Russell L. Ciochon, who heads the American team at the Ngandong site, with a gorilla skull. Credit: Tom Jorgensen

In the early 1930s, 14 Homo erectus fossils and 25,000 vertebrate remains were unearthed near the muddy Solo River at Ngandong in Java, Indonesia, by a research team from the Netherlands. Some 80 years later, this remains one of the world's largest caches of this early human. It could also be evidence of the species' swansong. A team has now returned to Ngandong, armed with the original Dutch survey documents, to answer some long-standing questions about the age of the fossils and the ancient sediment that they were buried in. Nature talked to Russell L. Ciochon, a palaeoanthropologist at the University of Iowa in Iowa City, while his team was digging at Ngandong.

Why is this fieldwork so important? H. erectus from Ngandong potentially lived in the last part of the Ice Age at the same time that Homo sapiens inhabited other parts of the Old World, and Homo floresiensis (the 'hobbit') was still living in caves on the Indonesian island of Flores. This was a very intriguing period in the saga of human evolution.

According to the Dutch team who discovered them, the H. erectus specimens were deposited by the Solo River. The fact that the present river is so near to the deposits that contained the H. erectus specimens could indicate that the deposits and fossils are far younger than the oldest known H. sapiens in Africa, which would mean that the two species actually coexisted. But attempts to date Ngandong over the past 30 years have proved inconclusive. This is partly because we don't know enough about the Ngandong geology and can't be sure that the dated samples from earlier excavations came from the discovery bed. Our team is the first to focus on the geological context of the fossils1. This was possible because geoarchaeologist Frank Huffman from the University of Texas at Austin obtained the long-forgotten survey documents from the 1930s2.

Indonesian team leader Yahdi Zaim (left) and Rob Scott excavate the bone bed. Credit: O. Frank Huffman

What have you found at the site?

During our expedition, we have recovered more than 800 fossils from a bone bed — a geological deposit with a dense collection of bones. The excavations have provided our geological team, which includes Yan Rizal from the Institute of Technology Bandung and University of Iowa geoarchaeologist Art Bettis, with details on the site's sediments that shed new light on how the bone bed was created. We believe that the detailed analysis of the site's geology and the circumstances of burial of the bone bed will provide the crucial information to evaluate the dating and other contentious issues surrounding the remains.

We found no stone artefacts at the site, but these are rare at most Javan H. erectus sites. It is one of the unique features of the Java record that remains to be explained.

What can the fossils tell us about the daily lives of these late members of H. erectus?

Although we don't have direct evidence, we believe that H. erectus exploited the resources in the area, probably by hunting or scavenging. Other clues about their lives can be found by looking at the non-human fossils, and what they can tell us about the ecology at the time. Two members of our team, Yahdi Zaim from the Institute of Technology Bandung and Robert Scott of Rutgers University in New Jersey, specialize in such fossils and found that they are mostly made up of large bovids — ancestors of the Javan banteng and water buffalo — as well as deer, Stegodon (an extinct elephant), rhinoceros, panther, crocodiles and turtles. The large percentage of bovids and deer could indicate that H. erectus lived in an open woodland or grassland environment.

Also, on the basis of estimates from partial skeletons of H. erectus from other sites, we think that the Ngandong H. erectus was probably between 1.66 and 1.85 metres tall — similar to the average human height in the United States today.

The Solo River, which runs near the Homo erectus site. Credit: O. Frank Huffman

What can the Ngandong excavation site tell us about the evolution and extinction of H. erectus?

The 14 H. erectus fossils are thought to represent a late stage in the evolution of the species. As a group, they have a significantly larger average brain size than that found in any other H. erectus fossils. Palaeoanthropologists don't like to use terms such as 'advanced' to distinguish one fossil group from another, but that term would certainly apply to the Ngandong fossils. After arriving on Java about 1.6 million years ago3, H. erectus apparently lived in 'splendid isolation' without competition from any other human species.

It is possible that when H. sapiens eventually reached Java, it could have competed with H. erectus for scarce island resources, leading to the extinction of the latter. But it is more likely that some unique geological or climatic event resulted in the extinction of Javan H. erectus, as is the case with most species. However, it does seem that the Ngandong group is the best evidence we have for the last occurrence of this species worldwide.

What has been the most exciting moment of the expedition so far?

The first high point came after several excavation pits were opened. We found boundaries of the original excavations not seen since the 1930s. This revealed untouched bone beds fitting the parameters originally described by the Dutch team. We are reliving the days of the discovery made nearly 80 years ago, and meeting present-day research objectives as we unearth the past.

What is a typical day at the site like?

We've been excavating for 24 days without a break. The days blur together and we often lose track of time. There is a routine to systematic palaeoanthropological excavation: opening an excavation pit, digging down to the bone bed, carefully mapping the strata as we proceed, exposing the fossils, assigning the fossil a number, charting its xyz coordinates, removing the fossil, and then sampling the strata for geological analysis and dating.

How does the Javan H. erectus compare with those from other sites in the world and where did it actually come from?

We spend a lot of time discussing this question. The earliest H. erectus fossils date to about 1.8 million years ago — appearing nearly simultaneously in both East Africa and at Dmanisi in the Republic of Georgia. So H. erectus from Africa and Georgia represent critical comparisons for any Javan H. erectus. It is thought that the species evolved in Africa and spread out of the continent to the Republic of Georgia and to Java over a very short timespan — perhaps less than 100,000 years. However, the most primitive and smallest H. erectus fossils come from Dmanisi. Anatomically, Dmanisi H. erectus shares features with both the African and Javan H. erectus, so it may actually be the centre of origin for the species. If you compare Ngandong H. erectus with those from elsewhere, it is clear that the most derived population is from Ngandong.