The timeline corridor
Everything in the portrait hall was here first. Not slightly first — first by a margin that is difficult to hold in your head, which is why this room draws it on a wall instead of stating it.
Life has been on Earth for at least 3,480 million years. For the first 2,922 million of them — 84% of the whole history of life — every living thing on the planet was microbial. There were no animals to film, no plants, no fungi on land. There was this museum's collection, and nothing else.
How to read this wall. The empty stretches are drawn to scale: every gap between two labels is the real interval between them. The labels are not — a label takes as much wall as its words need, and time did not allocate that. So trust the distances between labels, and not the labels' own heights.
4.54 bya
Molten, airless, and still being hit by the material it was assembled from.
Everything else on this wall happens on this planet. For most of the length of it, the only things alive are the ones this museum collects.
Dated from meteorites rather than from the Earth, because the Earth recycles its own crust and has kept no rock this old. The oldest surviving Earth minerals are zircon grains a little over 4.4 billion years old.
4.28 bya – 3.77 bya
If it is life, it is life almost immediately — within a few hundred million years of the planet being cool enough to have oceans at all.
That “if” is doing real work. At this age the evidence is a shape in a mineral, and shape is the weakest kind of evidence there is.
A span, not a date, because the rocks themselves are dated to a range this wide. Haematite tubes and filaments in Quebec's Nuvvuagittuq belt have been read as the mineralised remains of vent microbes. Comparable shapes can form without life, and these rocks have been cooked and squeezed for four billion years since.
3.7 bya
We hang the claim and the rebuttal together, the way the records hall hangs a challenger beside a title.
Both papers are careful. They disagree about whether a set of centimetre-high cones is biology or geometry, and the honest position at this depth in the record is that the question is open.
Cone-shaped layers in Isua metacarbonates, reported in 2016 as the world's oldest stromatolites — sediment trapped and bound by microbial mats. A 2018 reassessment in the same journal, working on the same outcrop after further exposure, argued the cones are deformation structures produced by the rock being squeezed, and that they are not layered the way a mat is. Both papers are cited here.
3.48 bya
Microbial mats
Layered domes, built by mats of cells that trapped sediment, grew up through it, and trapped more.
Living stromatolites still exist — Shark Bay is about eight hundred kilometres down the same coast. The building method has been in continuous use for three and a half billion years.
Stromatolites and geyserite from the Dresser Formation in the Pilbara, Western Australia. Geyserite forms only around hot springs on land, which places these communities in fresh water at the surface rather than at a deep-sea vent — a live argument about where life got started.
3.2 bya – 2.45 bya
Cyanobacteria
Photosynthesis already existed. What is new here is the reagent: water, which is everywhere, instead of hydrogen sulphide, which is not.
It is the single most consequential chemical innovation in the history of the planet, and the waste product is oxygen.
Drawn as a span because the date is genuinely open. Geochemical traces of oxygen production and molecular-clock estimates for the cyanobacterial lineage disagree by hundreds of millions of years, and the museum has picked the middle of a live argument. The event is not in doubt; its position on this wall is.
2.43 bya – 2.22 bya
Cyanobacteria
The first organisms to poison the entire planet did it with oxygen.
For everything then alive, oxygen was a corrosive waste gas, and there was suddenly no way to get away from it. What survived either learned to breathe the poison or retreated permanently into mud, rock and gut — where a great deal of it still lives.
The wall gives this two hundred million years of width because that is how long the atmosphere spent flickering on and off before it stayed.
Named as an event and dated as one for decades. Sulfur, iron and carbon systematics through two South African drill cores show the atmosphere oxygenating, falling back, and oxygenating again across roughly 200 million years, becoming permanent about 2.22 billion years ago — around 100 million years later than the textbook figure.
2 bya – 1.6 bya
An alphaproteobacterium
Every cell in your body is powered by the descendants of a bacterium that came inside another cell and never left.
Lynn Margulis put the argument in print in 1967. By her own account the paper was turned down by about fifteen journals first. It is now in the first chapter of the textbook.
Every organism in this zoo carries the same passenger.
The endosymbiosis itself leaves no fossil. The date is bracketed by molecular clocks and by the first fossils convincingly read as eukaryotes, and the two methods do not agree closely. What is not in dispute is that it happened: the mitochondrion keeps its own genome, and that genome is bacterial.
1.65 bya – 1 bya
The cells with a nucleus fan out into the shapes this museum collects: ciliates, amoebae, algae, the ancestors of fungi, plants and animals.
Every portrait in the hall belongs somewhere in this band. So does every visitor reading it.
A span covering the diversification of the eukaryotes, not the origin of any one of these animals. Most of the lineages in this museum leave no fossils at all, so their dates come from molecular clocks, and clock estimates for the same group can differ by several hundred million years depending on the calibration. Nothing on this wall dates an individual portrait.
Stentor · Volvox · Physarum · Spirostomum · Heliozoa · Vorticella · Paramecium · Tardigrades
1.05 bya
Bangiomorpha pubescens
Fossils this old are usually a smear of carbon or a disputed shape. This one is a filament of cells with holdfasts, spores and two distinguishable cell types — a specimen you can look at and describe.
It is also a chloroplast story. The lineage it belongs to carries a cyanobacterium that moved in, exactly as the mitochondrion did, and stayed.
Dated directly, by rhenium–osmium on the shales that hold it, which moved it from a long-quoted ~1.2 billion years to 1.047 billion. A red alga, and the oldest fossil showing both differentiated multicellular tissue and what is read as sexual reproduction.
