A T. rex tooth stuck in a duck-billed skull rewrites a small corner of predator-prey science
A Cretaceous-era skull with a tooth still lodged in its jawbone offers the first direct physical evidence of a T. rex attack on an Edmontosaurus, and a glimpse of who ate whom 67 million years ago.
On 14 July 2026, palaeontologists reported a piece of evidence that almost never survives 67 million years: a tyrannosaur tooth still embedded in the skull of the animal it was biting. The fossilised jaw of an Edmontosaurus, a duck-billed herbivore that shared the Late Cretaceous floodplains of North America with the apex predator Tyrannosaurus rex, carries a broken tooth crown lodged in its cheekbone, along with a healing scar that suggests the plant-eater lived, briefly, after the encounter. The find, announced this week, is being treated as the clearest direct physical evidence yet that T. rex actively hunted large ornithischians rather than only scavenging them.
The discovery matters because the predator-versus-scavenger debate has shadowed T. rex research for decades. Bite marks on prey bone have always existed; what they have not done, until now, is pin a tyrant's tooth inside a living victim's face at the moment of attack. The new specimen narrows the inferential gap: a hunting event is harder to dismiss as opportunistic carcass-feeding when the weaponry is still attached to the wound.
What the skull actually shows
The Edmontosaurus specimen preserves two injuries that researchers read as a sequence. First, the embedded tyrannosaur tooth, wedged into the jugal bone on the side of the face, with fracture patterns consistent with the kind of lateral bite T. rex is known to deliver. Second, bone remodelling around the wound: signs that the animal's immune system responded, the tissue healed, and the duck-bill survived at least weeks or months after the strike. Healing, in palaeontological terms, is the difference between a feeding trace on a dead animal and a record of a living one.
That detail reframes the long-running dispute. A T. rex that bit, lost a tooth, and watched its prey walk away is not a strict scavenger. It is also not, on this evidence alone, a confirmed pack hunter. The skull captures a single event, in a single geological moment, and the honest inference is that T. rex attacked large, living prey and sometimes failed to finish the job.
Why the scavenger hypothesis still has defenders
The dominant counter-reading, associated most prominently with the late palaeontologist Jack Horner, holds that the bone-crushing dentition and oversized olfactory bulbs of T. rex point to a carcass-specialist lifestyle. Under that view, tyrannosaurs were the Cretaceous equivalent of hyenas, dominating landscapes through opportunistic feeding rather than pursuit. Bite-mark surveys on hadrosaur and ceratopsian bones have been cited as evidence consistent with both hunting and scavenging, and the absence of direct predator-prey contact has, until now, left the argument open.
The new fossil does not demolish that position. It narrows it. A single embedded tooth in a single hadrosaur face is consistent with a hunting event; it is not, by itself, a population-level statistic about tyrannosaur behaviour. Researchers studying bite-mark frequencies across hundreds of specimens will want to know how common embedded teeth are, and how often the prey survived the bite. One data point shifts the framing; a distribution would settle it.
A structural shift in how the Late Cretaceous is read
For most of the twentieth century, dinosaur palaeobiology treated predation as inferred behaviour, reconstructed from tooth shape, biomechanical modelling, and the relative abundance of carnivore and herbivore fossils in the same formations. Behavioural claims about extinct animals rested on proxies. The Edmontosaurus skull is part of a slow shift across the discipline towards direct evidence: healed bite marks that prove a prey animal survived, stomach contents that fix a meal, footprints that capture a chase.
The same shift is visible in work on other megafauna from the period. Trackways in the western United States have recorded tyrannosaurs moving at speeds consistent with walking, not pursuit, complicating the cinematic image of a sprinting predator. Tooth-mark surveys on Triceratops frills and hadrosaur tails have begun to map which body regions tyrannosaurs targeted, and how prey responded. The new fossil sits inside that methodological turn: fewer computer models, more scars.
Stakes for the field and for the public imagination
The stakes for working palaeontologists are immediate. A direct predator-prey association between two iconic Late Cretaceous species tightens the ecological narrative of formations like the Hell Creek, where Edmontosaurus and T. rex are both abundant and where debates over community structure have run for decades. Specimens like the new skull will be cited in grant proposals, exhibition scripts, and undergraduate lectures for years.
The public stakes are less academic and more durable. T. rex is the most recognisable prehistoric animal on Earth, its image embedded in film, merchandise, and museum halls. Settling the predator-versus-scavenger question changes which story the public is told: the slow-burning carcass-economy, or the chase, the bite, the lost tooth, the herbivore that got away for a while. Either version is stranger than the Jurassic Park sprint. The evidence is now leaning, gently, towards the bite.
What remains genuinely uncertain is the frequency. A single healing skull confirms that T. rex attacked living hadrosaurs and that the prey sometimes survived; it does not, on its own, fix the ratio of hunting to scavenging across millions of years of evolution. The next step the sources point to is broader survey work, looking for more embedded teeth across museum collections, and the slow accumulation of scars into a dataset that can argue, statistically, for a lifestyle rather than a single dramatic moment frozen in stone.
This piece treats the new fossil as one specimen inside a long methodological shift in palaeontology, rather than as a definitive verdict on tyrannosaur behaviour. Where the dominant framing (active predator) and the dissenting framing (opportunistic scavenger) overlap, both have been given equal weight.
Wire provenance
This editorial synthesis draws on the following public wire/social posts:
- https://t.me/c/cluster-33bdbc063c/1
- https://en.wikipedia.org/wiki/Tyrannosaurus
- https://en.wikipedia.org/wiki/Edmontosaurus
- https://en.wikipedia.org/wiki/Hell_Creek_Formation