The bone came from a famous fossil site in South Dakota, USA. It is one of the best preserved Edmontosaurus skeletons ever found, with its outer skin texture still visible — that is why it is sometimes called a 'dinosaur mummy.'
The team used four different lab methods, including powerful microscopes, mass spectrometry, and protein sequencing. They detected pieces of collagen, the main protein in animal bones. They also found hydroxyproline, a special amino acid that almost only appears in collagen — strong evidence that the protein is real and not modern dirt.
For a long time, most scientists believed proteins fall apart within a few million years. This new paper, published in Analytical Chemistry on May 14, suggests that the right rock and water conditions can protect tiny protein fragments for far longer. It does not mean dinosaur DNA will be next — DNA is much fragile — but it widens the window for studying real dinosaur chemistry.
A multi-institution team led by the University of Liverpool reported on May 14 that it had recovered fragments of original collagen from a 66-million-year-old Edmontosaurus fossil — the same duck-billed plant-eater that browsed North America's last forests just before the Cretaceous–Paleogene extinction. The peer-reviewed paper appeared in the American Chemical Society journal Analytical Chemistry under the title 'Evidence for Endogenous Collagen in Edmontosaurus Fossil Bone.'
The specimen comes from the Hell Creek Formation of South Dakota and is one of the famous 'dinosaur mummy' Edmontosaurus, fossils preserved so well that skin texture, scale patterns, and even soft-tissue impressions remain visible. That extraordinary preservation made it a natural target for protein hunting, but the field has been burned before: previous claims of dinosaur collagen, from a 2007 T. rex paper onward, were repeatedly challenged on contamination grounds.
The new study layers four lines of evidence to defuse that worry. Optical and electron microscopy reveal fibrous, cell-shaped microstructures consistent with collagen scaffolding. ToF-SIMS mass spectrometry maps amino-acid fragments inside those microstructures rather than on their surfaces. A second mass-spec method — peptide LC-MS/MS at UCLA — independently picks up hydroxyproline, an amino acid that is essentially exclusive to vertebrate collagen and is not produced by modern bacterial or fungal contaminants. Finally, protein sequencing returns short peptide reads consistent with hadrosaurid collagen but inconsistent with any laboratory species used by the team.
The implications are conservative but real. The paper does not claim recoverable DNA, which decays orders of magnitude faster than protein. What it does suggest is that, under the right combination of rapid burial, low oxygen, mineral cementation and stable groundwater chemistry, fragmentary proteins can persist for tens of millions of years longer than the textbook 'few-million-year' ceiling. Several outside paleontologists called the work 'careful' and 'finally convincing,' and a follow-up by the Field Museum on a separate Edmontosaurus is already in review.
An Anglo-American team led by the University of Liverpool and including collaborators at UCLA and the Field Museum reported in Analytical Chemistry on May 14 that it has recovered convincing chemical signatures of endogenous collagen from a 66-million-year-old Edmontosaurus annectens specimen sourced from the Hell Creek Formation of South Dakota. The paper — titled 'Evidence for Endogenous Collagen in Edmontosaurus Fossil Bone' — pushes the empirical ceiling for biomolecular persistence past the Cretaceous–Paleogene boundary itself, a result the field has been chasing, and disputing, for nearly two decades.
The political history of such claims looms over the new work. After Mary Schweitzer's 2007 announcement of collagen peptides in a Hell Creek Tyrannosaurus, a series of replies argued that the spectra reflected contamination from modern ostrich, biofilm bacteria, or simple polysaccharide artefacts. The Liverpool group, led by analytical geochemist Stephen Taylor with hadrosaur specialist Phil Manning, designed the present study explicitly to anticipate every such objection by stacking four independent techniques on a single field-jacketed specimen excavated under low-contamination protocols.
Microscopy reveals fibrous, cellular-scale microstructures whose periodic banding is consistent with cross-linked Type I collagen. Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS) at Manchester maps amino-acid mass fragments inside, rather than on the surface of, those microstructures. UCLA's tandem LC-MS/MS picks up hydroxyproline — an amino acid essentially confined to vertebrate collagen, with no known microbial or fungal source. And a fourth pass, top-down peptide sequencing, returns short peptide reads whose alignment is incompatible with any species kept in the Liverpool laboratory environment. The cumulative case is harder to dismiss than any earlier single-technique paper.
The biochemical implications are bounded but consequential. Free collagen has a calculated equilibrium half-life of roughly 0.18 million years under standard wet conditions; mineral cementation, low-oxygen burial in iron-rich sediments, and isotopically stable groundwater can multiply that figure by orders of magnitude. The data force the field to abandon a textbook ceiling but stop well short of suggesting recoverable DNA or de-extinction. As Caltech's vertebrate paleontologist Hans Larsson, who was not involved in the work, put it for a Reuters reader: 'We are not getting Edmontosaurus back. We are getting Edmontosaurus chemistry back — and that's already a richer dataset than we expected to have in our lifetimes.'
A team led by the University of Liverpool, with collaborators at UCLA and the Field Museum, has detected fragments of original collagen — and the bone-specific amino acid hydroxyproline — inside an exceptionally preserved Edmontosaurus from South Dakota. The Analytical Chemistry paper, published May 14, uses four independent techniques to rule out modern contamination and undermines the long-held assumption that proteins cannot survive 66 million years.

Dinosaurs lived a very long time ago. They died many millions of years ago. Now we only have their old bones.
Scientists in England found something new. They looked inside a duck-billed dinosaur bone. The dinosaur is called Edmontosaurus.
Inside the bone they found tiny pieces of protein. Protein is the same stuff that builds our skin and bones today.
This is amazing because most people thought protein cannot last so long. The bone is 66 million years old. The work was shared on May 14.
1What did scientists find inside the bone?
2What is the dinosaur's name?
3How old is the bone?
4Which country are the scientists from?
5What kind of dinosaur is Edmontosaurus?
6Dinosaurs are still alive today.
7Edmontosaurus is a dinosaur.
8Protein helps build our skin and bones.
9The bone is 100,000 years old.
10The news came out on May 14.
11The dinosaur's name is ___.
12The bone is ___ million years old.
13Scientists found tiny pieces of ___ in the bone.