New hominin teeth from Ledi-Geraru, Ethiopia

Reviewing new evidence that suggests a presence of Homo and Australopithecus in the time before 2.5 million years ago.

Fossil teeth from Ledi-Geraru including LD 760 teeth at left, and LD 302-23, LD 750, and AS 100 teeth at right. Image: Brian Villmoare and coworkers (2025) CC-BY-NC-ND license.

I always love to see new fossils described for the first time. Last week Nature published a description of teeth from the Ledi-Geraru field area of Ethiopia by Brian Villmoare and a team of collaborators led by Kaye Reed and J. Ramón Arrowsmith. The team found the thirteen teeth in 2015 and 2018 and report that they come from a range of times from 2.78 million to 2.59 million years ago.

The thirteen teeth in the study were found in 2015 and 2018 at four different collection points, or localities, within the Ledi-Geraru area. They represent a minimum of four individuals that lived at different times across the roughly 200,000-year interval.

The study’s conclusion emphasizes diversity. The authors assess the four individuals as representing two different kinds of human relatives, one a form of Australopithecus and the other early Homo.

“These specimens suggest that Australopithecus and early Homo co-existed as two non-robust lineages in the Afar Region before 2.5 million years ago, and that the hominin fossil record is more diverse than previously known.”—Brian Villmoare and collaborators

I’m writing this brief summary to review the findings and share a perspective on this interesting work. The analysis of the teeth rides on small details of their shape. I’m sure that many people who read journalists’ accounts of the work will wonder whether these fairly subtle differences are really important. In my view, the authors express a clear hypothesis that follows from the comparisons they did with other fossil samples.

That doesn’t mean I necessarily agree, of course. There are a few comparisons I would really like to see that the research team didn’t do.

But that’s the purpose of publishing a description of these teeth, so that each fossil can be integrated into the broader record, enabling follow-up research of many kinds. I’m really pleased that Nature published these descriptions, making the fossils available to the broader community to study, instead of holding up the authors for more analyses and years of delay.

A new Australopithecus?

Nine of the teeth come from one locality—numbered LD 760—within the Lee Adoyta basin. Collected from an area of around 10 meters in diameter, the team interprets these as teeth of a single individual, just over 2.63 million years in geological age.

These are the teeth that have drawn the most attention to the new study. The team attributes them to Australopithecus, based on their relative sizes and absence of derived features found in two other genera, Homo and Paranthropus.

Australopithecus would not be at all unlikely considering that both earlier and later species are known from within 100 km of the Ledi-Geraru site. The earlier species, Australopithecus afarensis and Australopithecus deyiremeda, both are known to have existed before 2.9 million years ago. The later one, Australopithecus garhi, is known from one partial skull from around 2.5 million years ago. LD 760 is in the middle.

Villmoare and coworkers observe that the LD 760 teeth do not have features that are known to be diagnostic within any of these known species.

This is admittedly a tough argument for A. garhi since its only known specimen BOU-VP-12/130 has only upper teeth, and most of the LD 760 teeth are lower. Only the upper canine provides a comparison, and these are indeed different in size, and in some details of shape.

These comparisons give rise to the hypothesis that the LD 760 teeth might be something different from any of the known species of Australopithecus from the region.

A single premolar from another locality, LD 750, just under 2.63 million years, is also Australopithecus in the team’s view. The assessment focuses in part on the premolar’s large size, and the authors suggest it may belong to the same previously-unidentified species.

Evolving difference

Several evolutionary pathways might explain why these Ledi-Geraru teeth look a bit different from earlier and later Australopithecus species from the same region. The LD 760 individual lived three hundred thousand years or more after any known A. afarensis individuals. It represents a late population of A. afarensis or A. deyiremeda that had evolved some anatomical differenes from the earlier populations. (Many anthropologists would call that a new species by definition.) Or the Ledi-Geraru fossils may be an earlier population of A. garhi, one hundred thousand years or more earlier than the only known skull of that species.

Or both. The phylogenetic tree presented by Mana Dembo and coworkers in 2016 suggested that A. garhi was a sister species and possible descendant of A. afarensis. Maybe LD 760 was the missing link between them.

The authors listed these possibilities and I don’t think the data reject them. I would add some that they didn’t list. A regional population of Australopithecus africanus or Australopithecus sediba, which are otherwise known from South African sites, may have extended to northeastern Africa. Kenyanthropus platyops, so far known only from the Turkana Basin, may have had a presence further to the north. I’ll be very interested when it’s possible to do more comparisons with those species.

To me, every one of these possible evolutionary pathways are “might be” or “could be”. I don’t think I’m saying anything very original by listing these, and what’s most notable about the Ledi-Geraru evidence is that none of them are testable at the moment with the data from these teeth.

So what name should scientists use for the LD 760 individual?

That’s where the position of researchers who describe fossils is different from everyone else. A description must include some name.

