Plant-eating and meat-eating in Australopithecus

Scientists have many ways to squeeze evidence about ancient diets from the teeth of ancient human relatives. Some look at the mechanical damage on teeth from chewing food. Others examine microscopic fossils of plant parts embedded within dental calculus. Some researchers measure different aspects of the chemical makeup of bone or enamel. Integrating such different sources of information into a picture of an ancient species' diet can be a challenge. One kind of insight into the diet of an ancient animal is its trophic level—whether it ate mostly plants, mostly animals, or a mixture of the two.

Scientists have had one main avenue into understanding the trophic level and meat consumption of ancient animals: nitrogen-15. Animals have higher nitrogen-15 in their tissues than plants, animals that eat other animals tend to enrich nitrogen-15 even further. Measuring nitrogen-15 requires an adequate sample of nitrogen-containing compounds from the ancient animal. Bone is rich in nitrogen because of its abundance of the protein collagen, and the nitrogen comes from dietary protein. However, collagen breaks down and does not yield reliable nitrogen measurements after hundreds of thousands of years. That means that scientists have not been able to measure nitrogen-15 within many ancient human ancestors and relatives.

Earlier this year Tina Lüdecke and coworkers reported data on nitrogen-15 from Australopithecus africanus. Their numbers from seven teeth are the first measurements of nitrogen content from any hominins older than the Neandertals. To get this value from fossils more than two million years old, Lüdecke and her team developed high-precision measurements of very small amounts of organic material in enamel. Their finding of a relatively low nitrogen-15 value suggests that this species of Australopithecus probably didn't eat any large amount of meat.

This result was probably not a big surprise to anyone—indeed, it would have been much more surprising to find a substantial elevation in nitrogen-15 in the Australopithecus teeth. But the measurement is the first crack of a new window into the trophic level of many fossil human relatives.

How nitrogen isotopes connect to diet

When it comes to chemical analysis of ancient diets, the elemental workhorses have been carbon, nitrogen, and oxygen. Each has stable isotopes—forms that occur naturally and are different in atomic weight—that differ in abundance across natural environments and within different biological tissues.

Plants differ in the ratio of carbon-13 to carbon-12 based on the photosynthetic pathways they use. Animals obtain their carbon by eating plants, and this results in differences in ratio depending on diet. Water composed of oxygen-18 evaporates slightly less readily than water with only oxygen-16, and this difference can be detected in teeth of animals from arid or seasonal conditions.

Nitrogen-15 interacts less readily with metabolic enzymes than nitrogen-14. As a result, when animals eliminate nitrogen-rich waste products like urea, their tissues like bone and hair are left with a higher concentration of nitrogen-15. Animals take in nitrogen only from foods they eat. Carnivores tend to have higher nitrogen-15 concentrations in their tissues than herbivores. Marine animals have even higher nitrogen-15 levels due to many levels of the marine food chain.

Every amino acid molecule has a nitrogen atom. Across proteins they add up, and the most common and longest-lasting protein in ancient bones is collagen. Archaeological chemists have become adept at sampling collagen from ancient bone, giving a lot of information about groups of the latest Pleistocene and Holocene. From these kinds of measurements, scientists have learned that some European Neandertals relied strongly on meat from large mammals like mammoths.

Sampling nitrogen is a challenge for older fossils. Collagen eventually breaks down and its molecular remnants generally cannot be purified from bone after hundreds of thousands of years. Tooth enamel is very long-lasting but it contains little protein. So while scientists have been able to look at carbon and oxygen stable isotopes in ancient teeth, looking at nitrogen has been impractical.

Nitrogen from enamel

The methods developed by Lüdecke and team focus on the small amounts of nitrogen that are locked within enamel. In a 2023 paper led by Jennifer Leichliter, Lüdecke and team reported measurements that they took from enamel of modern African mammals, comparing directly to bone samples from the same individuals. In that work, they showed that their enamel method yielded similar measurements as collagen. That was a step toward a confident sampling of nitrogen-15 from fossil teeth.

