
Around 41,000 years ago, the very last Neanderthal took their final breath. At that moment, we聽became the only remaining hominins, the sole survivors of the once diverse family of bipedal apes.
We will never know exactly when or where this momentous event took place, but we do know the Neanderthals died out suspiciously close to the time when modern humans arrived in their territory. Exactly why they vanished has long been hotly debated, but astonishing revelations from the genomes of the last Neanderthals and hidden in a remarkable cave in France are now painting a detailed picture of these first encounters 鈥 and what might have happened next.
Advertisement
鈥淭his is a major turning point in our understanding of Neanderthals and their extinction process,鈥 says at the Centre for Anthropobiology and Genomics of Toulouse, France.
Our species, Homo sapiens, and Neanderthals share a common ancestor, but Neanderthals split from our lineage at least 400,000 years ago, evolving in Eurasia, from the Mediterranean to Siberia. Our species is younger, first appearing in Africa some 300,000 years ago and evolving into hominins that were anatomically much like us by at least 195,000 years ago. Modern humans left the continent in waves from around 170,000 years ago, and were thought to have reached western Europe roughly 43,000 years ago, when 鈥 according to the well-established view 鈥 they replaced the Neanderthals within just a few thousand years.
There are clear differences in the appearance of the two species. Neanderthals were shorter and stockier than us, with wider ribcages and pelvises, and their faces had pronounced brow ridges. Once regarded as brutish, unintelligent thugs, their reputation has been completely overhauled in recent decades, as evidence emerged that Neanderthals were skilled tool-makers who buried their dead, may have drawn on cave walls and invented complex substances such as birch tar, which was used as a glue, among other things. According to this new view, they weren鈥檛 all that different from us.
So why did they die out while modern humans conquered the world? Numerous explanations have been put forward, including climate chaos unleashed by the flipping of Earth鈥檚 magnetic field or . The success of H. sapiens, however, complicates any hypothesis involving such epic catastrophes 鈥 because why would modern humans have been unaffected?

Another popular idea is that Neanderthals were actually killed off by the incoming modern humans, whether deliberately or accidentally through the introduction of diseases, or that they were unable to compete with modern humans for resources. Still others argue that the Neanderthals didn鈥檛 really die out, but were instead assimilated into our lineage through interbreeding. It is certainly true that 1 to 4 per cent of the genomes of people today with non-African ancestry are derived from Neanderthals.
What would help clear things up is hard evidence of what happened when the two species encountered each other. And, unbelievably, archaeologists have found it 鈥 in a cave in France.
A surprise inside
Grotte Mandrin is a hillside rock shelter that overlooks the Rhone river valley in south-east France, long an important wildlife corridor linking the Mediterranean to northern Europe. For more than 80,000 years, this place has been a valuable location to hominins, commanding excellent views of herds of animals (although today it overlooks a busy motorway and railway line). Excavations in the cave began in 1990 鈥 and what Slimak and his colleagues have found since is rewriting what we know about when modern humans arrived in Europe and their interactions with the resident Neanderthals. 鈥淭he site of Grotte Mandrin keeps delivering surprises,鈥 says at the Natural History Museum in London.
From the stone tools and other artefacts left in the layers of sediment on the cave floor, we know that Neanderthals began using this shelter about 80,000 years ago and continued to visit periodically until the time of their extinction. But the archaeological excavations uncovered some unexpected things in the dirt.
The first surprise came in the 1990s, when 1500 tiny, triangular stone points resembling arrowheads were unearthed in a layer of sediment in the cave floor dating from 52,000 to 57,000 years ago. These were unlike any Neanderthal artefacts discovered in other layers in the cave 鈥 but they were almost identical to artefacts left by modern humans in roughly the same time period at a site called Ksar Akil in Lebanon, nearly 3000 kilometres away. Then, in 2018, a tooth belonging to a H. sapiens child was found in the same strata, sandwiched between layers of evidence of Neanderthal habitation.
Together, these findings demonstrated that modern humans had reached Mediterranean France around 54,000 years ago, than had been thought. This was 鈥渕assive news because it placed our species in western Europe well before the 鈥榬ed line鈥 of around 40 to 42,000 years ago which data up to then seemed to suggest was the upper limit for their presence鈥, says at the University of Cambridge.
