About three billion years ago, microbes in shallow seas began giving off oxygen, and the iron dissolved in the water rusted out of it in bands. The oldest object in the collection holds that moment, red and black, layer on layer. It is the reason for everything else here: without that oxygen, nothing that breathes would ever have lived. No fish, no dinosaur, no us.
The planet forms at midnight on January 1, and the Moon on January 4. The tiger iron forms in early May. The trilobite walks a seafloor on November 21. Megalosaurus dies on December 18, Stan on December 26. Our species appears at 11:25 in the evening on December 31, and everything anyone has ever written down happens in the last thirty-six seconds.
Every fossil here is older than the human species except one. Smilodon went into the tar around twenty-five thousand years ago, by which time people had been on Earth for more than a quarter of a million. Somebody watched that animal die.
Glossopteris grew in South America, Africa, India, Antarctica and Australia, and its seeds could not have crossed an ocean. It was among the first hard evidence that the continents move. Scroll back 260 million years and watch them come apart.
See it from orbit →That is what these ages look like when every year gets equal width. Everything after the tiger iron collapses against the right-hand edge, and four hundred million years of animals becomes a single mark. "Stretched" divides by ten at each step so the recent past can be told apart. Nothing else changes.
Where the Earth gave them up
Seven countries on five continents. TerraMar means land and sea, and this is the land part: every place a piece of this collection came out of the ground.
Where the whole animals are. The public places a visitor can stand in front of the original, a cast, the living thing, or what is left of it. This collection is teeth, claws and slabs; these are the rooms with the rest.
Drag to spin · tap a marker or a name
The worlds they lived in
The continents have not always been where they are. Ten Earths, drawn from our best reconstructions, with the pieces of this collection placed in the world that made them.
Scientific note: these ten Earths are stylized reconstructions, drawn by hand from published paleogeographic research and simplified for clarity. Positions before about 540 million years ago are not reliably known and are shown schematically. These reconstructions reflect the science as of 2026; they are revised when it changes.
See the planet change through all ten, from orbit →
The registry
Every object in the collection, in accession order. Tap an entry for the full record: what the object is, where it came from, what the animal was, who it knew, and how its age is known. The blue links walk the food web. The colored chips name shared worlds: groups that lived in the same place at the same time. Accession codes read collection, year, sequence: TMC-24-001 is the first piece of 2024. Entries marked cast are documented reproductions, here because the originals cannot be owned at any price.
Under every animal, a line in gold says the name out loud: AL-oh-SORE-us, not a phonetic alphabet, because the names are hard enough already and everyone can read capitals.
What is not here
A registry is also a record of its gaps. Sorted by age, the collection falls silent in places, and the glossary below defines things the shelves do not hold: no sauropod, no ornithischian, no synapsid, no bird, no soft animal, no holotype, no trace of what any animal did, nothing made by a hand. This is the list of those absences, why each exists, what would honestly close it, and what it would add to the story already here. It is long on purpose. TerraMar means land and sea; four of these are from neither. Some of the gaps will stay, because the right specimens are protected where they lie. That is as it should be.
4.567 billion years
- The gap
- The oldest object here is the stromatolite at 3.02 billion years. An ordinary chondrite meteorite is older than that by a billion and a half years: it is as old as the solar system, and the Earth’s own oldest surviving rock is younger than it by half a billion.
- Why it is empty
- Not scarcity. Chondrites are the commonest meteorites there are. The gap is a choice not yet made: it should be a classified stone from a named fall, with its type and locality on paper, not a dark rock with a story. Allende, which fell in Mexico in 1969, is the one: its white inclusions are the oldest solids ever dated, and the age of the solar system was measured from them.
- What would close it
- A classified chondrite, witnessed fall preferred, Allende first, with the classification and locality documented.
- Closed
- What it adds here
- It moves the first mark on the timeline off the Earth and past the planet’s birth. The stromatolite would no longer be the oldest thing in the collection, only the oldest thing that was ever alive, and “how the age is known” would have its founding example: the rock the solar system was dated from.
A planet that failed
~4.56 billion years
- The gap
- A pallasite: olivine crystals set in nickel-iron, cut thin and held to a light. It is the boundary between the metal core and the stony mantle of a small planet that was broken apart before the Earth was done forming. Nothing else in the solar system looks like it, and nothing here shows what the inside of a world is made of.
- Why it is empty
- Choice, and cost. Small pallasite slivers are sold everywhere, often in printed boxes. The slice worth having, thin enough to pass light and large enough to hold, costs several hundred dollars, and the collection is not buying a sliver to tick a box. The care is in the name, the classification and the size.
- What would close it
- A classified pallasite slice, Seymchan or Imilac, thin enough to pass light, name and mass on paper.
- Closed
- What it adds here
- Every other object from space on this list is about age. This one is about what a planet is: metal at the middle, stone outside, and the place where they meet. It would also be the first thing in the collection that is best seen with a lamp behind it.
Fell ~4,000 years ago
- The gap
- No iron meteorite. Irons are the metal cores of asteroids that broke apart, nickel-iron that cooled so slowly inside its parent world that, cut and etched, it shows a crosshatch of crystals no metal on Earth can grow.
- Why it is empty
- A choice not yet made, and a matter of look. Campo del Cielo, the shower of iron that fell on northern Argentina about four thousand years ago, is the obvious one: named, classified, and known to the people there long before it was written down in 1576. It should be a polished and etched slice or a whole piece with a shape of its own, never a trinket in a printed box.
- What would close it
- A classified Campo del Cielo iron, an etched slice or a characterful individual, with its mass on paper.
- Closed
- What it adds here
- The pallasite would show where a world’s core meets its mantle. This is the core itself, and the first thing in the collection heavy enough to surprise a hand.
The Moon
~4.4–3.9 billion years
- The gap
- No lunar material. Apollo rock is federal property and cannot be owned, so “moon rock” in the sense people mean is off the table forever. But the Moon sends itself: rock knocked off its surface by impacts falls here as lunar meteorites, mostly feldspathic breccias from the highlands, recognized because their chemistry matches the Apollo samples exactly.
- Why it is empty
- Cost per gram, and language. Most of what is sold as moon rock is either a lunar meteorite priced by the gram or nothing at all. The word for the real thing is lunar meteorite, with a Meteoritical Bulletin name and a classification, and it should be a slice or fragment with a mass, not a vial of dust.
- What would close it
- A classified lunar meteorite, feldspathic breccia, slice or fragment, name and mass on paper.
- Closed
- What it adds here
- TerraMar means land and sea. This would be the first object from neither, and the globe at the top of the page would need a new place to put it. It is also, for nearly everyone, the only way to own a piece of the Moon: not a sample brought back, but one that arrived on its own.
Mars
4.4 billion, or ~570 million years
- The gap
- No Martian material. A few hundred Martian meteorites are known, blasted off the planet by impacts and identified by gas trapped inside them that matches the atmosphere Viking measured in 1976. Most are shergottites, young volcanic rock; a handful are ancient crust.