Villmoare and collaborators find that the LD 760 teeth don’t share features that they consider to be diagnostic in A. afarensis, A. deyiremeda, or A. garhi. At the same time, the LD 760 teeth don’t have unique features of their own. They take a conservative approach: attribute the teeth to Australopithecus but not a new species. At least not yet.

Homo without a name

The same team took a similar approach to the mandibular fragment that they identified as Homo in 2015, LD 350-1. At an estimated age between 2.8 and 2.75 million years, this has long been the earliest specimen attributed to our genus. It lacks some of the derived features of later species of Homo, such as Homo habilis, and yet has no derived features of its own. Since its description, its identification has been limited to the genus level.

As I published with Lee Berger and Darryl de Ruiter back in 2015, my view is that the LD 350-1 fossil shares much of its Homo-like morphology of Australopithecus sediba. This species is generally recognized today as the closest relative of Homo, and so the similarity between its fossils and LD 350-1 may not be surprising. If they are not the same, I suspect that they come from related populations.

The new study reports three teeth that the team attributes to Homo. One is a broken premolar from the locality LD 302 (LD 302-23). This tooth comes from a geological unit that they reference as the “Gurumaha sedimentary package”, estimated around 2.78 million years old.

The other two teeth are first and second upper molars of one individual from the Asboli area of the survey area. These come from the Giddi Sands geological unit, estimated around 2.59 million years old. In the authors’ assessment, while the first molar is fragmented, the second molar is different in form from any of the Australopithecus species from the region and it shares aspects of crown shape with later fossils that have been attributed to Homo. It’s particularly similar in shape and size to the second molar of the 2.33-million-year-old AL 666-1 maxilla. This, too, is a specimen of early Homo without a species name.

The new teeth from Asboli are upper molars, and so cannot be compared directly with the LD 350-1 mandibular molars. The LD 302-23 lower third premolar can only be partially compared with the broken tooth in the LD 350-1 mandible. It’s a conservative proposal by Villmoare and coworkers in their new paper to suggest that these are not different species.

Opening the reviews

A neat thing about this research is the peer review file that is published alongside the article. It includes four reviews of the initial draft, author responses to those reviews, a little back-and-forth with two of the reviewers, and some explanations for how the final manuscript was changed to accommodate the reviews. These interactions give a real picture of how experts differ in their ways of assessing variation among fossils.

It’s clear in the end that they did not all accept the taxonomic interpretation that the authors prefer. They also weren’t happy with every illustration presented in the paper—often for reasons the authors cannot control. What makes this a great read is that it’s clear that the reviewers saw the value in having the work published, even if they might have done many things differently.

To me, that’s a really valuable outcome.

The peer review file also has some interesting nuggets about some challenges the team faced with access to comparative data for their work. Revealing those kinds of impediments to research has become one of the most compelling reasons to open up the review process entirely.

This upcoming semester I’ll be teaching a course for graduate and advanced undergraduate students looking closely at the diversity and adaptations of early hominins. This new research from Ledi-Geraru could not be more timely. It’s an exciting day in the lab when I can share publications like this with students. We’ll go through the reviews, talk about what these new teeth add to our existing knowledge, and reflect on the limitations of the data.

References

Alemseged, Z., Wynn, J. G., Geraads, D., Reed, D., Andrew Barr, W., Bobe, R., McPherron, S. P., Deino, A., Alene, M., J. Sier, M., Roman, D., & Mohan, J. (2020). Fossils from Mille-Logya, Afar, Ethiopia, elucidate the link between Pliocene environmental changes and Homo origins. Nature Communications, 11(1), 2480. https://doi.org/10.1038/s41467-020-16060-8

Hawks, J., de Ruiter, D. J., & Berger, L. R. (2015). Comment on “Early Homo at 2.8 Ma from Ledi-Geraru, Afar, Ethiopia.” Science, 348(6241), 1326–1326. https://doi.org/10.1126/science.aab0591

Villmoare, B., Delezene, L. K., Rector, A. L., DiMaggio, E. N., Campisano, C. J., Feary, D. A., Ali, B. M., Chupik, D., Deino, A. L., Garello, D. I., Hayidara, M. A., Locke, E. M., Omar, O. A., Robinson, J. R., Scott, E., Smail, I. E., Geleta Terefe, K., Werdelin, L., Kimbel, W. H., … Reed, K. E. (2025). New discoveries of Australopithecus and Homo from Ledi-Geraru, Ethiopia. Nature, 1–7. https://doi.org/10.1038/s41586-025-09390-4

Villmoare, B., Kimbel, W. H., Seyoum, C., Campisano, C. J., DiMaggio, E. N., Rowan, J., Braun, D. R., Arrowsmith, J. R., & Reed, K. E. (2015). Early Homo at 2.8 Ma from Ledi-Geraru, Afar, Ethiopia. Science, 347(6228), 1352–1355. https://doi.org/10.1126/science.aaa1343