The largest dinosaurs — sauropods like Argentinosaurus and Patagotitan — reached masses of 70–80 tonnes, dwarfing every land animal before or since. Several interlocking biological systems made this possible.

Bird-like air sac systems were the key. Like modern birds, large dinosaurs had lungs connected to a network of air sacs that extended into hollow bones throughout the skeleton. This meant up to 60% of their body volume was air — drastically reducing effective weight while maintaining a continuous, highly efficient one-way airflow through the lungs.

Hollow pneumatised bones gave structural strength at a fraction of the weight. A Brachiosaurus vertebra could be the size of a refrigerator yet weigh only a few kilograms, honeycombed with air pockets like a bird's bones.

Fast growth rates played a major role too. Analysis of bone growth rings shows large sauropods gained up to 2 tonnes per year during adolescence — growth rates more comparable to whales than reptiles. They reached full size in 20–30 years and simply kept the machinery running at scale.

Dinosaur fossils have been discovered on every continent including Antarctica, collected by thousands of independent scientists across two centuries. The evidence is layered, cross-referenced, and impossible to fabricate consistently at that scale.

Fossilised bones are the most direct evidence — mineralised originals or permineralised replacements where original calcium phosphate has been replaced atom-by-atom by silicate or carbonate minerals over millions of years. Tens of thousands of specimens exist in museum collections worldwide.

Eggs and nests have been found on every inhabited continent, many containing embryos still articulated inside the shell — including feathered embryos showing the direct link to modern birds.

Footprints and trackways preserve actual behaviour: we can see herds moving together, predators stalking prey, juveniles running alongside adults. Some trackways stretch for hundreds of metres.

The geological context is independently verified: the rock layers containing dinosaur fossils are dated by radiometric methods that consistently place non-avian dinosaurs between 230 and 66 million years ago.

The simple answer: most were neither — or more accurately, both. Modern palaeontology has moved well beyond the binary of reptile-cold vs mammal-warm.

Bone microstructure is the strongest evidence. Warm-blooded animals grow rapidly and continuously, producing fibrolamellar bone — a distinctive fast-growth tissue full of blood vessels. Dinosaur bone is overwhelmingly fibrolamellar, indicating rapid, continuous growth — a warm-blooded metabolic strategy.

Polar dinosaurs provide another clue. Species have been found in polar environments with months of winter darkness. Cold-blooded animals cannot maintain activity in such conditions. These dinosaurs show no sign of hibernation in their bone records.

Current consensus is that non-avian dinosaurs were mesotherms: faster metabolisms than modern reptiles, perhaps somewhat lower than modern birds and mammals — a middle strategy that worked exceptionally well for 165 million years.

Elaborate structures like Triceratops' frill and horns, Parasaurolophus' hollow crest, and Stegosaurus' back plates puzzled early palaeontologists. Modern analysis points to several overlapping functions.

Species and individual recognition is the most supported function. Animals needed to identify their own species quickly across varied habitats — elaborate headgear functioned like a species badge.

Sexual selection drove much of the extreme elaboration. Just as peacock tails are absurdly large because peahens choose the showiest males, dinosaur crests and frills grew beyond practical function because individuals with more impressive displays left more offspring.

Parasaurolophus' hollow crest was almost certainly a resonating chamber for vocalisation — CT scans reveal complex internal passages that would have produced a low, resonant call, consistent with long-distance communication in herds.

Palaeontologists don't guess at ages — they use multiple independent physical and chemical methods that cross-check each other.

Radiometric dating is the gold standard. Radioactive isotopes decay at precisely known rates. Uranium-238 decays to lead-206 with a half-life of 4.5 billion years. By measuring the ratio of parent to daughter isotope in volcanic rock surrounding a fossil, scientists calculate how long ago that rock formed.

Biostratigraphy uses the known ranges of index fossils — species with well-established first and last appearances — to bracket the age of associated finds. Cross-continental correlation of these markers provides a global chronological framework.

Magnetostratigraphy exploits reversals of Earth's magnetic field recorded in iron-bearing minerals. The sequence of normal and reversed polarity zones is known globally, providing an independent time scale.

Speed estimates come from multiple independent lines of evidence that give surprisingly consistent results across different methods.

Trackways are the most direct evidence of actual locomotion. By measuring stride length and estimating hip height from footprint size, palaeontologists can calculate minimum speeds using biomechanical formulae. Some theropod trackways indicate sustained speeds of 12 km/h — roughly a fast human jog.

Bone geometry predicts locomotor performance. Longer, more slender limb bones indicate cursoriality (running adaptation). Ornithomimosaurs had the most cursorial proportions of any large dinosaur — estimates suggest top speeds around 50 km/h.

Computer modelling of muscle attachment scars and bone stress limits sets hard ceilings. For T-Rex, models suggest top speeds of 17–25 km/h — fast enough to catch most large prey, but far from the 50 km/h sometimes claimed in popular culture.

Yes — not just some of them, but probably most theropods and possibly many other dinosaur groups. The evidence is now overwhelming and comes from multiple continents.

Direct fossil preservation in fine-grained Lagerstätten — deposits where exceptional preservation occurs — has produced hundreds of feathered dinosaur specimens from the Early Cretaceous Yixian Formation in Liaoning, China. These show not just feather impressions but actual pigment-bearing melanosomes, allowing partial colour reconstruction.

Quill knobs on the forearms of some Velociraptor specimens are anchor points for flight feathers — identical structures on modern birds. This means even specimens without preserved feathers can show evidence of them through skeletal attachment points.

The evolutionary logic is compelling: birds are theropod dinosaurs (a fact with overwhelming support), so feathers evolved within the dinosaur lineage. The question is not whether non-avian theropods had feathers, but how widespread and complex those feathers were.

The end-Cretaceous extinction 66 million years ago was caused by the Chicxulub bolide impact — a 10–15 km asteroid or comet that struck what is now the Yucatán Peninsula, Mexico. This is established beyond reasonable scientific doubt.

The impact evidence is global: a thin layer of iridium (rare in Earth's crust but common in asteroids) marks the Cretaceous-Palaeogene boundary in rock exposures worldwide. The Chicxulub crater itself is 180 km in diameter. Shocked quartz and glass spherules from the impact are found globally in the boundary layer.

The kill mechanisms were multiple and overlapping: an immediate thermal pulse from ejecta re-entering the atmosphere; a global dust and soot cloud blocking sunlight for months to years, collapsing photosynthesis; sulfur aerosols causing acid rain and further cooling; and wildfires ignited by the initial thermal pulse.

The Deccan Traps — massive volcanic eruptions in what is now India — were also underway and contributed to environmental stress, but the current consensus is that the impact was the primary cause of the abrupt, geologically instantaneous mass extinction.