- Why it is empty
- Timing, not scarcity. Tissint fell in Morocco in July 2011 and was seen falling, which has happened only a handful of times with Mars. Its rock crystallized around 570 million years ago, which would put a piece of another planet on this timeline inside the Ediacaran, the same age as the first animals on Earth.
- What would close it
- A classified Martian meteorite, Tissint preferred: witnessed fall, Moroccan, named, with mass recorded.
- Closed
- What it adds here
- Three of the worlds on this site are Moroccan. Tissint would be a fourth Moroccan world that is not on Earth. And its age is known the same way as the rest of the registry, radiometrically, which makes the point better than any sentence could: the clocks work on other planets.
4.03 billion years
- The gap
- The stromatolite is the oldest thing here that was alive. The oldest thing here that is Earth is the labradorite at 1.3 billion, and the planet has rock three times older than that. The Acasta Gneiss of the Northwest Territories is the oldest intact rock ever found: continental crust that formed while the planet was still being hit by the leftovers of its own making.
- Why it is empty
- It has simply never been chosen. Acasta is collected under permit from a known locality and sold in small documented pieces. It does not look like much, which is the whole point of writing a label for it.
- What would close it
- A documented piece of Acasta Gneiss, locality on paper, large enough to see the banding.
- Closed
- What it adds here
- The meteorite entry above moves the timeline off the Earth. This is the Earth’s own first surviving page, and it is Canadian like the labradorite, which gives the two oldest Earth rocks in the collection one country and two billion years between them.
~3.4–1.8 billion years
- The gap
- The oldest object here is tiger iron: the iron that fell out of the sea when the first oxygen rusted it clean. It is the work of the microbial mats, not the mats. A stromatolite in the strict sense, the laminated dome a mat builds layer by layer as it grows toward the light, is the oldest visible life on Earth, and the shelf does not hold one.
- Why it is empty
- The famous ones are protected: the 3.4-billion-year domes of the Pilbara and the living mats of Shark Bay stay where they are. But stromatolites two billion years old are cut and polished from named localities in Minnesota, Bolivia, and Morocco, cheaply, and in section the domes are unmistakable.
- What would close it
- A polished stromatolite slab cut across the domes, two billion years or older, locality named.
- Closed
- What it adds here
- The builder next to what it built. The tiger iron is the consequence; this would be the cause, and the two together are the whole first chapter: life, oxygen, rust, and a blue ocean.
~575–539 million years
- The gap
- Nothing between the labradorite at 1.3 billion years and the trilobite at 507 million. That silence covers the arrival of animals: soft, strange forms like Dickinsonia and Charnia, pressed into seafloor before anything had a shell or an eye.
- Why it is empty
- Nearly every legitimate locality is protected: the Flinders Ranges, Mistaken Point, the White Sea coast. Material that reaches the market usually cannot show how it left the ground. This is a gap the registry expects to keep.
- What would close it
- A documented impression with an export history that survives scrutiny, or a museum-produced cast of a type specimen. Nothing else.
- Closed
- What it adds here
- The trilobite is the first animal here. This is the animal before the first animal, and the only way to show that shells and eyes were inventions, not starting conditions.
~420 million years
- The gap
- Between the trilobite at 507 million years and the Devonian cephalopods there is a hundred million years in which the first jaws appeared, the first plants came ashore, and the first animals followed them. The registry says nothing. The animal that belongs here is a eurypterid: a sea scorpion, the first big predator with claws.
- Why it is empty
- Never chosen. Eurypterus remipes from the Bertie Group of New York, the state fossil, is common, legal, and honest, and a good one still has its paddles.
- What would close it
- A eurypterid from the Bertie Group, New York, complete or near it, with the quarry named.
- Closed
- What it adds here
- It is the first predator on the timeline that a child would recognize as one, a hundred million years before anything with teeth, and it is the first entry from the long middle of the Paleozoic that the shelves currently skip.
~385–375 million years
- The gap
- TerraMar means land and sea, and nothing here is the crossing. The Devonian world on this site holds cephalopods and nothing with a spine. The animals that carried a backbone out of the water, lobe-finned fish with bones in their fins that would become arms, are absent from a collection named for the two places they moved between.
- Why it is empty
- Tiktaalik came off Ellesmere Island under permit, belongs to a museum, and will only ever be a cast. But the lobe-finned fish that came before it are real and available: Osteolepis from the Orcadian lake beds of Scotland, whole on its slab, legal and cheap.
- What would close it
- An Osteolepis from the Achanarras beds of Caithness, with the bed named. A museum cast of Tiktaalik if one is ever offered.
- Closed
- What it adds here
- This is the thesis of the site made into an object: the animal that went from Mar to Terra. Every dinosaur, every mammal, and everyone reading this is downstream of that fin.
~340 million years
- The gap
- A crinoid. A sea lily, which is not a plant but an animal on a stalk, an echinoderm like a starfish that spent its life anchored to the sea floor with its arms open. The registry now holds a piece of the sea floor, the Ordovician brittle-star plate, and with it the first echinoderms in the collection. It still holds no crinoid.
- Why it is empty
- Only choice. The Crawfordsville beds of Indiana produce crinoids with the crown, the stalk, and the arms all intact, and they have been coming out of the same hillside since the 1860s.
- What would close it
- A Crawfordsville crinoid with crown and stalk, on matrix, with the quarry named.
- Closed
- What it adds here
- The Carboniferous on this site is a forest. This is the other side of the shoreline in the same age: the sea floor as a garden, and the first animal here that looks like it should be on the other timeline.
~328 million years
and Cenomanian
~95 million years
- The gap
- An octopus. It sounds absurd and it exists: the oldest known, from the Bear Gulch beds of Montana, is 328 million years old and has ten arms, and the Cenomanian limestones of Lebanon have given up a handful of true octopuses with arms, suckers, and ink sac pressed flat into the stone. Nothing here is soft. Every animal on the shelves left a hard part.
- Why it is empty
- Count. The Montana animal is one specimen in a museum. The Lebanese octopuses are a few dozen in the world, from quarries whose fish are sold but whose octopuses almost never are. The honest expectation is a cast, or a lifetime of waiting.
- What would close it
- A documented Lebanese octopus from the Hakel or Hjoula beds with export history, or a museum cast of one, from a named institution.
- Closed
- What it adds here
- The registry walks the cephalopods from the straight cone to the coiled shell. This is the branch that gave the shell up entirely and kept the brain. It would be the first soft-bodied animal here, and the strangest thing in the collection.
~315–310 million years
- The gap
- The registry holds a seed fern from the Breathitt Group of Kentucky and no tree from the forest it grew in. The trees were Lepidodendron and Sigillaria, lycopsids a hundred feet tall (30 m) with bark patterned like snakeskin, and Calamites, a horsetail the height of a house. They are the coal. The fern is what grew in their shade.
- Why it is empty
- Not scarcity. The roof shale above a Pennsylvanian coal seam is the mud that buried the forest standing on the peat, and bark, trunk casts, and horsetail stems come out of Appalachian mines and spoil constantly; most of the classic Carboniferous plants in museum drawers came from exactly there. What matters is the seam name written down, because the seam is the date.