The measure used in these comparisons is δ¹⁵N, or delta fifteen N, which is a measure of the relative abundance of nitrogen-15 in the sample compared to the atmospheric value. It is reported in parts per thousand (‰) or per mille. The scatterplot above shows that carnivores have higher relative abundance of nitrogen-15 in bone and enamel than herbivores. The nitrogen in carnivore bone and enamel comes from eating other animal species, which themselves have elevated nitrogen-15 compared to the plants they eat.

The comparison of nitrogen-15 and carbon-13 levels provides a broader perspective on the different dietary niches of African mammals today. Herbivores have low nitrogen-15 values and vary in carbon-13 depending on whether they are grazers eating a large fraction of grasses, like white rhino and black wildebeest, or browsers eating mostly tree leaves, like kudu or gorillas. Carnivores have high nitrogen-15 and vary in carbon-13 depending on their prey species: lions eat more grazers, leopards eat more browsers.

Baboons are omnivores. They eat both plant foods and animal foods, mostly invertebrates, small vertebrates, and eggs. The extent that baboons focus on animal foods varies enormously across studies of different field sites. Small vertebrates are a low proportion of the baboon diet almost everywhere, but some baboons get up to 20% of their bulk diet from insects. Insects and vertebrates make a greater nutritional than caloric difference to baboons, as they are important sources of protein, sodium, and fats.

The baboons included in the 2023 work by Leichliter and coworkers were yellow baboons from Kenya and Tanzania. Their δ¹⁵N and δ¹³C values from enamel overlap with the mixed plant feeders like springbok and elephants. They also overlap with leopards, suggesting that in this range of values, separating carnivores from omnivores and certain herbivores may be challenging.

Adding Australopithecus to the picture

The newest article from Lüdecke and coworkers added fossil species from Sterkfontein Member 4 into this picture. The age of these fossils is subject to some disagreement among specialists in geochronology. Some methods suggest that the Member 4 fossils are around 3.6 to 3.3 million years old, but a number of researchers disagree and instead assert that Member 4 accumulated between 2.6 and 2.4 million years ago. Whichever date is accurate, the Member 4 fossils include Australopithecus as well as several other species of primates: the baboon relatives Parapapio and Papio, as well as the colobine monkey Cercopithecoides. The site also has the sabertooth cat Dinofelis and many other carnivores, and a wide array of bovids.

The contrast in enamel nitrogen-15 isotopes between carnivores and bovids in living species is also manifested in the fossil species. That's a good indication that the ancient enamel is preserving the signal of trophic level that can be found in living species' tissues.

The fossil primates overlap in their nitrogen-15 values with living primates. In this study, Lüdecke and coworkers added a second sampling location for recent baboons, including the southern African Papio ursinus, and this species had lower nitrogen-15 than the east African samples. A similar range of variation was present among the Sterkfontein non-hominin primates.

Australopithecus individuals average a bit lower than the fossil baboons, but within their range, and a lot lower than any of the fossil or living carnivores. This was the major result reported as the conclusion of the study: the results show that the trophic level of Australopithecus was comparable to fossil baboons and within the range of fossil ungulates.

That low nitrogen-15 value for Australopithecus is quite different from Neandertals and early Upper Paleolithic-associated modern humans in Europe. These recent human groups both register higher nitrogen-15 values than many carnivores that inhabited their environments. Living human populations also tend to have high nitrogen-15—though admittedly their values can be more challenging to interpret once substantial marine or aquatic foods are included in diets, because of their high nitrogen-15 enrichment. All this is to say that Australopithecus was not very much like historic and recent humans.

What I found more interesting in the discussion by Lüdecke and coworkers is their approach to the variability of δ¹⁵N values in the Australopithecus teeth. With seven teeth sampled, this is one of the more numerous among all the species they examined, fossil or extant. The Australopithecus values span a factor of four. Lüdecke and coworkers suggest that the variability in δ¹⁵N values of these hominins may be information about their diets.

