Iron Age individual (16th-17th century) from Kalala Island in the Kafue River, Zambia, carries 40% non Bantu related ancestry that closely matches local Later Stone Age foragers. Admixture for this individual is estimated to be 850 years ago, several centuries earlier than reported for present day BaTwa from the same region. Focusing on the Bantu-related ancestry, haplotype-based analyses identify two main clusters among Iron Age, historical, and present-day south central Bantu groups associated with different Bantu-related migrations. These results reveal a layered history in which at least two Bantu expansions radiated from the southern Congo Basin, with south central Africa acting both as a crossroads of these movements and as a staging area for the subsequent southward expansion toward southern Africa.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.