Grotte Mandrin was first visited by Neanderthals, then modern humans for a period of one, possibly two generations. A millennium or two later, the Neanderthals returned, only to be again replaced by modern humans. This kind of occupation sequence hasn鈥檛 been found elsewhere and it complicates the idea of a simple population replacement to explain the Neanderthals鈥 disappearance. 鈥淢andrin is unique in that it鈥檚 the only site really where we have any evidence that the Neanderthals return,鈥 says Wragg Sykes.
Soot 鈥榖arcodes鈥
Equally astonishing is that Grotte Mandrin also offers a fine-scale view of these comings and goings, thanks to a technique pioneered by a member of Slimak鈥檚 team, S茅gol猫ne Vandevelde at the University of Paris, Panth茅on-Sorbonne. When cave visitors lit fires, . During periods when the cave was unoccupied, a layer of calcium carbonate would have formed over the soot, in as little as two months. This process happened repeatedly, resulting in alternating black-and-white crusts on the cave surfaces that, under the microscope, resemble a barcode, each stripe representing a different visitation.
Fragments of this crust broke off and became buried in the sediment layers on the cave floor. With the soot fragments as evidence of visitation, the sedimentary layer could then provide dating. Artefacts found in the layers completed the picture, allowing archaeologists to determine who was visiting the shelter when, over a period of 80,000 years.
Together, the evidence suggested that there was a maximum of a year between the first modern human fire in the cave, some 54,000 years ago, and the preceding fire lit by Neanderthals 鈥 providing conclusive proof that these two hominins were in the same area at the same time.

鈥淭he two groups鈥 share this territory,鈥 says Slimak. 鈥淚n that moment, the question is, what was the interaction with the local Neanderthals?鈥 Slimak thinks that the two species cooperated, pointing to evidence that the incoming modern humans used the same flint for their tools as the local Neanderthals, some from sites as far as 95 kilometres away. 鈥淚t鈥檚 likely that there must have been good relations, with shared territories and shared knowledge,鈥 he says.
What happened next is complex. After the first band of modern humans departed, 54,000 years ago, it took 1000 to 2000 years for Neanderthals to return to Grotte Mandrin. This group used the cave for the following 12,000 years, but when the next group of modern humans moved in, all traces of the Neanderthals disappear 鈥 not just from this cave, but from all over Europe, with the possible exception of some areas in the far north and south. So why did this encounter coincide with the disappearance of the Neanderthals, when they were able to survive for 12,000 years after their first meeting with H. sapiens?
Humans don't die in 'bim-bam-booms', they die in whispers
Here, too, Grotte Mandrin is offering some answers. In 2015, Slimak cleared away some dead leaves from outside the cave and spotted some teeth in the ground. They turned out to belong to a Neanderthal. 鈥淭hat was absolutely surprising,鈥 he says.
A painstaking excavation began, removing the sediment grain by grain with tweezers to preserve every bit of information. Thus far, the team has found 31 teeth, along with fragments of a skull, hand and other bones, all from the same individual.
Slimak called the Neanderthal fossil Thorin, after the Dwarf king of J. R. R. Tolkien鈥檚 The Hobbit. He sent a fragment from one of Thorin鈥檚 teeth for DNA analysis, but wasn鈥檛 optimistic: DNA isn鈥檛 typically preserved in warm climates and previous attempts to obtain genetic material from the site were unsuccessful. This time, however, the researchers struck gold.

The revealed that Thorin was male; analysis of the isotopes in his bones and the sediment surrounding him indicated that he had lived sometime between 42,000 and 45,000 years ago 鈥 making him 鈥渙ne of the very last Neanderthals鈥, says Slimak. But though dating incontrovertibly made Thorin a late Neanderthal, his genome didn鈥檛 match those of any of his species who lived at the same time. 鈥淭horin was completely divergent,鈥 says Slimak.
In fact, Thorin鈥檚 genome appeared similar to those of far more ancient Neanderthals, from around 105,000 years ago. Researchers struggled to make sense of this puzzle for years, before realising that Thorin was descended from a previously unknown lineage, a remnant of an ancient population that had diverged from the main Neanderthal population some 105,000 years ago and remained in extreme isolation, living in small groups, for more than 50,000 years.