“Overall, these results suggest that Australopithecus may have had more variable diets than any of the analyzed nonhuman primate, artiodactyl or carnivore species, including some individual-level diet niche separation.”—Tina Lüdecke and coworkers

Following up on this idea will require larger sample sizes for many of the species to better understand their range of variation. Especially valuable will be greater sampling of populations of baboons, which already show some large differences in the few samples that have been examined.

Sources of variation

The results of this study do not mean that Australopithecus from Sterkfontein Member 4 was a herbivore. Baboons are omnivorous, with insect consumption and some consumption of vertebrate meat both very important in their diets, and they have similar nitrogen-15 values as the Australopithecus teeth. I suspect that early hominins were a lot like chimpanzees in the amounts of meat and insects that they consumed. This would be consistent with the nitrogen-15 data.

It would also be consistent with the archaeological record. Hominins were consuming medium and large vertebrate prey from at least very dawn of stone toolmaking, more than three million years ago. Some of that evidence of early hunting and scavenging must have been made by species outside our genus, such as Australopithecus, Kenyanthropus, and Paranthropus. As I've written before, there is no evidence for preferential association of any genus of hominins with the earliest occurrences of stone tools.

With that in mind, the idea that different individuals of Australopithecus had substantially different intakes of nitrogen-15-rich foods is fascinating. Among the chimpanzees of Taï National Park, in Côte d'Ivoire, male and female adult chimpanzees have quite different access to meat. Male adults are often involved in coalitional hunting and use meat as a social currency. Female adults in this population and immature individuals of both sexes have less meat consumption. Scientists led by Vicky Oelze studied samples of hair from these and many other chimpanzees and found these differences in meat consumption echoed in the nitrogen-15 content of their hair. Maybe Australopithecus also had social dimensions of diet that led to differences in meat intake between male and female adults.

If sex was an important dimension of dietary variability, sampling protein residues from the same teeth as isotopes would add significant information to our understanding of early hominin societies. The enamel proteome is a limited source of information about genetic makeup, but one enamel protein—amelogenin—has isoforms on both the X and Y chromosomes that can be detected in ancient enamel. This variation can test the chromosomal sex of ancient individuals. Last month Palesa Madupe and coworkers reported on the sex of one Australopithecus tooth from Sterkfontein, showing the potential of this method for uncovering this aspect of biological variability from teeth at this site. In the future, isotopic data should be aligned with proteomic data—and genetic data, where it may be possible—to deepen the biological context.

Lüdecke and collaborators themselves suggest that ontogeny may contribute to the variation in nitrogen-15. Milk is a food source that is elevated in nitrogen-15, and teeth that form as children are breastfed tend to be enriched in this isotope. They point out that two of their sampled teeth were first molars, one with a very high nitrogen-15 value and the other much lower. Already evidence from a different sample of Sterkfontein teeth has suggested early weaning in some Australopithecus children, along with a seasonal pattern of dietary stress. The behavioral flexibility to transcend such events may have contributed to the variation in nitrogen isotopes as well.

Archaeologists long favored the idea that our own genus, Homo, relied more on meat than earlier fossil relatives. This notion that meat was an essential ingredient for evolution of humanlike features such as large brains and cooperative behavior was as old as Charles Darwin, but it has recently been challenged on several fronts. Certainly hunter-gatherers of the Late Pleistocene and Holocene have eaten a higher fraction of hunted meat than our earliest hominin relatives, but when did this shift happen? Recent work has questioned whether early members of our genus such as Homo erectus represented any increase in carnivory compared to earlier Oldowan toolmakers.

That debate has been unfolding with a yawning gap of no information about nitrogen-15 prior to the Neandertals. That's about to change.

Notes: Writing this post generated a lot of additional notes on various aspects of diet in early hominins. That includes sampling of calcium stable isotopes, which many researchers see has an additional way of learning about trophic level, but which pose some challenges of interpretation.

