Three years ago, a Chinese farmer made an unusual donation to a university museum—a giant, nearly intact human skull with strange proportions and an unusual backstory. The man’s family had been hiding the fossil since it was unearthed at a construction site in Harbin nearly 90 years ago.

After geochemical detective work to locate where the fossil was likely found, and painstaking comparison of its distinctive features with those of other early humans, some of the scientists investigating the find believe the cranium from Harbin could represent an entirely new human species—Homo longi or “Dragon Man.” If so, they further suggest it might even be the human lineage most closely related to ourselves.

“The discovery of the Harbin cranium and our analyses suggest that there is a third lineage of archaic human [that] once lived in Asia, and this lineage has [a] closer relationship with H. sapiens than the Neanderthals,” says Xijun Ni, a paleoanthropologist at the Chinese Academy of Sciences and Hebei GEO University. If so, that would make the strange skull a close relative indeed since most humans today still have significant amounts of Neanderthal DNA from repeated interbreeding between our species.

Claims of a new human species are sure to cause skepticism and spark debate. But it seems that wherever the 146,000-year-old fossil falls on the human family tree, it will add to growing evidence that a fascinating and diverse period of evolution was occurring in China from about 100,000 to 500,000 years ago.

And because excavations in China haven’t been as extensive as those in places like Africa, experts are only beginning to uncover the evidence.

Like its origins, the skull’s 20th-century story isn’t entirely clear. The family that donated the skull to co-author Ji Qiang, at Hebei GEO University’s museum, had been hiding it in a well for three generations. It was unearthed in the 1930s when a railway bridge was built along the Songhua River and the family, suspecting that it was important but unsure what to do with the fossil, had safeguarded the skull ever since.

Extensive analyses of the skull began soon after it reached the museum in 2018 and resulted in three separate studies, all including Ni, that appear this week in the open-access journal The Innovation.

Direct uranium-series dating suggests the skull is at least 146,000 years old, but a lot more work was needed to attempt to put the isolated fossil into context after 90 years.

The team used X-ray fluorescence to compare the skull’s chemical composition with those of other Middle Pleistocene mammal fossils discovered in the Harbin riverside area, and found them strikingly similar. An analysis of rare-earth elements, from small pieces of bone in the skull’s nasal cavity also matched those of human and mammal remains from the Harbin locale found in sediments dated to 138,000 to 309,000 years ago.

A very close inspection even found sediments stuck inside the skull’s nasal cavity, and their strontium isotope ratios proved a reasonable match for those found in a core that was drilled near the bridge where the skull was said to have been discovered.

Observing the skull’s unusual size was a far simpler matter; it’s the largest of all known Homo skulls. The big cranium was able to house a brain similar in size to our own. But other features are more archaic. The skull has a thick brow, big—almost square—eye sockets and a wide mouth to hold oversized teeth. This intriguing mix of human characteristics presents a mosaic that the authors define as distinct from other Homo species—from the more primitive Homo heidelbergensis and Homo erectus to more modern humans like ourselves.

Ni says the team compared 600 different morphological characteristics of the skull across a selection of some 95 varied human skulls and mandibles. They used a set of mathematical techniques on all this data to create branching diagrams that sketch out the phylogenic relations of the different Homo species.

That analysis suggested that there were three main lineages of later Pleistocene humans, each descended from a common ancestor: H. sapiens, H. neanderthalensis and a group containing Harbin and a handful of other Chinese fossils that have proved difficult to classify including those from Dali, Jinniushan and Hualongdong.

“Our results suggest that the Harbin cranium, or Homo longi, represents a lineage that is the sister group of the H. sapiens lineage. So we say H. longi is phylogenetically closer to H. sapiens than Neanderthals are.”

“Whether or not this skull is a valid species is certainly up for debate,” says Michael Petraglia at the Max Planck Institute for the Science of Human History, and the Smithsonian Institution’s Human Origins Initiative.

“It’s exciting because it is a really interesting cranium, and it does have some things to say about human evolution and what’s going on in Asia. But it’s also disappointing that it’s 90 years out from discovery, and it is just an isolated cranium, and you’re not quite sure exactly how old it is or where it fits,” says Petraglia, who was not involved with the study. “The scientists do the best they can, but there’s a lot of uncertainty and missing information. So I expect a lot of reaction and controversy to this cranium.”

Chris Stringer, a study co-author from the Natural History Museum, London, doesn’t necessarily agree with some of his colleagues that the skull should be classified as a distinct species. Stringer stresses the importance of genetics in establishing where species branch off from one another. He currently favors a view that the Harbin fossil and the Dali skull, a nearly complete 250,000-year-old specimen found in China’s Shaanxi province which also displays an interesting mix of features, might be grouped as a different species dubbed H. daliensis. But Stringer was also enthusiastic about what can still be learned from the Harbin skull, noting that it “should also help to flesh out our knowledge of the mysterious Denisovans, and that will form part of the next stage of research.”

The Denisovans, ancient humans who shared an ancestor with Neanderthals and ourselves, left behind evidence of their intimate relations with us in the DNA of modern peoples in Asia and Oceania. So far, however, little physical evidence of them has turned up, only three teeth and two small bone fragments from a Siberian cave.

Katerina Harvati is a paleoanthropologist at the University of Tübingen not associated with the study. Among her research subjects is the controversial skull from Apidima, Greece, that may or may not represent the oldest modern human ever found outside of Africa.

Harvati found the Harbin skull an intriguing mix of features previously associated with other lineages. “Middle Pleistocene human evolution is known to be extremely complex—famously called the ‘muddle in the middle,’” she says. “And it has been clear for some time that the Asian human fossil record may hold the key to understanding it.”

The studies of the Harbin skull, she notes, add some clarity to the picture thanks to extensive comparisons of morphological and phylogenetic analysis.

“The Harbin cranium is somewhat similar to other Asian fossils like Huanglongdong and Dali in showing unexpected combinations of features, including some previously associated with H. sapiens. The authors also identify similarities between Harbin and the (very few) known ‘Denisovan’ fossils. I think that these studies help bring the evidence together and point to a distinct lineage of Asian Middle Pleistocene hominins closely related to our own lineage as well as that of Neanderthals.”

The Dragon Man appears to be a 50-something male who was likely a very large and powerful individual. The authors suggest his small hunter-gatherer community settled on a forested floodplain in a Middle Pleistocene environment that could be harsh and quite cold. The fossil is the northernmost known from the Middle Pleistocene, which may have meant that large size and a burly build were necessary adaptations.

Petraglia agreed that populations living in the region were likely pretty small and probably isolated. “Maybe that’s what’s creating this diversity in this group of hominins,” he says, noting that Pleistocene humans are known from the rainforests of southern China to the frigid north. “They were cognitively advanced enough, or culturally innovative enough, that they could live in these extreme environments from rainforests to cold northern climates,” he says.

That theory fits with an evolutionary picture in which smaller populations evolve in isolation, intermittently expand over time and mix with others and then separate again into smaller groups that continue to adapt to their localized environments before again meeting and breeding with other groups.

The Harbin skull’s recent emergence, after thousands of years buried on a riverside and nearly a century hidden down a well, adds another intriguing piece to China’s Middle Pleistocene puzzle. It joins a number of other enigmatic fossils from populations that have resisted any easy identification, thought to have lived in transition between H. Erectus and H. sapiens.

“How do they fit in terms of their evolutionary relationships, to what degree are they interbreeding with the populations across Eurasia, and to what degree do they become isolated resulting in their distinctive features?” Petraglia asks. “This brings up a lot of interesting questions and in human evolution China is still really a great unknown.”