Our Human Ancestors Very Nearly Went Extinct 900,000 Years Ago, Genetics Suggest

A study proposes that the population that gave rise to modern humans may have been reduced to roughly 1,300 reproducing individuals

Equations for Human Bottleneck Study
The core formula of a new study in Science that suggests our ancestors may have survived a bottleneck 900,000 years ago Shanghai Institute of Nutrition and Health of the Chinese Academy of Sciences

For 75 years the Doomsday Clock has been a powerful metaphor, representing how close humanity is to extinction from threats like nuclear weapons. Currently set at 90 seconds to midnight, the clock suggests we are closer to catastrophe today than at any time since its 1947 debut.

But had the doomsday warning mechanism been around 900,000 years ago, it might have estimated only a single second remained before midnight signaled the end of humanity. An intriguing new genetic study suggests that our ancestors dwindled to just 1,280 breeding individuals during that era, and they nearly vanished from the earth long before modern humans even appeared on the scene.

The study, published Thursday in Science, analyzed the genetic lineages of 3,154 modern humans to trace their characteristics backward in time and model the population patterns likeliest to have produced their existing genomes. Wangjie Hu, of the Chinese Academy of Sciences, and colleagues suggest that between 813,000 and 930,000 years ago the population of ancient humans that would eventually give rise to our own species, Homo sapiens, experienced what geneticists call a “bottleneck.” For unknown reasons, perhaps difficult environmental conditions, their numbers plunged dramatically to a point where our lineage was within a whisper of total extinction. Based on the study’s estimates, some 98.7 percent of our human ancestors were wiped out.

“The estimated population size for our ancestral lineage is tiny, and the bottleneck lasted for a remarkably long time, if accurately modeled,” says paleoanthropologist Chris Stringer, of the Natural History Museum in London, who wasn’t part of the study. “If this bottleneck did occur, it would certainly have brought our ancestors very close to extinction.”

The geneticists suggest that the bottleneck may have led to increased inbreeding and a subsequent loss in human genetic diversity that has persisted to this day. They also theorize that the bottleneck may have given rise to a notable new hominin species. The timeline of the bottleneck matches some existing genetic estimates pegging that same time period as when a new hominin appeared that may be the last common ancestor of the later Pleistocene’s three big-brained species: Neanderthals, Denisovans and ourselves.

University of Wisconsin-Madison population geneticist Aaron Ragsdale, who wasn’t involved in the research, says the study raises some very intriguing questions about human evolution during a time period from which both genetic and fossil data are relatively scarce. “I am eager to see if their results are replicated using other methods,” Ragsdale says.

Population fluctuations, even those hundreds of thousands of years ago, leave signatures that can be identified in modern humans’ genomic sequences. To analyze them, a team of researchers led by Chinese geneticists developed a new tool called FitCoal. The researchers used the tool on more than 3,000 living individuals from 10 African populations and 40 non-African populations. FitCoal computations traced the populations’ many genetic mutations and their probabilities of occurring backward in time to arrive at estimates of population sizes that existed at various moments in evolutionary history.

“Our findings indicate that the severe bottleneck brought the ancestral human population close to extinction and completely reshaped present-day human genetic diversity,” the authors write in the study.

Ragsdale is not convinced that the situation was so dire. “I think it’s a bit of a stretch to conclude from these results that human ancestral populations were on the verge of extinction,” he says. Actual population sizes are often much larger than the “breeding” sizes methods like FitCoal provide, he notes. The suggestion of a small population for so many generations may be skewed by much shorter periods of reduced numbers, he adds.

What might have caused the population plunge? The answers aren’t found in the genetic data, but scientists do know that era saw a dramatic shift in the earth’s environments. The middle Pleistocene transition was a time of significant climate change—including a sharp cooling across the globe about 900,000 years ago that saw growing glaciers, chillier seas, extended droughts and stronger monsoons. Wildlife species of Africa and Eurasia underwent significant changes during this period.

Much research has shown how changing climate and environments have also helped drive major shifts in human evolution. But it does not appear that the era produced a global population plunge among the planet’s various hominin species that were not our direct ancestors.

Nick Ashton, a Paleolithic archaeologist at the British Museum not involved with the study, notes that numerous archaeological sites in Africa and Eurasia date to the period of the proposed bottleneck. That suggests that any population crash impacted only a limited group, perhaps in Africa, who may have been the ancestors of modern humans. “Looking for potential causes of the proposed bottleneck would be fruitful, whether this be regional drought, volcanic activity or other environmental factors,” he says.

Amazingly, the study suggests that our ancestors managed to survive in precariously small numbers for an extremely long time—an estimated 120,000 years. But when conditions again became conducive to human habitation, whether through beneficial climate shifts or, as the authors theorize, technological advances like human control of fire, our ancestors bounced back swiftly. By around 813,000 years ago, all ten African populations in the study appear to have increased by a factor of 20 times.

The Natural History Museum’s Stringer notes that, like other methods of reconstructing past populations, FitCoal relies on some assumptions and simplifications of factors like mutation rates. Since the authors have made FitCoal available to scholars, he adds, its accuracy will be further tested, and researchers may use it to investigate populations through other genomes like those of Neanderthals and Denisovans.

The study’s authors suggest this long period during which our ancestors survived in such small numbers may have led to the evolution of an entirely new species that may be the last common ancestor of modern humans and our close relatives the Neanderthals and Denisovans. Ashton says this idea is a good fit for some genetic evidence suggesting that the common ancestor of these species lived as recently as 500,000 to 700,000 years ago. “But there are other possible interpretations,” he notes.

For example, fossil studies, tracking morphological changes like skull and tooth shape, have suggested that the lineages had already diverged before any bottleneck, which means Neanderthals and Denisovans would have avoided its impacts.

The estimated bottleneck period is too old to yield any ancient DNA—at least with current methods. The oldest hominin DNA yet recovered is only 400,000 years old. In Africa, the cradle of humanity, climate makes preservation of ancient DNA particularly poor.

Facts on the ground, like stone tools and bones, are also hard to come by, but numerous sites do exist from this key period in time. Further examinations of skulls and bones found at various sites in eastern and southern Africa, China, Indonesia and Spain could help scientists tease out whether the evolutionary changes seen in our bones match those suggested by the genetic models. “The genetic theory needs to be thoroughly tested against the archaeological and human fossil evidence,” says Ashton. “This is best achieved through more refined dating of current and new sites that potentially date to the proposed bottleneck.”

Get the latest Science stories in your inbox.