What Did Dinosaurs Eat?

"What did dinosaurs eat?" sounds like a simple question, but the answer involves teeth, coprolites (fossilized dung), preserved gut contents, bite marks in bone, and careful inference from related living animals. Paleontologists don't guess at diet — they read the physical record. Here's what that record actually says.

How we know what dinosaurs ate

Before looking at specific diets, it helps to understand the evidence types:

• Tooth morphology. Serrated, blade-like teeth indicate meat-cutting. Flat, peg-like teeth indicate vegetation stripping. Leaf-shaped teeth with ridges indicate plant grinding. Tooth shape is the primary diagnostic for diet in most species.
• Coprolites. Fossilized feces can contain identifiable fragments of bone, plant matter, scales, or seeds. A 1998 paper describing a probable Tyrannosaurus coprolite containing crushed hadrosaur bone was the first direct physical proof that large theropods consumed bones, not just flesh.
• Gut contents. Occasionally, fossilized stomach contents are preserved. The small compsognathid Sinosauropteryx was found with a lizard preserved in its gut. Baryonyx was found with fish scales and partially digested Iguanodon bones.
• Bite marks on bone. Tooth drag marks, punctures, and scrape patterns on fossilized bones can be matched to specific predator tooth geometries. T. rex bite marks have been positively identified on Triceratops and Edmontosaurus specimens.
• Skull and jaw biomechanics. The shape of the skull, the attachment points of jaw muscles, and the leverage geometry of the jaw all constrain what a dinosaur could bite and how hard.

Carnivores: the evidence

Carnivorous dinosaurs are the group most people think of first. The classic examples are large theropods — Tyrannosaurus rex, Allosaurus fragilis, Spinosaurus aegyptiacus, Giganotosaurus carolinii — but smaller carnivores were far more numerous and ecologically important.

Tyrannosaurus rex

T. rex had banana-sized teeth with serrations designed for penetrating bone. Biomechanical studies estimate its bite force at 35,000–57,000 Newtons — the strongest of any confirmed terrestrial predator. The 1998 Sternberg Museum coprolite, 44 cm long and containing crushed Edmontosaurus bone, confirmed active predation. Healed bite marks on at least one hadrosaur specimen show that prey animals survived T. rex attacks, indicating active hunting rather than exclusive scavenging.

Spinosaurus aegyptiacus

Spinosaurus teeth are round in cross-section and conical — the classic shape for fish-eating, identical to modern gharials and herons. Baryonyx, a close relative, was found with fish scales and partially digested fish bones in its gut. Spinosaurus isotope analysis also suggests it spent significant time in water, consistent with piscivory (fish-eating) as a primary strategy.

Velociraptor mongoliensis

The "Fighting Dinosaurs" specimen, discovered in Mongolia in 1971, shows Velociraptor locked in combat with a Protoceratops. The raptor's enlarged second toe claw is embedded in the Protoceratops neck region. This specimen is direct evidence of predator-prey interaction, fossilized mid-conflict. Velociraptor was an active predator of animals roughly its own size or larger.

Herbivores: the evidence

By species count, herbivores outnumbered carnivores significantly — the same ratio we see in modern ecosystems. Sauropods, hadrosaurs, ceratopsians, stegosaurs, and ankylosaurs were all herbivores, occupying different ecological niches.

Sauropods (Brachiosaurus, Diplodocus, Patagotitan)

Sauropods had simple, peg-like or spatula-shaped teeth that were not designed for chewing — they stripped vegetation and swallowed it whole, relying on their enormous guts for fermentation. Diplodocus teeth are raked forward, ideal for combing foliage off branches. Brachiosaurus teeth are spoon-shaped, suited for cropping. Gastroliths (stomach stones) found with some sauropod specimens may have aided digestion, similar to modern birds.

Hadrosaurs (Edmontosaurus, Parasaurolophus)

Hadrosaurs had some of the most sophisticated chewing systems of any dinosaur — dental batteries of hundreds of closely packed teeth that ground plant material efficiently. Preserved gut contents from mummified hadrosaur specimens from Canada include seeds, needles, twigs, and conifer fragments. Their diet was varied and included tough, fibrous Cretaceous vegetation.

Triceratops horridus

Triceratops had a parrot-like beak and shearing cheek teeth. The beak cropped vegetation; the cheek teeth sliced it. Biomechanical modeling suggests Triceratops was capable of processing tough, low-growing plants — possibly palm fronds or cycads — that other herbivores couldn't manage as efficiently.

Omnivores: the evidence

True omnivory is harder to establish from fossil evidence than pure carnivory or herbivory, but several dinosaur groups show anatomical and preserved-content evidence of mixed diets.

Oviraptor philoceratops

Oviraptor ("egg thief") was named under the assumption it was stealing Protoceratops eggs. Later research found it was actually brooding its own nest. Its toothless beak and crushing jaw structure suggest a diet of hard food items — seeds, shellfish, and possibly eggs — making it functionally omnivorous.

Ornithomimosaurs (Gallimimus, Struthiomimus)

Ornithomimosaurs were ostrich-like toothless theropods with filter structures in their beaks, similar to modern flamingos. Analysis of their skull structure suggests they may have filtered small invertebrates and plant material from water. One Canadian specimen preserved gastroliths consistent with grinding plant matter, while isotope analysis hints at mixed diets varying seasonally.

Talking about dinosaur diet with kids

The question "what did this dinosaur eat?" is a natural entry point into evidence-based thinking. Ask a child to look at the teeth on a museum mount and predict the diet before reading the label. Flat and ridged? Probably plants. Serrated and blade-like? Probably meat. Conical and smooth? Possibly fish. This is exactly how paleontologists approach a new specimen — and it works as a conversation with a five-year-old.