Palaeontology: Fossil Formation
Why Almost Nothing Becomes a Fossil
Every fossil you have ever seen represents a creature that won, against odds so astronomical they make a lottery ticket look like a sure bet, the most improbable game in the history of life on Earth.
The Idea
The received story of fossilisation — creature dies, gets buried, minerals replace bone, millions of years pass, palaeontologist finds it — is technically accurate and almost completely misleading. It implies a process that, while slow, is reasonably reliable. It is not. The vast majority of organisms that have ever lived left absolutely no trace. Soft tissue rots within days. Bones are scavenged, crushed, dissolved by acidic groundwater. Even hard shells and teeth — nature's most durable materials — are usually destroyed by the very geological processes that might otherwise preserve them. For a fossil to form, an organism must die in exactly the right place: somewhere with rapid sediment deposition, low oxygen, and the right chemical conditions to mineralise tissue before decay wins. It then has to survive millions of years of tectonic movement, heat, pressure, and erosion — and finally, it has to erode out of rock at precisely the moment a human happens to be looking at that hillside. This is why the fossil record is not a blurry photograph of prehistoric life. It is more like a handful of frames from a film that ran for four billion years — and those frames are heavily biased toward creatures that lived in shallow seas, had hard parts, and happened to die near a riverbed. Everything else — most creatures, most ecosystems, most evolutionary experiments — is simply gone.
In the World
In 2007, palaeontologist Mary Schweitzer published findings that sent a ripple through the field: she had recovered what appeared to be soft tissue — flexible, translucent structures resembling blood vessels — from inside the femur of a Tyrannosaurus rex that had died 68 million years ago. The scientific community was immediately sceptical. Soft tissue surviving for 68 million years contradicted everything understood about organic preservation. Schweitzer's discovery held up to scrutiny, and subsequent work suggested that iron from haemoglobin may act as a fixative, effectively embalming tissue at the cellular level — a chemical fluke that turns out to occur under very specific conditions. What makes this story instructive is not just the discovery itself but what it reveals about our ignorance: palaeontologists had been cracking open dinosaur bones for over a century and not thinking to look inside at the fine structure, because everyone knew soft tissue couldn't survive. The assumption shaped the search. Schweitzer looked partly because she was a former medical technician, trained to notice biological material. This one exception — one animal, one bone, one unusual set of chemical conditions — has now opened an entire subfield of molecular palaeontology. It is a reminder that what we know about prehistoric life is shaped as much by what we think to look for as by what is actually there.
Why It Matters
There is a subtle but important cognitive habit that fossil formation can teach you, and it has nothing to do with dinosaurs. It is about understanding the difference between a record and reality — recognising when the data you have access to is a heavily filtered sample of a much larger truth. The fossil record's gaps are not random noise; they are systematic. Creatures with hard parts are overrepresented. Ocean floor fauna dominate. Tropical forests, which harbour extraordinary biodiversity, almost never fossilise because their warm, wet conditions accelerate decay. Once you see this, you start noticing the same pattern elsewhere: the historical record privileges the literate and the powerful; medical research has long overrepresented certain demographics; economic data captures market transactions but not unpaid labour. The question to ask of any record — fossil, historical, statistical — is not just what it shows, but what conditions determined what survived. The gaps are not absences of information. They are information.
A Question to Ponder
If the creatures best suited to leaving a fossil record are not necessarily the most representative of life in their time, what else do you rely on — in history, in data, in memory — that might be giving you a systematically skewed picture of what actually happened?
Get a new one of these every morning.
Start learning with Thinkable