717 mya – 635 mya
The longest of these glaciations ran for roughly fifty-seven million years. Life came through it in meltwater, under ice, and around vents — which is to say microbial life came through it, because there was no other kind.
The first animals appear within about sixty million years of the ice going.
Two glaciations — the Sturtian, beginning about 717 million years ago and lasting some 57 million years, and the shorter Marinoan ending about 635 — dated by rhenium–osmium and uranium–lead on four continents. How completely the ocean froze is still argued; that it froze at tropical latitudes is not.
558 mya
Dickinsonia
A flat, ribbed oval up to a metre across, lying on a sea floor covered in microbial mat.
Everything above this line on the wall happened without animals. That is not a gap in the record. It is most of the history of life.
Dates the oldest fossil that can be shown to be an animal, not the origin of animals, which is certainly older and is estimated from molecular clocks rather than read from rock. Cholesterol preserved in the fossil film settles what Dickinsonia was after seventy-five years of argument that included fungus, lichen and failed experiment.
The wall changes scale here. Everything below the Cambrian line is drawn four times larger than everything above it. Left at one scale, the entire fossil record of animals — every dinosaur, every forest, every ancestor you have a name for — would be a band about as tall as this paragraph.
538.8 mya
Shells, eyes, legs, jaws — hard parts, and therefore a fossil record that a person can walk into a museum and look at.
This is where most natural history museums begin, and it is worth knowing what that costs: by the time this line is drawn, six sevenths of the history of life is already over.
The base of the Cambrian, fixed by uranium–lead dating of ash beds at the boundary section in Namibia. The figure is precise to a few hundred thousand years, which is unusually good for a date this old and is why it is quoted to one decimal place.
295 mya
White rot fungi
Wood is difficult to eat. Lignin is the reason a tree can stand up, and for a long stretch of the Carboniferous nothing appears to have had the enzymes to take it apart.
So the trees fell over and stayed. The coal seams that powered the industrial revolution may be a sixty-million-year window in which the world’s decomposers had not yet caught up — and the window closes here, when a fungal lineage works out how.
A date reconstructed from 31 fungal genomes, not from a fossil: the enzymes that break lignin appear to originate around the end of the Carboniferous. The proposal that this is why the coal stopped forming is actively disputed — tectonics and climate produced the right basins at the right time as well, and the argument is not settled.
251.9 mya
Methanosarcina
Somewhere around nine in ten marine species did not survive it.
One reading of the carbon isotopes says the killer was not the volcanoes directly but what the volcanoes fed: an archaeon that picked up two genes by horizontal transfer and turned the sea floor’s buried carbon into methane, faster than the planet could absorb it.
It is a contested hypothesis and it is on the wall as one. But it is worth standing here for a moment with the possibility that the largest extinction in the history of the Earth was a bloom.
The date is well constrained — uranium–lead ages put the main extinction pulse at about 251.9 million years ago, over less than 60,000 years. The organism is the contested part. The proposal is that a methanogen acquired a gene for consuming acetate, bloomed on the ocean's organic carbon, and released enough methane to drive the crisis, with Siberian volcanism supplying the nickel its enzyme needs. Volcanism alone remains the more widely held explanation.
66 mya
The most famous catastrophe in the history of life, and on this wall it is very near the bottom.
The plankton were hit as hard as anything walking. Whole groups of shell-building microorganisms disappear at the boundary and the survivors rebuilt the ocean’s food web from underneath — the part of the story that does not get a skeleton in the entrance hall.
Argon–argon dating puts the impact and the extinction within about 32,000 years of each other — close enough that the two cannot be separated in the rock.
300,000 years ago
Three hundred thousand years. At the scale this part of the wall is drawn, that is a line about as thick as the rule it sits on.
The organisms in the next room have been here for three and a half billion.
Fossils from Jebel Irhoud in Morocco, dated by thermoluminescence on burnt flint to about 315,000 years ago, and read as early Homo sapiens across the whole of Africa rather than from a single cradle. Marks the oldest such fossils, not an origin date — there is no year in which a species begins.
Off the scale entirely
The last two events on this timeline cannot be drawn on it. At the scale of the section above, both of them are thinner than the rule they would be drawn on, and they are the two that this museum exists because of.
1676
Antonie van Leeuwenhoek
A draper in Delft, grinding his own lenses, looks into a drop of rainwater that has been standing in a pot and finds it full of animals.
The Royal Society did not believe him. They sent instructions to England’s best observers to try to repeat it, and it took the better part of two years before they were satisfied.
This is the founding moment of this museum. Everything in the portrait hall had been going about its business for billions of years, and this is the first time any of it was seen.
The letter of 9 October 1676, printed in English by the Royal Society the following year, describing living things in rainwater, well water, sea water, snow water and water in which pepper had been steeped. Earlier observations of other subjects exist; this is the letter that reports the animalcules.
1986
Prochlorococcus
It is roughly half a micrometre across, it lives in most of the sunlit ocean, and there are something like three thousand million million million million of them.
It took three hundred and ten years from the first sighting of any microorganism to notice the commonest one. It is in the records hall, holding a title.
Detected by flow cytometry in the Sargasso Sea in 1986 and described in Nature in 1988. Later work established it as the most abundant photosynthetic organism on Earth. The date given here is the observation; the description follows two years after.
Put the three numbers beside each other. They have been here for 3,480 million years. We have been here for one part in 11,600 of that. We have known they exist for 350 years, which is roughly one part in 10 million.
Everything that has ever been written about them, by anyone, was written in that last sliver. The portrait hall is a very early draft.
Powers of Ten does the same thing on the other axis — how small, rather than how long.