There is now a voluminous literature on carbon stable isotopes and dietary inferences in early hominins. The addition of nitrogen evidence will undoubtedly cause some reinterpretation of dietary sources of carbon. One unresolved aspect of the carbon evidence is the higher carbon-13 levels in southern African Australopithecus and Paranthropus, for which meat or insect consumption might be one source. Whether this may be consistent with the lower nitrogen-15 in these Sterkfontein Member 4 fossils is not yet clear.

Pickering and Domínguez-Rodrigo (2010) provided a summary of the ways that observations of hunting behavior in chimpanzees may provide models for thinking about hunting in early hominins. The social dimensions of hunting are an important aspect of their discussion.

References

Barr, W. A., Pobiner, B., Rowan, J., Du, A., & Faith, J. T. (2022). No sustained increase in zooarchaeological evidence for carnivory after the appearance of Homo erectus. Proceedings of the National Academy of Sciences, 119(5), e2115540119. https://doi.org/10.1073/pnas.2115540119

Ihobe, H., & Sakamaki, T. (2023). Hunting and Meat-Eating Behaviors of Bonobos at Wamba: Comparison with Other Bonobo Study Sites. In T. Furuichi, G. Idani, D. Kimura, H. Ihobe, & C. Hashimoto (Eds.), Bonobos and People at Wamba: 50 Years of Research (pp. 115–131). Springer Nature. https://doi.org/10.1007/978-981-99-4788-1_11

Joannes-Boyau, R., Adams, J. W., Austin, C., Arora, M., Moffat, I., Herries, A. I. R., Tonge, M. P., Benazzi, S., Evans, A. R., Kullmer, O., Wroe, S., Dosseto, A., & Fiorenza, L. (2019). Elemental signatures of Australopithecus africanus teeth reveal seasonal dietary stress. Nature, 572(7767), 112–115. https://doi.org/10.1038/s41586-019-1370-5

Leichliter, J. N., Lüdecke, T., Foreman, A. D., Bourgon, N., Duprey, N. N., Vonhof, H., Souksavatdy, V., Bacon, A.-M., Sigman, D. M., Tütken, T., & Martínez-García, A. (2023). Tooth enamel nitrogen isotope composition records trophic position: A tool for reconstructing food webs. Communications Biology, 6(1), 1–13. https://doi.org/10.1038/s42003-023-04744-y

Lüdecke, T., Leichliter, J. N., Stratford, D., Sigman, D. M., Vonhof, H., Haug, G. H., Bamford, M. K., & Martínez-García, A. (2025). Australopithecus at Sterkfontein did not consume substantial mammalian meat. Science, 387(6731), 309–314. https://doi.org/10.1126/science.adq7315

Madupe, P. P., Munir, F., Dickinson, M., Taurozzi, A. J., Mackie, M., Tawane, M., Mollereau, C., Hlazo, N., Penkman, K., Schroeder, L., Zanolli, C., Olsen, J. V., Ackermann, R. R., & Cappellini, E. (2025). Results from an Australopithecus africanus dental enamel fragment confirm the potential of palaeoproteomics for South African Plio-Pleistocene fossil sites. South African Journal of Science, 121(1/2), Article 1/2. https://doi.org/10.17159/sajs.2025/18571

Oelze, V. M., Fahy, G., Hohmann, G., Robbins, M. M., Leinert, V., Lee, K., Eshuis, H., Seiler, N., Wessling, E. G., Head, J., Boesch, C., & Kühl, H. S. (2016). Comparative isotope ecology of African great apes. Journal of Human Evolution, 101, 1–16. https://doi.org/10.1016/j.jhevol.2016.08.007

Pickering, T.R. & Domínguez-Rodrigo, M. (2010). Chimpanzee Referents and the Emergence of Human Hunting. The Open Anthropology Journal, 3, 107–113. https://doi.org/10.2174/1874912701003010107

Watts, D. P. (2020). Meat eating by nonhuman primates: A review and synthesis. Journal of Human Evolution, 149, 102882. https://doi.org/10.1016/j.jhevol.2020.102882