- What would close it
- A section of lycopsid bark with the leaf scars visible, a Calamites stem, and a piece of coal from the same seam, seam and county named.
- Closed
- What it adds here
- The tree, the forest floor, and what the forest became, from one seam. It joins the Carboniferous world and the seed fern directly, and it turns the coal in the story from a word into an object: the fern grew in this tree’s shade, and this is what the tree turned into.
~309 million years
- The gap
- A Mazon Creek nodule. The ironstone concretions of Illinois split open to show soft animals that fossilize nowhere else: jellyfish, worms, shrimp, and the Tully monster, an animal described in 1966 that no one has been able to place on the tree of life since.
- Why it is empty
- Choice. Common nodules with ferns or shrimp are inexpensive; the Tully monster is not rare so much as sought. The beds are from a strip mine’s spoil, collected for decades, and material is documented by pit.
- What would close it
- A Mazon Creek nodule, both halves, with a named animal, pit recorded. The Tully monster is the one to wait for.
- Closed
- What it adds here
- It is the second soft-bodied entry on this list and the only one that is affordable. It also puts an unsolved problem on the shelf, which a registry that keeps saying how things are known should be willing to do.
- The gap
- The same seed ferns as the Kentucky specimen, from the Llewellyn Formation at St. Clair, Pennsylvania, but preserved in white and, in the best pieces, in rainbow. The color is a film of the mineral pyrophyllite that grew over the leaf as it decayed, and it shifts as the stone is turned.
- Why it is empty
- A choice not made. These are common, legal, and sold everywhere from fossil shows to the internet, and the good ones have been seen and passed over. The care is in getting a piece with the iridescence intact, not painted or coated, and the locality named.
- What would close it
- A St. Clair fern slab, Neuropteris or Alethopteris, with natural iridescence, from a named pit.
- Closed
- In September 2026, by a slab of Ptychocarpus from the Llewellyn Formation near Hazleton, in the Eastern Middle Anthracite Field of Luzerne County (TMC-26-029). Not quite what this entry asked for, and better. Not St. Clair but the same formation a little to the north; the field is named, the pit is not. And the main plant is not a seed fern but a true fern, a tree fern, with a Macroneuropteris leaf beside it: so the slab holds both kinds of coal-forest plant at once. The color is the fossil’s own, and it is the rarest thing about it: fossil ferns are everywhere and black; this district is the one place that gives them up in blue, violet and gold, and only a fraction of those. The collector it came from sealed the face with a UV-resistant clear coat to keep it from fading, and the record says so, because that is what this entry was careful about.
- What it adds here
- The thing that looks fake is the most honest object on this list. It gives the Carboniferous a second locality and the seed fern a second way of being preserved, and it is the one entry here that a child will pick up first.
~295–272 million years
- The gap
- Nothing between the seed fern at 310 million years and the Megalosaurus at 167. The hole swallows the whole Permian and Triassic, and the end-Permian extinction with them: the worst that ever happened, and the registry holds nothing from either side of it.
- Why it is empty
- The animal that belongs here is Dimetrodon, the sail-back everyone files under dinosaurs. It is a synapsid: on the line to mammals, closer to you than to any dinosaur that ever lived. Teeth from the Texas red beds are common; a good neural spine from the sail, documented to locality, is not.
- What would close it
- A Dimetrodon sail spine section or jaw fragment from the Texas red beds, with the formation named.
- Closed
- What it adds here
- The glossary already tells the reader that Dimetrodon is closer to them than to any dinosaur. Right now that is a sentence. This would make it an object the kids can put a hand on, and the first thing in the collection on the line that leads to us.
~225–210 million years
- The gap
- Nothing Triassic, and the plants are one fern. Petrified wood from the Chinle Formation of Arizona is both: conifer logs that stood on a river floodplain when dinosaurs were new and small, turned to agate cell by cell.
- Why it is empty
- Petrified Forest National Park is protected. The private ranch land around it produces the same wood legally, and a cut slab with the ranch named is not hard to find. The gap is only that it has never been chosen.
- What would close it
- A Chinle Formation wood slab or round from named private land in Arizona, polished on one face and left natural on the other.
- Closed
- What it adds here
- The first thing here from the Triassic, the age the dinosaurs were born into, and the second plant. The seed fern is the forest that became coal; this is the forest that became stone.
~200–180 million years
- The gap
- An ichthyosaur, the fish-shaped reptile of the Lyme Regis and Whitby cliffs. The collection has no Early Jurassic and no British original material, and this animal is both. It is also the same chapter as the Megalosaurus: Mary Anning was pulling these from the Dorset shore in the same decades Buckland was describing the Oxford jaw.
- Why it is empty
- An earlier version of this entry held out for a jaw and refused the easy thing. That was wrong. A Lias vertebra is honest, common, and costs less than dinner, and it puts a British original on the shelf now. The jaw, the paddle, the run of articulated vertebrae are the upgrade, not the entry.
- What would close it
- An ichthyosaur vertebra from the Lias of Dorset or Yorkshire, bed named, now. A jaw section or paddle from the same beds when the right one appears.
- Closed
- What it adds here
- Half this family is British, and the collection holds no British original material. This would be the first, and it ties Anning’s shore to Buckland’s jaw, which is already on the shelf.
~150 million years
and Eocene
~52 million years
- The gap
- No bird, and no feather. Birds are the dinosaurs that lived through the day the K–Pg entry below describes, and a registry with two tyrannosaur casts and nothing with feathers is telling half the story. Archaeopteryx from Solnhofen is the animal; a feather from the Green River lake is the same fact, fifty million years later.
- Why it is empty
- The Solnhofen originals are counted on two hands and are in museums. Green River feathers are not: they come from the same Eocene lake as the fish already here, on the same pale stone, with the barbs visible.
- What would close it
- A Green River feather on matrix, quarry named, now. A museum cast of an Archaeopteryx slab, from a named institution, when one is offered.
- Closed
- What it adds here
- The feather joins TMC-26-011 directly, same lake, same rock. Together with the boundary layer below it says the one thing the Cretaceous entries cannot: that the extinction had survivors, and that they are outside the window.
~155–148 million years
- The gap
- No sauropod bone. The family that produced the largest land animals that ever lived is in the collection only as eggshell. Camarasaurus and Brontosaurus from the Morrison Formation would sit between the Megalosaurus and the big ammonite, which at this scale are neighbors.
- Why it is empty
- Size. A sauropod bone worth having is a limb bone, and a limb bone is furniture. Morrison material is legal and documented from private ranch land; the problem has only ever been the shelf.
- What would close it
- A Morrison Formation sauropod limb bone, tibia or femur, with the quarry named. Camarasaurus first.
- Closed
- What it adds here
- The largest thing that ever walked, beside the largest shark that ever swam, and a parent for the eggshell on the shelf.
