More than ten million years ago, South America's food chain had a ruler, and it wasn't a mammal. A study published August 16, 2026, in the Journal of Vertebrate Paleontology used bite marks preserved directly on fossilized bone to demonstrate that giant crocodylians, not the saber-toothed mammalian predators sharing their habitat, sat at the true top of the Miocene South American food web.

A continent covered by a vanished mega-wetland

During the Miocene, a vast lake and wetland system, sometimes referred to as proto-Amazonia or the Pebas Mega-Wetland System, covered an enormous stretch of the South American continent. That environment supported an extraordinarily rich community of large animals, including giant turtles, an entire assemblage of crocodylian species, and a diverse cast of predators hunting along its shores: anacondas, saber-toothed marsupial and placental predators, flightless "terror birds" known as phorusrhacids, and sebecids, a group of terrestrial, land-dwelling relatives of crocodylians that had evolved to hunt on foot rather than in water.

With that many predator species competing within a single ecosystem, determining which one actually functioned as the apex predator, the animal sitting at the genuine top of the food chain, has remained a genuinely difficult question for paleontologists to answer using fossil evidence alone.

Why direct evidence of predation is so hard to find

Oscar Wilson, a postdoctoral researcher at the University of Helsinki who led the research, explained the fundamental challenge underlying this kind of study: direct behavioral observation is obviously impossible across a gap of ten to sixteen million years, meaning researchers have to rely on far more indirect forms of evidence to reconstruct which animals were actually eating which other animals in a given ancient ecosystem. Bite and tooth marks preserved directly on herbivore fossils represent one of the only forms of genuinely direct evidence available for reconstructing these ancient predator-prey relationships, since they record an actual physical interaction between two specific animals rather than simply inferring diet from anatomy or general ecological reasoning.

To pursue that evidence, Wilson's research team analyzed fossil specimens held in museum collections, dating to between 10.5 and 16 million years old, drawn primarily from the La Venta fossil assemblage in Colombia, one of the richest and most thoroughly studied Miocene fossil sites in South America.

What the bite marks actually showed

The research team examined four specimens bearing clear evidence of predation: a skull, a leg bone, and two jawbones, each showing distinctive marks consistent with a massive crocodylian bite. According to Sci.News's detailed coverage, researchers classified these marks using established trace-fossil categories, including Nihilichnus, Linichnus, Knethichnus, and Brutalichnus, formal designations paleontologists use to describe specific types of bite trace patterns based on their shape and the kind of damage they left in the bone.

One of the clearest examples came from a skull belonging to Pericotoxodon platignathus, a large native South American ungulate. A circular puncture on the skull's frontal bone measured 29.2 millimeters across with two opposing vertices, a mark researchers classified specifically as Nihilichnus hastarius. Both the puncture's size and shape matched the teeth of Purussaurus neivensis, described by researchers as the largest crocodyliform known from the entire La Venta fossil assemblage.

Meet the actual apex predator

Purussaurus neivensis, a giant relative of modern caimans, reached approximately seven meters in length and weighed roughly 1,800 kilograms, dimensions researchers have compared to the length of a school bus. According to IndianDefenceReview's coverage of the study, the specimens examined showed punctures, pits, fractures, grooves, and areas of crushed cortical bone, all consistent with crocodylian feeding behavior specifically, rather than damage that could plausibly be attributed to any of the ecosystem's mammalian or avian predators instead.

Wilson framed the underlying ecological logic behind why a reptilian predator, rather than the array of mammalian carnivores sharing the same habitat, ultimately dominated: "There was so much prey available that mammalian predators couldn't have eaten it all. Our new study shows that crocodylians in fact were the most important predators in this region, particularly when it comes to preying on large animals."

