Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.


Hominin footprints at Laetoli reveal a walk on the wild side

The human version of walking on two legs, known as striding bipedalism, is unique among mammals. It requires the ability to balance a tower of loosely connected body parts over a single foot, as the other foot swings forwards to complete the stride. Conventional wisdom holds that this ungainly form of locomotion had a single evolutionary origin in an ancestral hominin, followed by about six million years during which further anatomical adjustments accumulated — a linear model of evolution in which early hominin bipedalism became progressively more similar to our own over time. However, fossils discovered during the past decade show that multiple versions of bipedalism existed simultaneously during one or more periods of hominin evolution. Writing in Nature, McNutt et al.1 suggest that evidence of locomotor diversity in hominins has been overlooked for many decades.

More than 3.6 million years ago, ash fallout after a volcanic eruption blanketed the landscape at Laetoli in northern Tanzania. Animals left footprints in the ash layer as they searched for food, water and protection. Among the prints2 of ostriches, giraffes, hyenas and chalicotheres (imagine a knuckle-walking horse with claws) are some that are instantly recognizable as hominin footprints. These traces, called site G footprints (similar footprints were also discovered later at site S at Laetoli), were left by an early hominin that had feet shaped like ours, and that walked using a biomechanical pattern very similar to our own. These iconic footprints2 helped to prove that striding bipedalism appeared millions of years before big brains and skilled toolmaking.

Only a few experts are familiar with the enigmatic footprints found at Laetoli site A. These prints were partially excavated in the 1970s, documented and then quickly reburied to protect against erosion2. Unlike the footprints at sites G and S, the site A prints have an unusual shape and document a bipedal walking movement that occurred in a peculiar cross-stepping manner, in which each foot crosses over the body’s midline to touch down in front of the other foot (Fig. 1). One explanation for the enigmatic site A prints was that they were made by a bear walking bipedally. An alternative proposal was that the prints from sites G and S and those of site A were made by different kinds of hominin3,4. Scientists were not convinced by either explanation. Ultimately, the site A prints were more easily forgotten than explained.

Figure 1

Figure 1 | Footprint analysis. a, McNutt et al.1 examined an enigmatic 3.6-million-year-old footprint track, preserved in volcanic ash, at site A in Laetoli, Tanzania. The authors conclude that the tracks were made by a hominin (a member of the family tree that includes humans) using an unusual cross-stepping motion in which the foot crosses to the other side of the body at each step. b, Other hominin footprints (left print shown) of the same age exist at another Laetoli site (site G), but these tracks are not associated with cross-stepping. c, A close-up of a left footprint from site A reveals a wider and shorter footprint than that of site G prints, raising the possibility that two different hominin species were there at the same time. d, The authors investigated whether the site A tracks might have arisen from a bear walking bipedally. However, such bear prints (left print shown) do not match the characteristics of the site A tracks. McNutt and colleagues conclude that the angle between the big toe and second toe of the site A prints is more ape-like than like that of a typical hominin. Might this mean that the big toe had a thumb-like grasping capacity? e, Chimpanzees have that type of big toe, which can leave an impression that is at a notable angle to the side of the foot, as in this left print.Credit: a, b, d: E. J. McNutt et al./Nature; c: John Reader/SPL; e: Stephanie Melillo, in collaboration with the Loango Chimpanzee Project

New excavations of the site A footprints by McNutt et al. reveal a combination of features diagnostic of hominins. The big toe and second toe are similar in length; the impression made on the ground by the big toe is much larger than that made by the second toe; the impressions made by the toes and the rest of the foot are continuous; and the heel is wide (Fig. 1). Still, the site A footprints are unlike those of any other hominin. The footprints themselves are oddly wide and short, and the feet responsible for their creation might have had a big toe that was capable of thumb-like grasping, similar to the big toe of apes.

The authors point out that cross-stepping is possible only thanks to unique aspects of hominin skeletal anatomy that position the feet extremely close to the body’s midline, helping us to balance during walking when we are supported by a single foot. If the walker responsible for the site A prints had a foot positioned under the hip joint, as is the case for bears and other mammals, sequential footprints would be situated much more widely apart than they are in the site A tracks.

A bipedal bear might sound like a circus act, but the development of this proposal in the 1980s was insightful. Because of the way a bear’s foot is shaped, it is possible to mistake a bear’s right footprint for a primate’s left footprint. This means that if the footprints were made by a bear, the site A track could have been made without cross-stepping. But if these prints were made by a hominin, as McNutt et al. convincingly show, then the cross-stepping mystery remains. It is difficult to imagine cross-stepping as the normal gait of a biped. Was the site A individual injured or stumbling? The number of possible scenarios is limited only by our imagination. However, the authors show that footprints made by cross-stepping humans remain recognizably human, so the unusual appearance of the site A prints is probably not due to the cross-stepping movement alone.