~140–125 million years
- The gap
- In 1842 Richard Owen invented the word dinosaur for three animals: Megalosaurus, Iguanodon, and Hylaeosaurus. The first is here, as a cast. The other two are not, and together they are the only place the story of the word can be told in full.
- Why it is empty
- Iguanodon is possible: Wealden material from the south of England comes up with locality. Hylaeosaurus is one animal in London and will only ever be a cast, if a cast of it exists at all.
- What would close it
- An Iguanodon tooth or bone from the Wealden, documented. A Hylaeosaurus cast, if one is ever produced. And, above both, an original Megalosaurus tooth or bone from the Stonesfield Slate, which would replace nothing and be taken the day it appeared.
- Closed
- What it adds here
- The Megalosaurus cast is the first dinosaur ever named. With Iguanodon beside it, the collection would hold two of the three animals the word was invented for, and could tell the story of the word itself.
~100–95 million years
- The gap
- Nothing in the collection ever left the ground. The Kem Kem world on this site already holds Spinosaurus and the sawfish, and the sky over that river was full of pterosaurs whose teeth are among the commonest fossils the beds produce.
- Why it is empty
- Not chosen. A Kem Kem pterosaur tooth, long and slightly curved, is inexpensive and honest, and it comes from a world that is already built.
- What would close it
- A pterosaur tooth from the Kem Kem beds, Coloborhynchus or a relative, with the locality named.
- Closed
- What it adds here
- A world on this site gains its sky. It also makes the Quetzalcoatlus cast further down a second flying thing rather than a first, which is the right order: the common animal before the impossible one.
- The gap
- Spinosaurus is here as a tooth. The animal is not a tooth; it is a sail, a row of spines standing along its back, the tallest ever measured about five and a half feet (1.65 m), and nobody agrees what the sail was for. Display, heat exchanger, fat store, swimming fin: four serious answers, all defended in print. A single dorsal vertebra with its neural spine standing three feet above the centrum is the sail in one object.
- Why it is empty
- Geometry, and speed. A blade that tall and that thin snaps near its base long before it reaches a market, so Moroccan spinosaurid material arrives as teeth, or as the spool-shaped centrum with the spine gone. When something whole comes out of the Kem Kem it is gone before it is public: to a museum, to science, or into a private collection nobody hears about. And the animal itself is a reconstruction. The type specimen burned in Munich in 1944, and every Spinosaurus since has been rebuilt from Stromer’s 1915 drawings and from pieces. An original with the spine intact is a museum’s, not a family’s. The search is for the rare cast taken from a real one, made by someone who will say what was original and what was not.
- What would close it
- A full-scale cast of a Kem Kem spinosaurid dorsal vertebra with its neural spine, taken from a real specimen and not from another cast, restoration disclosed in writing. An original if the world ever changes.
- Closed
- Closed by the sail vertebra, TMC-26-026: a full-scale cast of a dorsal vertebra with its neural spine standing, 42¾ inches on its custom stand, taken from real Moroccan bone by Gaston Design, with the missing last third of the blade rebuilt from Stromer’s plates and the restoration stated plainly. Why this one and not another: we found no other. No other full-scale sail vertebra taken from a real specimen was to be had, and an original with its blade whole is a museum’s. It meets every word of what would close this gap.
- What it adds here
- The Kem Kem world on this site gains the animal’s defining structure rather than its smallest part, standing as tall as it stood, and a registry that keeps saying how things are known gets an object whose whole subject is a hundred years of argument.
~80–70 million years
- The gap
- The registry has the straight cone and the coiled one. It does not have the shapes in between: the heteromorph ammonites, Baculites straight as a pencil, Didymoceras in a loose corkscrew, Nipponites tied in a knot. A lineage that had perfected the spiral and then, at the end, let go of it.
- Why it is empty
- No good reason. Baculites from the Pierre Shale is honest, common, and beautifully preserved. The stranger forms are rarer and more often reconstructed, which is where the care has to go.
- What would close it
- A heteromorph with its suture visible and its restoration disclosed. Didymoceras is the one to hold out for.
- Closed
- What it adds here
- The orthocone, the goniatite, and the two ammonites already walk the shell from straight to coiled across 350 million years. This is the last chapter of that line: the spiral abandoned just before the asteroid ended the whole group.
~75 million years
- The gap
- Four ammonites are here. None of them is a jewel. In the Bearpaw Formation of Alberta the shell of Placenticeras was buried under conditions that turned its mother-of-pearl into ammolite, a gemstone made of the animal itself, red and green and gold as the stone is turned.
- Why it is empty
- Only choice. Alberta guards its fossils and keeps them, which is right, but ammolite is the exception: it is mined under provincial license and leaves the country legally, which is why it can be here when the Gorgosaurus cannot.
- What would close it
- A piece of ammolite on its own matrix, unpolished on one face if possible, from a named mine, with the shell structure still visible.
- Closed
- What it adds here
- It answers a question the site keeps circling: whether a stone can tell a story. This one is an animal that became a gem without ceasing to be the animal, and it is the only Alberta fossil the collection will ever legally hold.
~79–75 million years
- The gap
- Three T. rex objects and no other tyrannosaur. The family is more interesting than its last member: Gorgosaurus, Daspletosaurus, Albertosaurus, lighter and earlier, running the northern plains ten million years before Stan.
- Why it is empty
- The best of them come from Alberta, where fossils belong to the province and stay there, which is right. What is legally available is Judith River material from Montana, where the same animals cross the border into private land.
- What would close it
- A tyrannosaurid tooth or bone from the Judith River Formation, Montana, with the county named. A large one.
- Closed
- What it adds here
- Stan and Sue are the end of a story. This is its middle. A Gorgosaurus tooth beside the two T. rex casts turns a name into a lineage, which is what the connections on this site are for.
~76–66 million years
- The gap
- No ceratopsian. The registry’s dinosaurs are all theropods: teeth, claws, jaws, faces built to frighten. The horned dinosaurs are the other tradition, and the frilled ones with windows in the bone, Torosaurus, Chasmosaurus, Pentaceratops, are the only dinosaurs that read as sculpture rather than threat. Fifteen genera from the northern plains alone, and none of them here.
- Why it is empty
- Nothing but priority. Triceratops teeth and frill fragments from Hell Creek are common and legal, chasmosaurine frill sections come out of the Judith River on private land in Montana, and a brow horn is a real purchase but a possible one. Diabloceratops, the one with the horns on its frill, lives in a Utah monument and belongs to the cast entry below.
- What would close it
- A Triceratops tooth now, a frill section with the vascular texture visible next, and a brow horn core from Hell Creek or Judith River, county named, when the right one appears.
- Closed
- What it adds here
- It would be the first dinosaur here that was not a predator and the first that is beautiful before it is frightening. It also gives the tyrannosaurs something to have been looking at.
- The gap
- Stan and Sue are here. What they ate is not. Edmontosaurus, the duckbill, came out of the same rock in herds, and the registry holds nothing that stood on the prey side of the food web: no hadrosaur jaw with its grinding battery of teeth, no limb bone from the animal the tyrannosaurs were built to catch.