A predator massive enough to shape entire populations

Beyond simply identifying Purussaurus as a dominant hunter, the study's authors suggest its ecological role may have extended considerably further, potentially influencing the actual population dynamics of the herbivore species it preyed upon. Given the animal's sheer size and the frequency of bite marks attributable to it found across the examined fossil sample, researchers propose Purussaurus may have functioned as a genuine keystone predator, one whose hunting pressure helped regulate herbivore population sizes across the broader wetland ecosystem, a role comparable in ecological function, if considerably larger in physical scale, to how modern apex predators like lions or wolves can shape prey populations in their own respective habitats today.

Wilson noted that the frequency of marks specifically attributable to Purussaurus across the examined fossil sample supports this broader interpretation, indicating the animal wasn't simply an opportunistic predator taking advantage of already-weakened or dying prey, but an active, important hunting force actually shaping the composition and size of the large herbivore populations sharing its wetland habitat.

Not just a hunter, but a provider for other predators too

The research also points to a secondary ecological role Purussaurus likely played within this ancient food web: providing scavenging opportunities for other predators lower in the food chain. According to Nautilus's coverage of the findings, sebecids, the terrestrial crocodylian relatives that hunted on land rather than in water, likely benefited from carcasses and remains left behind after Purussaurus had fed, suggesting the giant crocodylian's kills helped support a broader scavenging economy across the ecosystem, beyond simply removing prey animals from the population through direct predation alone.

A predator whose reputation keeps growing with each new study

Purussaurus has drawn increasing scientific attention in recent years as researchers have progressively documented just how physically extreme the genus actually was. Separate biomechanical research on a related species, Purussaurus brasiliensis, calculated a bite force of approximately 69,000 newtons, or roughly 15,500 pounds of force, a figure researchers have described as more than double the bite force estimated for Tyrannosaurus rex, and among the most powerful bites ever documented for any four-limbed animal in the fossil record. That prior biomechanical work, combined with this new study's direct fossil evidence of the animal's actual predatory activity within a specific ecosystem, gives researchers an unusually complete picture connecting Purussaurus's raw physical capability to its actual, observable ecological role.

Why this challenges assumptions about ancient food chains generally

The study's broader significance lies partly in what it reveals about a common assumption in paleontology: that mammals, having eventually come to dominate most of Earth's major terrestrial ecosystems following the extinction of non-avian dinosaurs, would naturally have occupied the top predator role in most ancient environments researchers study. This research demonstrates that assumption doesn't hold universally, at least not within this specific Miocene South American wetland ecosystem, where reptilian crocodylians, rather than any of the ecosystem's various mammalian predators, occupied the genuine apex position.

That finding adds to a broader and still developing scientific picture of South America's unusual evolutionary history during this period, a continent that remained geographically isolated from North America for millions of years before the eventual formation of the Isthmus of Panama, allowing distinctly different evolutionary experiments, including this crocodylian-dominated predator hierarchy, to play out compared to contemporaneous ecosystems on other continents where mammalian predators more typically held the top position.

What this means for understanding Miocene South America going forward

With direct fossil evidence now establishing Purussaurus neivensis as the genuine apex predator within this specific proto-Amazonian wetland ecosystem, researchers have a considerably clearer foundation for reconstructing the broader food web dynamics that shaped Miocene South America more generally. Understanding precisely which species held predatory dominance, and how that dominance shaped surrounding herbivore populations, offers paleontologists a more complete picture of how this now-vanished mega-wetland ecosystem actually functioned, more than ten million years before the modern Amazon basin, and its considerably different ecological hierarchy, eventually took its place.

---

*Sources cited in this article include the peer-reviewed study "Physical evidence of crocodylian predation on ungulates in the Middle Miocene of Colombia," published August 16, 2026, in the Journal of Vertebrate Paleontology, and reporting from ScienceDaily, Phys.org, EurekAlert, Sci.News, IndianDefenceReview, Nautilus, and the University of Helsinki covering research led by Dr. Oscar Wilson. All figures reflect reporting available as of August 19, 2026.*