Excavations of Homo Erectus remains Ngandong Java

Excavations of Homo Erectus remains in 2010 in Ngandong, Java
An early human species may have survived in Java, Indonesia, until 108,000 years ago. Homo erectus endured so long that individuals may have interbred with more recent hominins, like the mysterious Denisovans.

It has long been unclear when H. erectus died out. Now a reanalysis of the youngest known remains may have pinned down the extinction date.

H. erectus was one of the first species in our genus, Homo. It evolved in Africa about 2 million years ago, then spread across Europe and Asia. Compared with earlier species, H. erectus had relatively large brains and used tools skilfully - although it was surpassed in both respects by later groups like Neanderthals and modern humans. H. erectus may be our direct ancestor.

H. erectus died out before modern humans reached Java, so it is unlikely the two species crossed paths. That means our species isn't in the frame for its extinction.

Last of their kind

The last known H. erectus lived near the Solo River in Ngandong, Java. The remains were discovered in the 1930s but scientists have struggled ever since to date them.

Previous studies had implied that H. erectus overlapped with our species, but most now agree that this was not the case.

In 2011 a team led by Susan Antón of New York University concluded that the last H. erectus lived between 143,000 and 546,000 years ago - significantly older. Though there was a huge margin of uncertainty it was still before modern humans arrived in the area.

A team led by Kira Westaway of Macquarie University in Sydney, Australia, has now narrowed down the age. The key was that the fossils were found in one of a series of terraces formed by the river. The higher terraces are older and lower ones are younger, so by dating each of the terraces they could identify a window of possible ages for the remains.

"You're effectively constraining the age of Ngandong with an upper and lower maximum age range," says Westaway. "That's what we tried to do, to pin Ngandong down to where it sits in the landscape."

Along with other dating methods, this allowed the team to constrain the ages of the H. erectus to between 117,000 and 108,000 years ago. Westaway calls this "a convincing timeline for the Ngandong evidence".

"This is a hugely important site and it's really great to have it revisited and in a very thorough way," says Antón. However, she says the "super-precise" date needs to be treated with caution, because each of the dating methods used requires a lot of modelling to interpret the raw data.

A fateful meeting

Currently there is no evidence that modern humans were in Java that early. "That might change in a few years if somebody finds a much older modern human site," says Westaway. Still, for now it seems the two didn't meet. That means they cannot have interbred, and there is no reason to think humans were responsible for the extinction of H. erectus.

"But the age does open up the opportunity that there could have been potential overlap with the Denisovans," says Westaway. The Denisovans are known from a handful of remains, which have yielded DNA. They roamed Asia and interbred with the ancestors of people in China and South-East Asia.

"There's a good chance they came down into South-East Asia and could have interacted with H. erectus," says Westaway. Conceivably the Denisovans outcompeted H. erectus, but there is no evidence of this.

Interbreeding is also a possibility. "I think it's almost unavoidable, given the timeframe and given the geographic position," says Antón.

Journal reference: Nature, DOI: 10.1038/s41586-019-1863-2