- Why it is empty
- Only priority. Hell Creek is among the best-documented formations that private land produces, and the same clocks already on this site would date the bone. A hadrosaur jaw section or a large limb bone would sit beside the tyrannosaurs and complete the room.
- What would close it
- An Edmontosaurus jaw section or limb bone from Hell Creek, with the locality named.
- Closed
- It arrived in September 2026: a partial fibula of Edmontosaurus annectens, 32.1 inches of original bone from the right hind leg, Lance Formation, Weston County, Wyoming, with the county confirmed in writing. The gap asked for Hell Creek; the Lance is the same Maastrichtian plain under Wyoming’s name for it, the same age, the same animals, across a state line the dinosaurs did not know about, and the entry says so rather than pretending otherwise. Why this one: it is the first original dinosaur bone in the collection, and the gap was written too small for what closed it. This animal was not the other half of anything. It was as long as the tyrannosaurs, it outnumbered them by the herd, it had reached every searched continent, and it was standing on this plain in the last years before the impact. Three feet of its leg now sit beside three of its neighbors’ teeth. The ankle end was never recovered and stays an open want under the same number.
- What it adds here
- It would be the first thing here that Stan could have eaten. Every food-web link on this site runs predator to prey, and Hell Creek currently has no prey.
~67–66 million years
- The gap
- A tooth from Nanotyrannus. For forty years the question has been whether the small tyrannosaur of Hell Creek is its own animal or a young T. rex, and in 2025 a study of the Dueling Dinosaurs specimen argued, with a fully grown skeleton, that it is its own. The argument is not over. The registry holds nothing that is still being argued about.
- Why it is empty
- Never chosen. Hell Creek teeth sold as Nanotyrannus are common and cheap; what matters is the locality and an honest label that says what the name is, which is a claim under review.
- What would close it
- A Hell Creek tyrannosaurid tooth of the Nanotyrannus type, blade-like and finely serrated, county named, labeled as the argument it is.
- Closed
- What it adds here
- Beside Stan and Sue it puts the reader inside a scientific disagreement that is happening now, which the glossary and “how the age is known” exist to make possible. A registry that only holds settled things is a catalog.
~67–66 million years
- The gap
- Before 1988 nobody had a complete Tyrannosaurus rex arm. The two-fingered stubs were the joke about the animal for eighty years, told without the evidence. Then a Montana rancher out with her family near Fort Peck saw bone in a cliff, and what came out was the first intact forelimb and hand anyone had recovered, on a specimen now on public display in Washington as the Nation’s T. rex. Stan and Sue are here as a tooth and a cast. The animal that settled the arm is not.
- Why it is empty
- The bone is public property and will never leave the museum, which is right. A claw from it was reproduced once, under the museum’s license, and then production stopped, in numbers nobody published. Sue’s teeth and claws are copied constantly and are easy to find; this one, from the more important arm, almost never was. So the hunt is for one of the few that exist, wherever one has come to rest, with the license still on it and its history intact. And there is a second gap inside the first: no one recorded which claw it is, hand or foot, digit or side, so even the found object will arrive with a question attached.
- What would close it
- A licensed replica of a claw from USNM 555000, located wherever one still exists, with its paperwork. Nothing that copies it, and nothing from a different animal wearing the name.
- Closed
- It arrived in August 2026: the licensed full-scale replica of a claw from USNM 555000, the Nation’s T. rex, found where one still existed, with its paperwork. Why this one: there was one first complete T. rex arm, and one first set of claws on it, and no original will ever leave the museum. This is as close to that claw as anyone who does not work there can stand. It was made from a scan of the original rather than a mold, which the entry records; it is no longer produced and the license has lapsed, so what exists now is all there will be; and it arrived with its question attached, because nobody had written down which claw it is; the anatomy answered it, the inner claw of the left hand, and the entry says how. The first entry on this list to close, and the one the list was written around.
- What it adds here
- A third Tyrannosaurus beside Stan and Sue, and a different kind of famous: not the largest or the most complete, but the one that closed a question open since 1905. Finding it would say something about the collection, too: that it will search for the scarce copy of the right animal rather than take the easy copy of the famous one.
~67–66 million years
- The gap
- Tyrannosaurus is here as a tooth and as casts of teeth. The jaw is a different order of thing. The lower jaw of a large adult runs three feet and more, a battery of teeth the size of bananas set in bone, and it is an object almost no one alive has touched. It sits behind glass in perhaps two dozen museums in the world. It does not sit in houses.
- Why it is empty
- Because there is no ordinary road to it. An original mandible from a named, cataloged Tyrannosaurus belongs to a museum or to a price that is not measured in a family’s numbers, and the few that have passed through private hands did so at auction, in the millions, in the news. The other road is a cast, and almost every cast in the world is not it: resin poured off resin, third and fourth generation, softened at every step, a shape of a jaw rather than the jaw. The only object worth the wait is one step from the bone, molded by the institution that holds the specimen or the shop that worked on it, from a named animal, licensed and numbered and made in quantities a museum controls. Those do not appear for sale. They are found, over years, one at a time, and when one surfaces it is spoken for before the sentence ends.
- What would close it
- The mandible of a named Tyrannosaurus rex, one step from the bone: an original, if the world is ever that kind, or a first-generation licensed cast with the molding institution and the source specimen on paper. Nothing further from the animal than that.
- Closed
- What it adds here
- A tooth is a fact. A jaw is a presence. This would be the first thing in the collection that exists otherwise only in a museum, and the only one where a hand on it is a hand one step from the largest predator that ever walked North America. For most people alive, that is not a purchase. It is the thing you cannot do, which is what this collection is for.
~66 million years
found 1902, named 1905
- The gap
- The registry holds copies of famous animals. It does not hold a copy of a holotype: the one specimen a species’ name is anchored to, the bone every later find is compared against to earn the name. For Tyrannosaurus rex that is CM 9380, found by Barnum Brown in the Hell Creek badlands in 1902, described by Osborn in 1905, and in Pittsburgh since 1941. Without that skeleton there is no T. rex. Stan, Sue, and the Nation’s are T. rex because they match it. (One type specimen is already on the shelves, and it arrived that way by accident: the Megalosaurus dentary is the lectotype of its species, the bone chosen later to carry the name. A holotype proper, named as the name-bearer from the start, is the gap.)
- Why it is empty
- Holotypes do not leave their museums, and copies of them are made rarely and mostly for other museums, because the value of a holotype is scientific rather than decorative and the market has never learned to want them. Casts of elements of CM 9380 exist, made by the Carnegie for study and exhibition; they surface seldom and almost never with paperwork. The search is for one that does.
- What would close it
- A museum-made cast of any element of CM 9380, or of the holotype of any animal already in the collection, institution and element on paper.
- Closed
- What it adds here
- The argument this registry makes for casts, taken to its end. A copy of the specimen that made the knowledge carries the knowledge; a copy of the specimen that made the name carries the name. It would also give the three tyrannosaurs already here the thing they are all measured against.