After heated debates in the 1970s to 1980s, most palaeoanthropologists reached a consensus that all fossil bones and footprints dated to the middle Pliocene epoch (roughly 3.7 million to 3 million years ago) represented the hominin species Australopithecus afarensis. This species was the earliest hominin known at that time and the presumed ancestor to all later hominin species. However, fossils discovered in the past two decades challenge the hard-won consensus5,6.

The 3.4-million-year-old ‘Burtele foot’ from the Woranso-Mille project area in Ethiopia7 is particularly difficult to reconcile with the prints from sites G and S and similarly aged fossil bones from Hadar, Ethiopia. The Burtele foot shows some features diagnostic of bipedal hominins, but they occur in combination with a short big toe that is angled away from the foot and a second toe that curves towards the first. This arrangement indicates that the inner side of the Burtele foot would have been capable of hand-like grasping7, which is an important adaptation to moving through trees that apes share8. This fossil reveals a strikingly different version of bipedalism from the one inferred for A. afarensis.

Could the site A footprints and the Burtele foot represent the same hominin species — one that is distinct from A. afarensis? This would be the simplest explanation. These fossils come from a similar place and time, and both provide evidence of a foot that is more ape-like than is that of A. afarensis. However, differences between the Burtele foot and the site A prints appear when the details are examined closely. Most notably, the first and second toes in the site A footprints are of similar lengths, whereas the Burtele foot’s big toe is relatively short compared with its second toe. It is unclear whether differences such as this could exist among individuals of the same species. No living animal has a combination grasping–bipedal foot, so we have no reference for the kind of print such a foot would make.

It seems that two possibilities remain as probable explanations for the site A prints. They could have been created by a hominin species other than A. afarensis (perhaps the same as that represented by the Burtele foot). Alternatively, they could have been created by an A. afarensis individual walking in an atypical manner other than that tested in the study by McNutt and colleagues. Virtual approaches that simulate the process of footprint creation in extinct species might help us to decide between these options in the future.

If the footprints at sites G and S and those at site A were made by different species, then the Laetoli footprint tuff (rock made of volcanic ash) captures multiple hominin species living in the same habitat and at the same geological instant. This level of precision is exceedingly rare in the fossil record. It would demonstrate species coexistence in a way that implies competition for ecological resources, and it would provide a new view of the evolutionary forces at play during the early periods of human evolution.

But which hominin species existed alongside A. afarensis? The site A footprints and the Burtele foot provide evidence that another species was present, but these fossils are floating in a taxonomic limbo because foot bones and footprints are not conventionally used to define species. A number of species names are directly attached to skulls, jaws and teeth from middle Pliocene sites in eastern Africa, but researchers have questioned whether some or all of these fossils are distinct enough from A. afarensis to be considered different species. Palaeoanthropologists have debated how to link footprints, foot bones and species names ever since early hominins were discovered in eastern Africa. The state of the current fossil record ensures that this debate will continue for years to come.

Nature 600, 388-390 (2021)



  1. 1.

    McNutt, E. J. et al. Nature 600, 468–471 (2021).

    Article  Google Scholar 

  2. 2.

    Leakey, M. D. & Harris, J. M. Laetoli. A Pliocene Site in Northern Tanzania (Clarendon, 1987).

    Google Scholar 

  3. 3.

    Tuttle, R. H., Webb, D. M. & Tuttle, N. I. in Origine(s) de la Bipédie chez les Hominidés (eds Coppens, Y. & Senut, B.) 187–198 (CNRS, 1991).

    Google Scholar 

  4. 4.

    Tuttle, R. H., Webb, D. M., Tuttle, N. I. & Baksh, M. in Topics in Primatology Vol. 3 (eds Matano, S., Tuttle, R. H., Ishida, H. & Goodman, M.) 221–242 (Univ. Tokyo Press, 1992).

    Google Scholar 

  5. 5.

    Wood, B. & Boyle, E. K. Am. J. Phys. Anthropol. 159, S37–S78 (2016).

    PubMed  Article  Google Scholar 

  6. 6.

    Kimbel, W. H. in Handbook of Paleoanthropology (eds Henke, W. & Tattersall, I.) 2071–2105 (Springer, 2015).

    Google Scholar 

  7. 7.

    Haile-Selassie, Y. et al. Nature 483, 565–569 (2012).

    PubMed  Article  Google Scholar 

  8. 8.

    Lovejoy, C. O., Latimer, B., Suwa, G., Asfaw, B. & White, T. D. Science 326, 72 (2009).

    Article  Google Scholar 

Download references

Competing Interests

The author declares no competing interests.


Nature Careers


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing


Quick links