The Footprints That Were Too Big for the Expected Species

Twenty-one footprints, preserved in ancient lakeside silt in northern Kenya for 1.43 million years. The anatomy said one species. The body size said another. The study published in PNAS by Kevin Hatala of Chatham University and the Max Planck Institute for Evolutionary Anthropology and twelve colleagues does not resolve that contradiction. It documents it, and in doing so changes what scientists thought was possible for at least one of the candidates.

A Lakeshore Scene That Survived

The GaJi10 site lies within the Koobi Fora Formation in East Turkana, northern Kenya, one of the world’s most productive regions for studying human evolution. The site was first identified in 1978, but expansions of the excavation in 2016 and 2023 exposed substantially more of the footprint surface. Researchers have now documented 21 hominin tracks organized into five continuous trackways, provisionally representing eight individuals.

The tracks were made approximately 1.43 million years ago, a date established by correlating the volcanic deposit approximately 40 centimetres above the footprint surface with the regionally known Akait Tuff. At the time, the area was the shoreline of ancient Lake Lorenyang, and the hominins moved through shallow water estimated at two to ten centimetres in depth. Fine sand subsequently filled the impressions and preserved their shape before later sediments buried the surface and locked it in place.

The hominins were not alone. Alongside the 21 human relative tracks, researchers documented 146 animal footprints: hippopotamuses, bovids, birds, and equids. The surface records a busy lakeshore, visited within a short window of time by multiple species. Most of the hominins were moving westward. One walked in the opposite direction.

Feet That Moved Differently From Ours

The research team created detailed three-dimensional models of each footprint and compared them with fossil tracks from across East Africa, human footprints from other regions, and hundreds of experimentally produced tracks from modern people. The differences from modern human walking were consistent and significant.

The GaJi10 footprints show a flatter arch than modern humans produce. The heel-to-toe movement pattern differs from our own. Most strikingly, the big toe in many of these tracks was angled substantially outward from the rest of the foot: in some prints, the angle reached nearly 40 degrees. Modern human big toes generally align close to the remaining toes.

These characteristics match the anatomical signature of other East Turkana footprints that have previously been attributed to Paranthropus boisei, a robust hominin with a large face and powerful chewing apparatus that lived in East Africa from roughly 2.3 to 1.2 million years ago. The GaJi10 tracks also share features with the 3.66-million-year-old Laetoli footprints in Tanzania, generally associated with Australopithecus afarensis, indicating that this style of foot movement had a long history in the human family. Footprints attributed to Homo erectus at other East Turkana sites, by contrast, show a more modern human-like pattern, with a more aligned big toe and a different weight-transfer profile through the step.

So far, the anatomy of the GaJi10 tracks points clearly toward P. boisei.

The Body Size That Doesn’t Match

The body size estimates from the same footprints tell a different story. Because the species cannot yet be conclusively identified, the team calculated size using two models, one based on body proportions expected for P. boisei and one based on Homo erectus proportions.

Under the P. boisei model, the seven individuals for whom reliable estimates could be made would have stood between approximately 1.31 and 1.81 metres tall and weighed between 39.8 and 74.6 kilograms. All seven height estimates exceeded previously published average values for P. boisei. Four exceeded the tallest skeletal individual in the comparison set. Six of the seven mass estimates fell above the highest published species average.

This does not fit the picture of P. boisei that skeletal remains have established. The species has been understood as relatively compact and not particularly tall. A set of individuals in which every single member exceeds the known average, and several exceed the known maximum, suggests either that current estimates of P. boisei body size are significantly incomplete, or that these tracks belong to a different or unexpected population.

Using Homo erectus proportions instead produces estimated heights of approximately 1.55 to 1.81 metres and masses of 55 to 79.6 kilograms. These values are toward the large end for H. erectus, but they fall within the range already documented in skeletal remains. The body size problem is considerably less severe under the H. erectus model.

This creates the central puzzle of GaJi10. The foot anatomy favors P. boisei. The body size is more naturally accommodated by Homo erectus. Both interpretations have something going for them, and neither resolves cleanly into a complete picture.

A Group Moving Together

The footprint surface also preserves something that fossilized bones almost never can: information about social behavior in the moment it occurred. Five of the larger individuals appear to have moved westward in broadly similar directions. Their tracks do not significantly overlap, which the researchers interpret as evidence that these individuals crossed the surface at approximately the same time rather than hours or days apart.

Under the P. boisei body-size model, the estimated sizes of the five individuals suggest that several may have been adult males. Adult males of most hominin and great ape species tend to be significantly larger than females. If these five were male, and if they crossed at the same time, the surface preserves a rare snapshot of early hominin social tolerance: multiple adult males in the same space at the same moment, either simply tolerating one another or belonging to a shared social group.

The researchers are careful about this interpretation. Body size differences cannot be converted directly into a precise social system. The specific nature of the relationships between these individuals cannot be recovered. What the tracks establish is that they were there at the same time, in the same small stretch of lakeshore, and that they were large.

Why the Lake Margins Mattered

GaJi10 is one of a growing number of Early Pleistocene footprint sites found along former lake and river shorelines in East Turkana. The concentration of evidence in these environments is not coincidental. Wetland margins offered reliable water, and for P. boisei, papyrus and other sedges may have been important food sources. The shallow water itself could provide shellfish and fish alongside the plant resources of the shoreline.

Evidence from other East Turkana sites suggests that P. boisei maintained a relatively small home range closely connected to stable lakeshore vegetation, while Homo erectus exploited a broader range of environments and increasingly incorporated animal foods into its diet. Both species returned to these lakeshore zones repeatedly. Earlier discoveries in the same formation have even preserved tracks from both species on a single surface, demonstrating that the two hominins sometimes passed through the same space within hours or days of each other.

The precise relationship between them remains unknown. Competition, coexistence, indifference: the tracks record presence, not relationship. They do not tell us whether these hominins were aware of each other, whether they avoided contact or sought it, or what it meant to be two different species of early human relative sharing a lakeshore in the East African Rift Valley 1.43 million years ago.

An Identification Still Open

The Hatala team makes no definitive species claim for the GaJi10 footprints. Anatomically, the tracks resemble P. boisei. In size, they better fit Homo erectus, or they suggest a P. boisei population substantially larger than previously documented. A third, currently unknown hominin cannot be formally excluded, although decades of work in this geological layer have produced fossil evidence for only two species.

What the footprints have already established is independent of the species question. Something large was walking along a lake margin in northern Kenya 1.43 million years ago. Several of these individuals moved through the same space at nearly the same time. They moved slowly, through shallow water, alongside hippopotamuses and cattle ancestors and birds. And the silt remembered every step.

Sources

Hatala, K.G., Roach, N.T., Harris, J.W.K., Kiura, P., Ndiema, E.K., Sila, B., Kipkebut, D., Waweru, J., Kisale, P., Longaye, A.A., Sale, H., and Behrensmeyer, A.K. (2026). Insights into hominin body size, locomotion, and behavior from Early Pleistocene trackways in northern Kenya. Proceedings of the National Academy of Sciences, 123(31): e2530996123. https://doi.org/10.1073/pnas.2530996123