~67–66 million years
- The gap
- The Dimetrodon entry says the synapsid line leads to the reader, and then the list drops it for two hundred million years. The mammals of Hell Creek were there the whole time, small, nocturnal, living in the same mud as the tyrannosaurs, and one of them inherited the world.
- Why it is empty
- Not chosen. Teeth and jaw fragments of the multituberculate Meniscoessus come out of the Hell Creek screen-wash constantly, from the same private land as the dinosaur bone.
- What would close it
- A Hell Creek mammal jaw or tooth, Meniscoessus or another named genus, locality documented.
- Closed
- What it adds here
- The line that leads to us, picked back up. It is the smallest thing on this list and the one that survived the boundary layer, and it belongs beside the tyrannosaurs on that ground alone.
~72–66 million years
- The gap
- A mosasaur skull. The Khouribga sea is the best-represented world in the collection, four animals and five teeth, and the animal that owned it is here only as a tooth. A skull of Thalassotitan or Mosasaurus hoffmannii would be the centerpiece of the whole ocean.
- Why it is empty
- Not finding one. Phosphate skulls are on the market constantly. The difficulty is finding one whose restoration is disclosed: most are composites of several animals with the seams filled, sold as one. The glossary entries for composite and restoration apply to this object more than any other.
- What would close it
- A mosasaur skull with a written condition report: what is original, what is restored, and from how many individuals.
- Closed
- What it adds here
- The Khouribga sea is the best-built world on the site, and a skull would give it a face. It would also be the first object here large enough to be a room’s center rather than a shelf’s.
- The gap
- Every object here was a body. Nothing recorded behavior. A body fossil shows what an animal was; a trace fossil shows what it did.
- Why it is empty
- Oversight. Trace fossils are among the most affordable objects in the field, and rarely worth faking. The Sidi Chennane beds that produced both plesiosaur teeth have coprolites described in the literature as recently as 2025.
- What would close it
- The first record of an animal doing something, from a named formation.
- Closed
- In September 2026, by titanosaur eggshell from the Late Cretaceous of Aix-en-Provence, southern France (TMC-26-028). These are pieces of eggs laid by titanosaurs, the sauropod family that produced the largest animals ever to walk, each egg about the size of a cannonball, laid in nests on the floodplains of what is now Provence. A bone shows what an animal was. An egg shows the most important thing it ever did. This entry once asked for a track or a coprolite. Tracks are too often painted to be read, and the collector has no wish to own the other. The egg is the better answer: fragile, beautiful, and the record of a life beginning rather than a meal ending.
- What it adds here
- The first thing on the shelf that is not a body: a nest, and on the ruler, the animal that walked away from it.
~80–66 million years
- The gap
- Some animals cannot be owned in the original at any price, which is exactly what the cast rule on this site is for. Therizinosaurus, with three-foot claws on its hands. Quetzalcoatlus, a pterosaur the height of a giraffe. Diabloceratops, horns on the frill. And the Fighting Dinosaurs of Mongolia, a Velociraptor and a Protoceratops buried mid-struggle, the single most dramatic fossil ever found.
- Why it is empty
- Mongolia does not export fossils, and it should not. Quetzalcoatlus comes from a national park in Texas, Diabloceratops from a national monument in Utah. For all of them a museum-produced cast is not the compromise; it is the only honest object that exists. And a cast on its matrix, the stone reproduced with the bone, is a different thing from a bone on a stand: it shows the burial, not just the animal.
- What would close it
- Documented museum casts from named institutions: a Therizinosaurus claw, a Quetzalcoatlus element, a Diabloceratops skull, and, if one is ever made available, the Fighting Dinosaurs in matrix.
- Closed
- What it adds here
- The registry already says reproductions are here because the originals cannot be owned at any price. These are the animals for which that is most literally true, and the Fighting Dinosaurs would be the first object in the collection that is a moment rather than a thing.
66.0 million years
- The gap
- The asteroid runs through half the records on this site and is not represented by anything. No piece of the impactor survives to be collected. What does survive is its signature: spherules of melted rock that rained out of the sky, shocked quartz, and the thin clay layer carrying the iridium that proved the whole story.
- Why it is empty
- It has been overlooked in favor of the animals it killed. Boundary material with a documented section is available and modest, and it is the only object that would belong to every world in the collection at once.
- What would close it
- A sample of the K–Pg boundary layer, spherule bed or iridium clay, from a named section, with the section logged.
- Closed
- What it adds here
- Every Cretaceous record on this site ends on the same day. This is the day. It would belong to the Kem Kem, Khouribga, and Hell Creek worlds at once, and it is the era boundary on the timeline made into something you can hold.
~55–45 million years
- The gap
- The largest shark that ever lived is here as a tooth and has no ancestors on the shelf. Otodus obliquus is where the line begins: a broad-toothed shark with cusps on either side of the blade that, over thirty million years, lost the cusps, grew serrations, and became megalodon. And it comes from the Khouribga phosphates, the sea already built on this site.
- Why it is empty
- It costs less than lunch. This is the cheapest entry on the list and one of the most connected, and it has simply never been picked up.
- What would close it
- An Otodus obliquus tooth from the Khouribga phosphates, cusps intact, locality named.
- Closed
- What it adds here
- It links TMC-24-003 backward to the Khouribga world in a single object, which is exactly the kind of connection this site was built to draw and cannot currently draw. A second, later tooth, Otodus chubutensis, would complete the line.
- The gap
- Most of what is here is a piece: a tooth, a claw, a bone, a shell. A piece asks you to imagine the rest. A whole animal, head to tail in one stone, does not.
- Why it is empty
- It was not, as it turned out. This entry once said nothing here was an entire creature, and that was wrong.
- What would close it
- A complete animal, articulated, from a named locality.
- Closed
- Already, by the Diplomystus from Fossil Lake, Wyoming (TMC-26-011): a whole fish, every bone of it lying where it lay in the lake fifty million years ago, skull to tail fin. It is not alone. The trilobite (TMC-26-013) and the brittle stars (TMC-26-023) are whole animals too. What this entry used to ask for, an insect in amber, now has its own entry below.
- What it adds here
- The first skeleton on the shelf that needs nothing imagined: the whole animal, as it was.
~44 million years
and Miocene
~20 million years
- The gap
- Everything here that was alive is bone, tooth, shell, carbon, or stone that took their place. Nothing still has its body. Baltic amber is fossil resin from an Eocene forest, and the insects caught in it are complete: wings, eyes, legs, the hairs on the legs. Dominican amber, younger, does the same in a clearer stone. It is also the Cenozoic, which the registry barely holds.
- Why it is empty
- Fakes, and priorities. Amber is among the most imitated fossil materials there are, and the inclusions are the part most often faked. The test is documented origin, a real inclusion identified at least to order, and a piece large enough to hold rather than a bead. Burmese amber is left off this list on purpose. And for now, the money is going to space.
- What would close it
- Baltic or Dominican amber with a documented insect inclusion, identified, in a piece meant for a hand and a loupe, not a necklace.
- Closed
- What it adds here
- The first thing in the collection that still has its eyes.
