The debate about when exactly Homo sapiens left Africa remains a hot topic. Over recent years, growing evidence has been published of sites with H. sapiens from earlier than the 50,000 years the “Out of Africa” model establishes for the earliest migrations of our species out of the African continent. Among this evidence is that from the Fuyan Cave site in Daoxian, southern China, published in the year 2015 by the Centro Nacional de investigación sobre la Evolución Humana (CENIEH) researchers María Martinón-Torres and José María Bermúdez de Castro, in the journal Nature, as part of a collaboration with scientists from the Institute of Vertebrate Paleontology and Paleoanthropology (IVPP), in Beijing.

At the start of this year, a study published in the Proceedings of the National Academy of Sciences USA (PNAS) by Xue-feng Sun, a researcher at Nanjing University, and his team, questioned the ages of no fewer than five sites in China, one of which was Fuyan. This study reviewed the datings for these sites and, in the case of Fuyan, presented two new "human" teeth, which they attributed to the same sample as that published by Martinón-Torres and Bermúdez de Castro. Through genetic analysis and radiocarbon dating of these new teeth, Sun and his collaborators concluded that the Fuyan teeth, originally dated to between 80,000 and 120,000 years old by the IVPP team, were barely 10,000 years old.

Now, the same CENIEH researchers, once again in collaboration with scientists at the IVPP in Beijing, have published a reply in PNAS where they question the quality of the study conducted by Sun and his colleagues. Apart from problems with the dating and the context of the fossil remains, the main criticism lies in the fact that one of the "human" teeth analyzed by Sun's team and from which they extracted human DNA, is in reality the tooth of a herbivore, which was probably a deer.

“Science moves forward with data, and to be able to rebut a hypothesis, the data need to be reliable”, says the CENIEH director, María Martinón-Torres, who is the lead author of this article. “Extracting human DNA from a tooth which is not human calls into question the credibility and scientific standards of the entire work”, she continues.

Surprisingly, in their own response in PNAS, led in this case by Darren Curnoe of the Australian Museum in Sydney, these researchers insist on their taxonomic identification, even though they do not offer any data. “We continue to await a minimal description, a morphological and metric comparison, a microtomography, an image analysis supporting the claim. Something. Science is data-driven”, comments Martinón-Torres.

The CENIEH and IVPP researchers have compiled assessments from dozens of specialists all over the world, who have ratified that the tooth is not human. “Moreover, the context in which these two new teeth were found is unknown”, emphasizes Bermúdez de Castro, coordinator of the CENIEH Paleobiology Program. “Sun, Curnoe and their team claim that they were found at the same level as the sample we published in 2015, but at no time have they made contact with or coordinated with Liu Wu and Xiujie Wu at the IVPP, who are in charge of the excavation, and are therefore familiar with the site and its stratigraphy”.

As the CENIEH researchers explain, the study by Sun is plagued by problems overall. “In addition to a lack of professionalism and rigor, this way of proceeding is a far cry from the code of good conduct expected in scientific work”, they add.

The dating, also questioned

For their part, a research team at Oxford University has criticized the dating methods employed by Sun and his colleagues in their paper. In a letter sent separately to PNAS, Tom Higham and Katerina Douka, experts in radiocarbon dating, point out that the methods employed by the Nanjing University team are not as solid as is recommended for dating samples of this age.

In particular, the chemical pre-treatment methods recommended for dating bones and carbon from the Pleistocene are absent. According to Higham and Douka, this means that the dates proposed in Sun's study are not reliable and, in any case, they should be regarded as minimum estimated dates.