Purussaurus

Reconstruction of Purussarus neivensis attacking Pericotoxodon platignathus. (Credit: Miguel Hernandez)

In A Nutshell

  • Fossilized bite marks on four ancient animal bones from Colombia show the first direct physical evidence that giant prehistoric crocodiles hunted large plant-eating mammals.
  • The likely attacker was Purussaurus, a crocodile estimated at 1,793 kilograms and nearly seven meters long, far bigger than today’s largest living crocodile.
  • One fossil shows marks from two different predators, suggesting Purussaurus made the kill and a smaller, land-dwelling crocodile relative scavenged the remains afterward.
  • Researchers matched the bite marks to Purussaurus teeth and calculated bite force estimates nearly twice as strong as a modern saltwater crocodile’s.

Millions of years ago in Colombia, something bit clean through the skull of an animal nearly the size of a rhino. Scientists just found out what it was.

A new study in the Journal of Vertebrate Paleontology presents the first direct physical evidence that giant prehistoric crocodiles actively hunted large, plant-eating mammals at La Venta, a site in Colombia’s Tatacoa Desert. Researchers examined four fossils from three large plant-eating species and found tooth punctures, crushing damage, and dragging marks in the bone, most pointing to one likely attacker: a giant crocodile of the genus Purussaurus, estimated to weigh around 1,793 kilograms and stretch nearly seven meters long.

Today’s largest living crocodile, the saltwater crocodile, tops out around 1,150 kilograms. Purussaurus was heavier, longer, and, judging by these fossils, hunted animals far bigger than any modern crocodile dares to.

A Punctured Skull and a Matching Leg Bone

Researchers analyzed four specimens held across museum collections in Colombia, France, and the United States. Each came from La Venta, one of the most richly preserved prehistoric ecosystems in South America, home to 156 identified species of ancient vertebrates, including nine crocodile relatives.

Among the finds was a partial skull from a large, rhino-sized hoofed animal called Pericotoxodon platignathus, estimated to have weighed between 931 and 982 kilograms. Researchers found a nearly circular puncture almost 30 millimeters across on top of that skull, a signature consistent with the ridged teeth of a large crocodile. A partial upper leg bone, likely from the same species but a smaller individual weighing around 616 kilograms, showed five additional punctures surrounded by radiating cracks, evidence that enormous force had hit fresh bone rather than bone that had already fossilized.

Purussaurus
Fossil bite marks reveal a giant crocodile hunted ton-sized mammals in Miocene Colombia, the first direct proof of its kind. (Credit: Oscar Wilson)

Bite Force Calculations Point Straight to Purussaurus

Pinning down which ancient predator made the marks meant comparing their size and shape against every possible candidate in the La Venta ecosystem. Prehistoric South America had no lions, no wolves, no large bear-like animals at this time. Instead, the dominant meat-eaters included pouched mammals similar to marsupials, large flightless terror birds, and multiple crocodile relatives. None had teeth large or round enough to match except Purussaurus, whose thick, cone-shaped front teeth sometimes exceed 30 millimeters across at the base. Researchers compared the puncture dimensions against 22 isolated Purussaurus teeth from museum collections and found a consistent match, then used the size of the punctures to work backward and estimate how hard the animal had to bite to leave them. The strongest estimate landed close to the bite force previously calculated for a related giant crocodile species, and roughly twice what a modern saltwater crocodile can manage.

A Second Crocodile Species Left Its Own Marks on the Carcass

Two jaw bones from separate elephant-like hoofed animals called astrapotheres carried deep punctures, bone crushing, and, in one case, a sharply curved scratch consistent with the violent twisting motion modern crocodiles use when subduing prey. But not all the damage on these jaws came from a single feeding event. On one astrapothere jaw, researchers identified two distinct types of marks that could not have come from the same animal. On the inner side: large, deep punctures matching a Purussaurus attack. On the outer side: a small oval depression filled with ten thin, parallel scratches pointing to something different, a smaller, land-dwelling crocodile relative with blade-like, serrated teeth, most likely Langstonia huilensis, a roughly 4.2-meter-long hunter or scavenger that walked on land, somewhat like a modern Komodo dragon. Researchers interpret this as a two-stage event: Purussaurus likely attacked the animal, and Langstonia later fed on its remains.

Three of the four specimens showed bite marks concentrated around the head and jaw, the same spot modern crocodiles target when ambushing large prey near water. None of the bite marks showed signs of healing, suggesting the damage occurred around the time of death rather than during an earlier encounter the animals survived. Combined with where the marks land, the authors lean toward predation as the likeliest explanation, while acknowledging that certainty is hard to reach from fossils alone.