鈥淭here is no integration between 鈥榗lassic鈥 Neanderthals and Thorin鈥檚 population,鈥 says Slimak. 鈥淚t鈥檚 no longer just a question of Neanderthals and Homo sapiens, it鈥檚 a question of different populations of Neanderthals, very distinct, very isolated.鈥
Dying in isolation
Analysis of Thorin, coupled with revisiting the genome of another late Neanderthal who lived around 43,000 years ago in what is now central France, indicated to Slimak and his team that there was at least one, or more likely two, previously unknown groups of isolated Neanderthals that persisted right up to the end of the species.
These insights into Thorin鈥檚 population also fit with discoveries about the social structure of other Neanderthal groups. that analysed the genomes of 13 Neanderthals who lived between 44,000 and 59,000 years ago from two sites in the Altai mountains in Siberia found high levels of homozygosity 鈥 inheriting identical genes from both parents, or inbreeding 鈥 indicating that they probably lived and reproduced in small groups of about 20 individuals at a time and had little interaction with outsiders. 鈥淥verall, this fits with the wider picture that late Neanderthals were, in general, not very interconnected with each other,鈥 says Wragg Sykes.
But why? Slimak suspects that the first incursion by modern humans, though short, had a destabilising effect on Neanderthal populations, noting that the Neanderthals at Grotte Mandrin initially shared their territory and sources of materials with the modern humans 鈥 but then stopped. 鈥淎fter the incursion of Homo sapiens into that territory, this population begins to live in very fragmented territories, with no exchanges with other populations,鈥 he says.
The lack of interchange between Neanderthal groups across Europe had implications for the survival of the species, not least because the associated inbreeding could have degraded the genetic fitness of the population. This isolation would also mean that Neanderthals didn鈥檛 benefit from sharing ideas. At Grotte Mandrin, the effect of this social isolation is reflected in the artefacts found at the site. The few stone weapons lifted from the sediment layer of Thorin鈥檚 lineage are large and rather clumsy triangular shapes, quite unlike the artefacts found at other Neanderthal sites from the same time.
They stand in even starker contrast to the tiny, arrowhead-like objects seen in the modern human layers from before and after Thorin鈥檚 time, which, if used as projectiles, would have enabled the much safer hunting strategy of killing animals at a distance. This sophisticated weaponry seems to have arisen through modern humans sharing information via a vast social network 鈥 after all, arrowheads of precisely the same design were found in Lebanon. 鈥淭hat means that there are鈥 very important networks that are used by this population to keep contact,鈥 says Slimak. 鈥淭he social network is very strong and very large.鈥 He argues that this standardisation of tool-making techniques made modern humans more efficient; their vast social networks allowed them to spread their best practices, giving H. sapiens an advantage over Neanderthals operating in isolation. 鈥淭his efficiency acts like a wave, and all the other past humans have been taken by this wave,鈥 says Slimak.
A one-sided genetic exchange
Yet despite the deep isolation of some late Neanderthal groups, they still sometimes interbred with modern humans. What is curious, however, is that though Neanderthal DNA appears in H. sapiens fossils, there is no sign of modern human DNA in any late-surviving Neanderthals sequenced so far. 鈥淲e know that these groups overlap for thousands of years in Europe, yet we find this largely one-directional gene flow from the Neanderthals to modern humans and not the other way around,鈥 says at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany.
There are many possible explanations for this one-sided genetic exchange. It could simply be due to sparsity of data, as relatively few Neanderthal genomes have been sequenced. Alternatively, Neanderthal DNA may have been more important for the survival of modern humans than the other way round 鈥 for example, bolstering immune system function 鈥 so was preserved in later generations by natural selection.
Or, says Hajdinjak, 鈥渋t could be social practices鈥 that resulted in babies born through interbreeding being kept in modern human groups but not Neanderthal ones. Slimak suspects that Neanderthal women were joining the modern human groups, but the reverse didn鈥檛 happen.
鈥淭heir already small populations were losing reproductive age individuals to the other species, without any replenishment in return 鈥 a recipe for demographic disaster,鈥 says Stringer. Whatever the reasons for the genetic imbalance 鈥 social, biological or both 鈥 it contributed to the demise of the last Neanderthals, he says.
It all suggests that the end of the Neanderthals was a long, messy process. Looking for a single catastrophic explanation for Neanderthal extinction, such as a volcanic explosion, is wrong, says Slimak. 鈥淗umans don鈥檛 die in 鈥榖im-bam-booms鈥欌 they die in whispers.鈥