In 1990, a farmer in central China stumbled on a skull so battered it looked more like a pancake than a fossil. For decades, scientists squinted at the flattened bones of “Yunxian 2,” guessing it belonged to Homo erectus — the workhorse of human ancestry.
Now, with digital forensics reconstruction tools that would make a CSI tech jealous, researchers have ‘uncrushed’ Yunxian 2. What they found could blow up our timeline of human evolution.
Dragon Man’s ancestors
The team rescanned Yunxian 2 in high resolution to digitally unflatten the cranium. They used CT segmentation to separate bone from sediment, then reassembled displaced fragments — leaning on pieces from the more damaged Yunxian 1 where needed.
They didn’t stop at a 3D model. The authors placed 533 anatomical landmarks on Yunxian 2 and on a large comparative sample. Then they ran morphometric and phylogenetic analyses, including parsimony and Bayesian tip-dating, to test where Yunxian 2 best fits among known fossils.
The skull reconstruction points to an Asian lineage that includes Homo longi and the mysterious Denisovans. It essentially suggests that modern humans, Neanderthals, and Denisovans started going their separate ways far earlier than we thought — more than a million years ago.
“This changes a lot of thinking because it suggests that by 1m years ago our ancestors had already split into distinct groups, pointing to a much earlier and more complex human evolutionary split than previously believed. It more or less doubles the time of origin of Homo sapiens,” Chris Stringer of London’s Natural History Museum, a coauthor on the new study, told The Guardian.
The fossil’s features are a mash-up: a long, low braincase with heavy brow ridges, but also a surprisingly big brain for its age — about 1,143 cubic centimeters, close to early modern humans. The face is broad with flat cheekbones and a high nasal bridge (but without the midfacial prominence we find in Neanderthals), traits that link it to Homo longi, also known as “Dragon Man.”
In other words, Yunxian 2 looks a bit like the weird cousin who shows up at a family reunion with the unmistakable cheekbones of your aunt but the brow of your grandfather.
And this is where it gets even stranger: Denisovans — the mysterious humans known mostly from their DNA and a few fragmentary fossils (a finger bone, a molar, and later, the Xiahe jaw from the Tibetan Plateau) — fall inside the Homo longi family tree. That means, in this reconstruction, they’re closer to us than Neanderthals were.
The Muddle in the Middle
If this holds up, it rewrites what paleoanthropologists sometimes call the “muddle in the middle,” the confusing and chaotic record among human fossils from 1 million to 300,000 years ago, which lacks clear classification.
The new analysis suggests that Neanderthals split off around 1.38 million years ago, while the ancestors of modern humans and Denisovans stayed together until about 1.32 million years ago. Compare that to the standard genetics-based model, which puts the split closer to 600,000 years ago.
Essentially, the new tree reveals three big evolutionary jumps:
- Neanderthals split off around 1.38 million years ago.
- The longi and sapiens clades diverged around 1.32 million years ago.
- The sapiens clade itself began around 1.02 million years ago.
“This fossil is the closest we’ve got to the ancestor of all those groups,” said Stringer.
The conventional DNA-based picture says an ancestral population split into (1) modern humans and (2) a branch that later divided into Neanderthals and Denisovans. The new tree instead peels Neanderthals off first, with modern humans and Denisovans diverging later — over a million years ago.
That raises an unsettling possibility: maybe the birthplace of our lineage wasn’t in Africa, where the oldest clearly modern human fossils date to about 300,000 years ago. Instead, the ancestor of Homo sapiens could have lived in western Asia, before later populations moved back into Africa to evolve into us. “It looks more likely that the ancestor was outside of Africa, perhaps in Western Asia,” Stringer told New Scientist.
Still More Questions
Not everyone is convinced. Geneticist Aylwyn Scally warned that human history looks less like a tree and more like a messy “tangled network.” He added, “Genetics is a better guide to such relationships than morphology — especially when you have only partial skeletons.”
Others see promise but want more evidence. “If confirmed by additional fossils and genetic evidence, the divergence dating would be surprising indeed,” said Frido Welker of the University of Copenhagen. Paleoanthropologist Sheela Athreya was more blunt: “Middle Pleistocene evolution represents an enduring mystery.”
Yunxian 2 may not give us the final word on where humans came from. But it does give us a powerful reminder that our past isn’t a straight line — it’s a kaleidoscope of ancient populations, splintering and reconnecting across continents.
If Yunxian 2 sits near the origins of both the Homo longi/Denisovan and Homo sapiens lineages, “it may represent one of the most important windows yet into the evolutionary processes that shaped our genus around one million years ago,” according to Stringer.
The new findings appeared in the journal Science.