~33 million years
- The gap
- Between the asteroid and the Ice Age the registry holds a fish, a shark tooth, and nothing that walked. The White River Badlands of South Dakota and Nebraska are the answer: the best mammal beds in North America, and the only place a family can own a complete skull of an extinct animal for the cost of a single dinosaur tooth.
- Why it is empty
- Not chosen. Oreodonts, sheep-sized grazers that lived in herds, come out of private ranch land as whole skulls, legally and with locality. Mesohippus, the three-toed horse the size of a dog, comes out of the same beds as jaws.
- What would close it
- An oreodont skull from the White River Group, ranch and county named. A Mesohippus jaw from the same beds.
- Closed
- What it adds here
- The first complete skull of an original animal in the collection, and the horse would make a story the kids already know, the horse getting bigger and losing its toes, into something on a shelf.
3.2 million to 300,000 years
- The gap
- The timeline ends at a gold dot for humans, and the registry holds no human ancestor and nothing a person ever made. The Dimetrodon entry began a line and this list has followed it through the Hell Creek mammal; it stops before it arrives.
- Why it is empty
- Originals are not possible and should not be. Every hominin fossil belongs to the country it was found in. But museum casts of the famous adults, Lucy, the Neanderthal from La Chapelle-aux-Saints, Homo erectus from Turkana, are made for exactly this, and a documented Acheulean handaxe from an old European collection is legal, common, and half a million years old.
- What would close it
- A museum cast of a named adult hominin skull or skeleton, institution documented. A handaxe from a named site with a collection history.
- Closed
- What it adds here
- The handaxe would be the first object on this site made by a hand. The skull would be the reader’s own family on the shelf, and the end of the sentence that begins with the sail.
~45,000–8,000 years
The North Sea
- The gap
- TerraMar means land and sea, and every object here came out of the land. Nothing was lifted from the sea floor. But there is a place where the two are the same thing: Doggerland, the plain that joined Britain to the continent until the ice melted and the North Sea rose over it eight thousand years ago. Mammoths, bison, woolly rhinoceros, and the people who hunted them lived on ground that is now a hundred and thirty feet down (40 m), and trawlers dredging the Brown Bank and Dogger Bank bring their bones up in the nets.
- Why it is empty
- Care, not law. Dredged North Sea bone is sold openly in the Netherlands and Britain, dated by carbon, and comes with the bank it was trawled from. The care is at this end: in California the object must be bone, not tusk and not tooth, and it must come with its dredging record, because a bone with a story and no paper is only a bone. The search is for a piece that is unmistakably the animal, from a named bank, with the trawl on paper.
- What would close it
- A mammoth, bison, or woolly rhinoceros bone dredged from a named bank of the North Sea, dated, documented, legal in California. A Doggerland flint tool if one ever comes with its record.
- Closed
- What it adds here
- The first object in the collection that came from under the sea, and it came from a drowned land, which is the name of the collection made literal. For a family that loves the water more than the land it stands on, this is the entry, and the Ice Age entry beside it would gain an animal that a fisherman, not a quarry, gave up.
~50,000–10,000 years
- The gap
- Sixty-six million years of mammals taking over the world, and the registry holds a fish, a shark tooth, and one cat. The Ice Age is almost the only chapter where an animal can be pictured breathing next to a person. A mammoth. A woolly rhinoceros. A dire wolf. A cave bear. Something from the last cold.
- Why it is empty
- Law, and that is fine. In California the ivory statute reaches mammoth tusk and, as written, teeth. So the object is bone: a tibia, a femur, something from the leg that carried the animal across the tundra. The woolly rhinoceros is the same story twice. Its horn was keratin, not bone, and survives only in permafrost, so the real ones are in museums; and the same California law that reaches ivory names rhinoceros horn, with no exception written in for a rhinoceros that has been extinct for ten thousand years. The object is a skull cast, horn and all, from a named museum. A dire wolf skull from La Brea is a museum cast and nothing else, from the same pits as the Smilodon. Cave bear teeth from the caves of Romania are legal and inexpensive.
- What would close it
- A mammoth or bison limb bone from a documented Alaskan or Yukon locality, legal in California. A woolly rhinoceros skull cast with the horn, institution named. A La Brea dire wolf skull cast. A cave bear canine, cave named.
- Closed
- What it adds here
- Smilodon is the only animal here that lived beside people. These are the others. The rhinoceros would be the first animal in the collection whose most famous part is the one that cannot be owned, which the label should say. Together they make the last page of the timeline a world rather than a single cat, and the line in gold stops being alone.
1662
- The gap
- The dodo. Every extinction on this site was the planet’s doing; this one was ours, within written history, on an island where sailors wrote down what the bird was like. There is almost nothing left of it: one head and one foot in Oxford, scattered bones, no complete skin anywhere.
- Why it is empty
- There is no original to own. But a maker in England produces casts from the Oxford material, a few a year, and that is the honest object: the bird as it survives, not as it is imagined.
- What would close it
- A cast of the Oxford dodo head, maker and source documented.
- Closed
- What it adds here
- The timeline runs three billion years and ends with a gold dot. This is the first extinction the gold dot caused, and the registry should be willing to say so.
England
~200–66 million years
- The gap
- Half the story of how fossils were understood happened on one island, and the collection holds no British original: no Lias, no Wealden, no Chalk, no Jurassic Coast. In 1842 Richard Owen coined the word dinosaur for three English animals, Megalosaurus from the Stonesfield Slate of Oxfordshire, Iguanodon from the Sussex Weald, and Hylaeosaurus from the same quarries. In the same decades Mary Anning was taking the sea dragons out of the Lyme Regis cliffs: the first complete ichthyosaur, and in 1823 the first plesiosaur anyone had ever seen. Every one of those animals is absent here in the original. The Megalosaurus is a cast of the Oxford jaw. Nothing on the shelves was pulled from an English cliff.
- Why it is empty
- It has been coming, and it is not all Waiting. Isle of Wight Iguanodon bone comes out of the Wealden on the foreshore with the bed named, because the collectors there care about that. Lias ichthyosaur and plesiosaur vertebrae from Charmouth and Whitby are common and honest. Megalosaurus is the hard one: original Stonesfield bone or a tooth surfaces rarely, and when it does it will not wait. Hylaeosaurus is one animal in London and exists for a collection only as a cast, and there are matrix casts of the English founders, large ones, that show the slab as well as the bone.
- What would close it
- In order of likelihood: a Lias ichthyosaur or plesiosaur vertebra, bed named, now. An Isle of Wight Iguanodon bone, documented. A Megalosaurus tooth or bone from the Stonesfield Slate, which would be taken the day it appeared. A Hylaeosaurus matrix cast from a named institution. Any documented British original counts.
- Closed
- What it adds here
- It puts the country where the science began on the globe at the top of the page, and it turns the Megalosaurus cast from a lone Englishman into the first of the three the word was invented for. The sea dragons add something the land animals cannot. The plesiosaurs on this site are Moroccan, from the last sea before the asteroid. Anning’s plesiosaur is the same family a hundred and thirty million years earlier, on the other side of an ocean that did not yet exist. Marine reptiles crossed the world in a way no dinosaur could, and an English vertebra beside the Khouribga teeth would draw that line across the whole Mesozoic.
