Did range expansion leave early modern Europeans genetically indistinguishable from Neanderthals?
In a preprint posted on bioRxiv.com, American geneticist David Reich has proposed that a critical event in Neanderthal evolution was a range expansion of early modern humans into Europe 300,000 years ago1. As early modern human groups spread, they interbred with indigenous archaic people, leading ultimately to a resulting population that carried around 3 to 7 per cent Homo sapiens ancestry.
Ancient DNA from two hominins found at the Sima de le Huesos cave site in northern Spain has posed a puzzle. The remains date to 430,000 years BP2. Nuclear DNA suggests a closer affinity to Neanderthals than to Denisovans3. Neanderthals are thought to have diverged from Denisovans before 430,000 years BP4, and it is to be expected that the Sima hominins should be more closely related to a European species than to an Asian species. However, analysis of ancient mitochondrial DNA suggests that the Sima hominins were more closely related to Denisovans than they were to Neanderthals, and Neanderthals were more closely related to Homo sapiens than they were to the Sima hominins or Denisovans5. Ancient Y-chromosomal DNA analysis reveals a similar picture. The Denisovan Y chromosome diverged from the common ancestor of the Neanderthal/Homo sapiens Y chromosome 700,000 years BP, but the Homo sapiens and Neanderthal Y chromosomes diverged only 370,000 years BP6.
Mitochondrial and Y-chromosomal DNA make up only a small fraction of the human genome, so these results are unlikely to overturn those from nuclear DNA. An explanation is nevertheless required. One possibility is that the mitochondrial lineages of Late Pleistocene Neanderthals were acquired after 430,000 years BP. The likeliest interpretation of this is that an African population associated with the Homo sapiens clade migrated to Europe and interbred with early Neanderthals. As a result, the original mitochondrial lineages were replaced by the African lineages3.
This model has been tested against mitochondrial DNA from a 124,000-year-old Neanderthal thigh bone from the cave site of Hohlenstein-Stadel, southwestern Germany. The mitochondrial DNA was found to have diverged from that of Homo sapiens360,230–467,720 years BP, and from that of other Neanderthals 218,980–316,080 years BP. These dates set upper and lower limits of 467,720 years BP and 218,980 years BP, respectively, on the date of such a migration from Africa. Given the small size of the Neanderthal population in Europe, even a very limited introgression could have eventually resulted in the replacement of the indigenous lineages7.
The same explanation could apply to the Y chromosomal lineages. Here, an additional factor in the replacement could have been reproductive fitness. Deleterious mutations to Neanderthal Y chromosomes could have resulted in Neanderthal males with the early Homo sapiens version gaining a reproductive advantage6.
In both cases, it was suggested that the replacement was due to a ‘fitness gap’, i.e., the reproductive or selective advantage early modern human genetic lineages possessed over archaic lineages to explain why original Neanderthal mitochondrial and Y-chromosomal lineages were replaced by Homo sapiens lineages. Deleterious mutations are more likely to accumulate in small populations, where selection pressure to remove them is weaker.
There is evidence for interbreeding between Neanderthals and early Homo sapiens in Eurasia: studies suggest that 3 to 7 per cent of the Neanderthal genome introgressed from early Homo sapiens8–10.
Reich challenges the view that an archaic-modern fitness gap was sufficient to retain modern mitochondrial and Y-chromosomal DNA. He claimed that the selective advantage of the modern genes was too small, giving less than a 3 per cent chance that both modern mitochondrial and Y-chromosomal DNA would fix in the archaic population. The proposal that this would happen purely because Neanderthals suffered from a high genetic load of deleterious mutations fails under realistic scenarios for several reasons.
The actual fitness gap was too small. Previous models assumed a selective advantage of 1–2 per cent in favour of modern human DNA, but when Reich considered realistic mutation rates, demography, and biological constraints, he obtained a much smaller fitness gap: 0.32 per cent for mitochondrial DNA and 0.09–0.17 per cent for Y-chromosomal DNA. Such a small selective advantage would likely be overwhelmed by genetic drift in realistic population models. Mitochondrial DNA had a 23 per cent chance of fixing and Y-chromosomal DNA a 10-13 per cent chance of fixing; the observed fixation of both is therefore just 2-3 per cent.
Another factor is Bateson-Dobzhansky-Muller hybrid incompatibilities: when a population splits into two daughter populations, they begin to diverge as each picks up different genetic mutations. Once multiple mutations have occurred, interbreeding between the two populations can cause problems. An example of Bateson-Dobzhansky-Muller incompatibility occurs in hybrids between the platy and the swordtail, two species of freshwater fish related to the guppy. Many (though not all) platys have spots on their dorsal fin, but the swordtail lacks these spots. In some backcrossed hybrids, larger dorsal fin spots develop, and these eventually develop into malignant melanomas. The gene responsible for the spots in platys is counterbalanced by a repressor gene that regulates the effects of the spot-producer. The corresponding genes in the swordtail do not produce these effects, and if a hybrid receives the platy spot-producer but not the repressor, it will develop the malignancy11. Bateson-Dobzhansky-Muller incompatibilities frequently impact mitochondrial-nuclear DNA interactions, creating hybrid fragility and making fixation even less likely.
