How to Learn About Dinosaurs

Dinosaur is a precise term rather than a general word for large extinct animals, and applying it correctly clears up an enormous amount.

Pterosaurs flew and were not dinosaurs. Plesiosaurs and mosasaurs swam and were not dinosaurs. Mammoths were not dinosaurs and missed them by over sixty million years. Dimetrodon, the sail-backed animal in every toy set, is more closely related to you than to Tyrannosaurus.

Meanwhile birds are dinosaurs, in a technical and not a poetic sense. The group did not die out; one branch of it is currently outside.

The numbers in this subject are beyond intuition, and the resulting error is treating the whole Mesozoic as one period where everything coexisted.

The correction that lands hardest: Stegosaurus and Tyrannosaurus never met, and the gap separating them is larger than the gap between Tyrannosaurus and us. Dinosaurs are separated from each other by more time than separates them from the present.

Building a rough timeline with a few anchor points, and placing everything relative to those, is the only reliable way to hold this.

Fossilisation is extraordinarily unlikely. It needs rapid burial, the right chemistry, and then survival through millions of years of geology without being destroyed. Almost every organism that has ever lived left no trace at all.

This is not a technicality; it shapes every conclusion. Species known from one partial skeleton still get drawn as complete animals, and the parts nobody has found are precisely the parts most likely to be revised. Soft-bodied organisms are systematically underrepresented, so the record over-samples things with hard parts.

Bones say what an animal was; tracks, nests, dung and bonebeds say what it did. Speed, gait, whether animals moved in groups, what they actually ate, whether they cared for young - none of that comes from a skeleton.

This is the part of the subject that has changed most, and it is where the interesting current arguments are. Nesting colonies changed how dinosaurs were understood far more than any new skeleton did.

This field turns over fast, and much of what an adult learned as a child is now known to be wrong. Feathers on many theropods. Warm-bloodedness as a spectrum rather than a binary. The whole dinosaur family tree rearranged in the last decade.

Treating the current picture as provisional is not scepticism, it is accuracy. Being able to say what is well established, what is likely and what is contested is a better answer than confidence about all three.

The end-Cretaceous extinction is the best-understood mass extinction there is and still argued about, because an asteroid impact and a massive volcanic province arrived close enough together to be difficult to separate.

Working through how the iridium layer, the crater and the volcanic record are weighed against each other is a compact lesson in how competing explanations get tested. It is also worth knowing that the end-Permian extinction killed a far larger share of species and gets a fraction of the attention.

The reason this subject is worth taking seriously past childhood is that it is a working demonstration of how much can be reconstructed from very little, and of how conclusions get revised when new evidence arrives.

There is no dinosaur DNA and there will not be. It degrades far too fast, and the molecular work is on much more durable things like collagen fragments.