Purussaurus
Jawbone of Xenastrapotherium kraglievichi that features marks of both predation by Purussaruus neivensis and possible scavenging by Langstonia huilensis. (Credit: Oscar Wilson)

Giant Crocodiles Ruled Prey That Modern Predators Never Faced

Before North and South America joined and animals could migrate between continents, South American ecosystems worked very differently from today’s. Large plant-eaters were abundant, but their predators were not wolves or big cats. Crocodile relatives, terror birds, and pouched predators filled that role instead. Something similar still happens today: Nile and saltwater crocodiles regularly prey on large hoofed animals, including cattle, water buffalo, and in documented cases, hippos and black rhinos. Purussaurus appears to have filled that same niche at La Venta, only at a scale that dwarfs anything alive now. This study offers rare, direct fossil proof of how that relationship played out, recorded permanently in bone that has survived twelve million years to reveal not just who was eating whom, but how.


Paper Notes

Limitations

Researchers acknowledge several important constraints on their conclusions. Distinguishing predation from scavenging using fossil evidence alone is inherently difficult, as the authors note, because the animal cannot recover from either. While anatomical location of bite marks and comparison to modern crocodile behavior support predation as the most likely explanation in most cases, the authors cannot definitively rule out post-mortem scavenging for some specimens. Body mass estimates for the prey animals, derived from tooth measurements and bone dimensions using previously published mathematical formulas, are acknowledged as potentially producing overestimates, particularly for the notoungulate and astrapothere species. Bite force calculations presented in the study are described as conservative minimum estimates based on bone compression strength, not precise measurements of actual force applied. Researchers also note that no systematic search for bite marks across the entire La Venta fossil collection has been conducted, meaning the frequency of such interactions in this ecosystem remains unknown. Additionally, the specimens were identified incidentally rather than through a structured survey, which may influence how representative the findings are.

Funding and Disclosures

Lead author Oscar E. Wilson was funded by a Research Council of Finland grant for a project described as studying non-analogue ecosystems of the past, grant number 340775/1346292. Co-author Jorge W. Moreno-Bernal was supported by a French National Research Agency grant, ANR-24-ERCS-007-01. Co-author John-Paul Zonneveld received research support from an NSERC Discovery Grant, RGPIN 2021-04054. No potential conflicts of interest were reported by any of the authors.

Publication Details

Authors: Oscar E. Wilson, Jorge W. Moreno-Bernal, Catalina Suarez, Frank Cardozo Tovar, Andrés Vanegas, Rubén Vanegas, Ethan A. Shirley, Patricia A. Holroyd, and John-Paul Zonneveld | Institutional affiliations include: University of Helsinki (Finland); Universidad del Norte and Muséum national d’Histoire naturelle (Colombia/France); Instituto Argentino de Nivología, Glaciología y Ciencias Ambientales (Argentina); Museo Paleontológico de Villavieja and Museo de La Tatacoa (Colombia); University of Michigan; University of California, Berkeley; and University of Alberta (Canada). | Journal: Journal of Vertebrate Paleontology | Paper title: “Physical evidence of crocodylian predation on ungulates in the Middle Miocene of Colombia” | DOI: https://doi.org/10.1080/02724634.2026.2701154 | Published online: August 16, 2026 | Submitted: October 17, 2025; accepted June 22, 2026


About StudyFinds Analysis

Called "brilliant," "fantastic," and "spot on" by scientists and researchers, our acclaimed StudyFinds Analysis articles are created using an exclusive AI-based model with complete human oversight by the StudyFinds Editorial Team. For these articles, we use an unparalleled LLM process across multiple systems to analyze entire journal papers, extract data, and create accurate, accessible content. Our writing and editing team proofreads and polishes each and every article before publishing. With recent studies showing that artificial intelligence can interpret scientific research as well as (or even better) than field experts and specialists, StudyFinds was among the earliest to adopt and test this technology before approving its widespread use on our site. We stand by our practice and continuously update our processes to ensure the very highest level of accuracy. Read our AI Policy (link below) for more information.

Our Editorial Process

StudyFinds publishes digestible, agenda-free, transparent research summaries that are intended to inform the reader as well as stir civil, educated debate. We do not agree nor disagree with any of the studies we post, rather, we encourage our readers to debate the veracity of the findings themselves. All articles published on StudyFinds are vetted by our editors prior to publication and include links back to the source or corresponding journal article, if possible.

Our Editorial Team

Steve Fink

Editor-in-Chief

John Anderer

Associate Editor