Previous theories assumed the Y chromosome was prone to accumulating harmful mutations in small populations. However, Reich noted that the Y chromosome contains extensive palindromic repeats (inverted sequence blocks). Palindromic repeats undergo arm-to-arm gene conversion, a non-reciprocal recombination process where genetic information transfers between inverted sequence blocks (arms) separated by a spacer. This mechanism maintains near-identical sequence homology (often greater than 99 per cent) in non-recombining regions like the Y chromosome, purging harmful mutations.
Finally, modern humans actually have a slightly higher ratio of non-synonymous (amino-acid-changing) mutations in their lineages than Denisovans, which is the exact opposite of what would be expected if archaic humans had a high genetic load of deleterious mutations.
Having rejected the fitness gap hypothesis, Reich postulated a bicontinental range expansion of early Homo sapiens 300,000 years ago. This population was spread across North Africa, Northeast Africa, the Near East, and East Africa as far south as the Horn of Africa. These people probably resembled the hominins from Jebel Irhoud, Morocco. The largely intact Irhoud 1 cranium has primitive features, including moderately developed browridges and a long, low cranial vault, although the forehead lies within the modern range. The expansion proceeded in two directions: northwest into Europe, and south into southern Africa. The European archaics are described as ‘Neanderthal ancestors’ or simply ‘archaics’, but are best thought of as early Neanderthals. They did not yet possess the full suite of ‘classic’ Neanderthal characteristics, but as noted, contemporary Homo sapiens also retained primitive features.
When a population expands into a vast territory already occupied by a resident group, it behaves like a wave. Even if the two groups only interbreed a tiny amount at the very edge (the wavefront), the expanding group constantly moves forward into fresh, unmixed archaic territory. Over many generations, the moving wavefront (early Homo sapiens) picks up more and more local (early Neanderthal) genes. By the time the expansion reached the edges of Europe, the Homo sapiens nuclear genome had been repeatedly diluted by archaic DNA. The final population carried around 93-97 per cent Neanderthal ancestry genome-wide, effectively making them genetically indistinguishable from Neandertals.
The obvious question, then, is why were the mitochondrial and Y-chromosomal genomes not similarly flooded by their Neanderthal counterparts? Reich proposes that specific population dynamics prevented the archaic versions from entering the wavefront gene pool. At the very start of the migration, early Homo sapiens are the absolute majority within their own expanding groups. Therefore, modern mitochondrial and Y-chromosomal DNA start at 100 per cent frequency in the moving group. To maintain the status quo, the expanding group needs a mechanism to block archaic lines from entering their lineages. Reich suggests four possible mechanisms:
• Scenario 1: Haldane’s Rule & Daughter Skew: Hybrid incompatibilities between the groups could cause hybrid male babies carrying an archaic Y chromosome to be fragile or less fertile. Alternatively, hybrid incompatibilities on the X chromosome could cause hybrid fathers to overproduce daughters, skewing the offspring sex ratio to at least 60 per cent female, naturally eliminating the Neanderthal Y-chromosomal line over time.
• Scenario 2: Clan Prestige (Cultural Selection): Early human cultures often tracked lineages through either the maternal or paternal lines. If individuals who could trace their direct lineage back to the prestigious founding Homo sapiens clans had a moderate (9 to 14 per cent) reproductive advantage in terms of better access to partners, resources, or status, that edge would drive modern mitochondrial and Y-chromosomal DNS to 100 per cent fixation.
• Scenario 3: Two Alternating Waves: The expansion might have happened in two distinct pulses separated by thousands of years. One wave had a social structure where males left their own group to live with a Neanderthal group, and the other wave had a social structure where it was the females who left to join a Neanderthal group. The first wave fixed modern mitochondrial DNA, and the other fixed modern Y-chromosomal DNA.
• Scenario 4: Differential Offspring Care: A cultural scenario where hybrid communities selectively chose not to raise first-generation offspring that shared the same sex as their archaic parent, although Reich noted that this was the least anthropologically plausible of the four scenarios proposed.
Reich suggests that his range expansion not only explains ancient DNA results from Europe, but also genetic models that suggest a deep genetic substructure in present-day populations. Studies suggest that lineages leading to present-day populations diverged into two ancestral branches, which remained separate for hundreds of millennia before recombining. Estimates as to when the split occurred range from 1.7 to 1.0 million years BP, with the recombination occurring 300,000 to 125,000 years ago. However, it was not an equal recombination: there is an 80:20 ratio between the ancestries inherited from the two branches12–14.
Reich notes that the upper end of the date range coincides with his proposed expansion into Europe. He suggests this was only one half of the expansion; early Homo sapiens populations also moved south and admixed with archaic populations from which they had originally diverged 1,400,000 to 700,000 years earlier. The incoming early Homo sapiens were less closely related to these archaics than to Neanderthals, and hence were less permeable to admixture with them. Thus, the final population which became our own ancestors had a much higher proportion of ancestry from the expanding early Homo sapiens lineage.
In support of his model, Reich suggests that his bicontinental range expansion correlates with shared archaeological traits between West Eurasia and Africa that were far less present in East Asia. The timing of the genetically inferred expansion coincides with the spread of the Levallois prepared stone core technique across Africa 400,000 years BP (Middle Stone Age), the Near East, and Europe, where it was adopted by Neanderthals 480,000 – 300,000 years BP (Middle Palaeolithic). These technologies were used much less intensively, or not at all, in East Asia, where the evidence that does exist has been disputed.
By itself, the presence or absence of Levallois technology is not conclusive, as it is possible its apparent spread reflects multiple independent invention rather than a population expansion. However, Reich also cites improved hafting, controlled use of fire, use of ochre, and larger, more organised campsites. All these traits became widespread both in West Eurasia and in Africa and were much less present or entirely absent in East Asia.
Again, perhaps, a note of caution should be exercised in suggesting that East Asia was a technological backwater. The teardrop-shaped Acheulean hand axe was the signature technology of humans for one and three-quarters million years, until it was superseded by Levallois technology. In the 1940s, the American archaeologist Hallam Movius noted that while Acheulean hand axes are ubiquitous in Africa and western Eurasia, they are largely absent east of the Indian subcontinent15. What became known as the Movius Line is widely accepted, but British archaeologist Robin Dennell16,17 argues that it is not a real phenomenon and that the concept is “a house built on sand that should be forgotten”. He suggests East Asia during the Lower Palaeolithic was not an isolated backwater and hence not inherently different to the rest of Eurasia. Dennell notes that there is no unambiguous evidence that any of the Mainland Southeast Asian core and flake industries cited by Movius are from the Middle Pleistocene and hence contemporaneous with Middle Pleistocene Acheulean assemblages in South and Southwest Asia. The rarity of hand axes in China may be apparent rather than actual and may reflect an incomplete archaeological record.
With these caveats in mind, the technologically driven expansion of a population of early Homo sapiens is compelling. We now await the release of a peer-reviewed version of Reich’s preprint and the lively debate it is certain to trigger.
References:
1. Reich, D. Hypothesis: A modern human range expansion ~300,000 years ago had a critical role in Neandertal origins. Preprint at https://www.biorxiv.org/content/10.64898/2026.03.11.711219v2 (2026).
2. Arsuaga, J. et al. Neandertal roots: Cranial and chronological evidence from Sima de los Huesos. Science 344, 1358–1363 (2014).
3. Meyer, M. et al. Nuclear DNA sequences from the Middle Pleistocene Sima de los Huesos hominins. Nature 531, 504–507 (2016).
4. Prüfer, K. et al. The complete genome sequence of a Neanderthal from the Altai Mountains. Nature 505, 43–49 (2014).
5. Meyer, M. et al. A mitochondrial genome sequence of a hominin from Sima de los Huesos. Nature 505, 403–406 (2014).
6. Petr, M. et al. The evolutionary history of Neanderthal and Denisovan Y chromosomes. Science 369, 1653–1656 (2020).
7. Posth, C. et al. Deeply divergent archaic mitochondrial genome provides lower time boundary for African gene flow into Neanderthals. Nat. Commun. 8, 16046 (2017).
8. Kuhlwilm, M. et al. Ancient gene flow from early modern humans into Eastern Neanderthals. Nature 520, 429–433 (2016).
9. Hubisz, M., Williams, A. & Siepel, A. Mapping gene flow between ancient hominins through demography-aware inference of the ancestral recombination graph. PLOS Genet. 16, e1008895 (2020).
10. Harris, D. N. et al. Diverse African genomes reveal selection on ancient modern human introgressions in Neanderthals. Curr. Biol. 33, 4905–4916 (2023).
11. Coyne, J. & Orr, H. Speciation. (Sinauer Associates, Sunderland, MA, 2004).
12. Fan, S. et al. Whole-genome sequencing reveals a complex African population demographic history and signatures of local adaptation. Cell 186, 923-939.e14 (2023).
13. Ragsdale, A. et al. A weakly structured stem for human origins in Africa. Nature 617, 755–763 (2023).
14. Cousins, T., Scally, A. & Durbin, R. A structured coalescent model reveals deep ancestral structure shared by all modern humans. Nat. Genet. 57, 856–864 (2025).
15. Movius, H. The Lower Palaeolithic Cultures of Southern and Eastern Asia. Trans. Am. Philos. Soc. 38, 329–420 (1948).
16. Dennell, R. Hallum Movius, Helmut de Terra and the Line than Never Was: Burma, 1038. in Living in the Landscape: Essays in honour of Graeme Barker (eds Boyle, K., Rabett, R. & Hunt, C.) 11–34 (McDonald Institute for Archaeological Research, Cambridge, 2014).
17. Dennell, R. Life without the Movius Line: The structure of the East and Southeast Asian Early Palaeolithic. Peking Man Relat. Stud. 400, 14–22 (2016).