South America
- The gap
- A continent with its own dinosaurs, its own giant crocodiles, its own Cretaceous, and nothing from it here. Giganotosaurus, the Crato Formation insects, the Patagonian sauropods that are the largest animals that ever lived.
- Why it is empty
- Brazil has prohibited the export of fossils since 1942 and Argentina since 2003, and both are right to. What exists legally is old: Crato and Santana material that left before the law, with paperwork that shows it. Most of what is sold does not have that paperwork, which is why this entry is not marked Now.
- What would close it
- A pre-1942 Crato Formation specimen with documented collection history, or a museum cast of a named Argentine dinosaur from a named institution. Nothing without the paper.
- Closed
- What it adds here
- A fifth continent on the globe. It is also the place where the registry’s rule, that where a thing came from matters more than what it is, will be tested hardest.
Puerto Rico
- The gap
- The island where one of us was born, and it has almost nothing in it that a collector would recognize: volcanic rock from a Cretaceous island arc, and limestone from the Oligocene and Miocene seas that formed over it. But that limestone is made of what lived in those seas, reef corals above all, and a coral from a Puerto Rican sea is a piece of the water one of us was born beside.
- Why it is empty
- Nobody looks. There is barely a market, which is the good news and the difficulty at once. Shark teeth from the island are scarce and rarely surface; fossil coral is the honest object, and it does come up. The Caribbean keeps a larger surprise in reserve: mosasaur remains have been reported from Cretaceous rock in Cuba, so who is to say what the arc holds. Anything here will come from a documented locality, with the island’s own protections for its heritage checked first.
- What would close it
- A fossil coral from the Oligocene or Miocene limestones of Puerto Rico, locality documented, collected legally. A shark tooth if one ever appears with its record.
- Closed
- What it adds here
- The globe at the top of the page gains a place that is not a fossil site but a birthplace. For this collection that is the whole point.
Named, not held
The story of the planet is bigger than any shelf. These are the animals and plants the records above name but the collection does not hold: some protected by the countries they come from, some belonging to the public, some so rare that every one is in a museum, some simply not here yet, and some gone within the time people have been writing things down. Wherever the text names one, the name is a link to it here. Each entry says when and where it lived, why it matters, why it is not on a shelf, and how far in time it stood from the objects that are.
Glossary
The words the labels are written in. Not a dictionary, but a way in: each term here is something the registry above actually uses, explained the way a museum wall would explain it.
About TerraMar
A collection born from a child's wonder, built to preserve three billion years of Earth's story alongside one family's own.
This registry is described rather than shown, on purpose.
A tooth is a tooth until you know that three inches is near the ceiling for its kind. A rock is a rock until you know it was once alive, and that the oxygen in the room came from what it did. A stone that fell from the sky is a stone until you know it is older than the Earth under your feet, older than any number anyone can hold in their head. What makes an object matter is never the look of it. It is what it was first. Was it alive, did it walk, did it breathe. If it is a copy, who made it, from what, and why. The registry is built to answer those questions before it asks you to look at anything, because imagination creates ownership. A picture only confirms it.
That is why the glossary exists, and why the names are written out so a child can say them, and why every entry says how its age is known and what has been restored. The words carry the weight. Imagery is the last part of this, not the first, and when it comes it will not be a photograph of a pretty fossil, because that is not how this story is told. Until then, read, and see it.
The registry tells you what a thing is. Orbit puts it back into the world that made it, and then lets you stand in that world.
Ten Earths wrap a sphere, and scrolling is time: begin at the birth of the planet and move down toward now, and the continents pull apart under your hand. But the reconstruction does not stop at coastlines. Each world carries its own air and its own clock. The Archean day is thirteen hours long, because the Moon was closer and the Earth spun faster, and the page says so as you pass. Oxygen begins as a trace and climbs; carbon dioxide begins at seven hundred thousand parts per million and falls to four hundred and twenty, which is why it is drawn on a stretched scale, because no honest straight line can hold both ends. A thermometer reads what a body would feel at the point beneath you: in orbit it reads the Sun, about two hundred and fifty above in light and two hundred and fifty below in shadow, swinging as the terminator passes under you; descend and it settles to that world's global mean, and says whether that mean is measured, estimated, or contested.
Sixteen named formations are outlined on those worlds: the Pilbara craton, the House Range embayment, the Kem Kem rivers, the Appalachian coal swamps, the Hell Creek plain. So when you come close to an object the page can tell you the ground it came out of, by name, and keep that outline under you as you descend. Four flood-basalt scars burn where the great eruptions were happening at that moment: the Deccan Traps are erupting in India while the asteroid falls on the other side of the planet, and the page says plainly that which one did more of the killing is still argued.
The sky is real, not wallpaper. The Moon starts close and drifts out. The young Sun is fainter and warmer and sweeps a different path. The stars slide until no constellation survives. Rock falls constantly in the early years, meteor showers swell and fade, a comet passes with its tail turned away from the Sun, a lunar eclipse arrives when the geometry actually allows one, and at sixty-six million years the asteroid strikes where it truly hit, with ejecta, fireball, shockwave, fire and dust, and the thermometer spikes with the pulse before falling below freezing under the dark. Descend to the ground and there is a block of text you can reach no other way, about what it was like to be there. The whole soundscape is synthesized in code, not sampled: space, deep ocean, forest, river.
And this is where the collection is shown rather than described. Every pinned object rises into your hands as a hologram drawn from its own proportions, turnable, at the size it really is. Orbit is the home of the imagery. The films live here and nowhere else, loaded only when you open a panel, never on arrival, so the registry stays words and this page carries the picture. Films, scans, whatever the object earns.
It is built for everyone at once. A child can scroll it and watch the continents move. A paleontologist can check the formation outlines against what they know. Someone who came for the sky can stay for the sky. Nothing is dumbed down and nothing is gatekept, and almost everything you can see, you can touch: the Moon, the Sun, the asteroids, the meteors, the comet, every pinned object. Tap one and it has something to say.
That includes the visitor. A small alien craft crosses now and then, rarer once you are down on the ground, and it is there on purpose. Whatever you are when you arrive, scientist or nine-year-old, the thing that brought you is curiosity, and curiosity is a child's instinct that most of us never actually lose. The alien is the page admitting that out loud.
Every event that crosses the sky names its own category as it passes: dated, evidence-based, illustrative, not science. That is the part that has no equivalent anywhere else. A planetarium asks you to believe all of it equally. This page tells you, event by event, exactly how much to, and it is still the most beautiful thing here.
A measurement is a number until you stand next to it. Three inches means nothing. Three inches of tooth, in a jaw, in an animal forty feet long, standing beside you, means something you cannot unfeel.
Scale is one continuous camera and three stops: the animal, the place the object sat inside that animal, and the object itself at true size. The camera never cuts. A tooth becomes a jaw becomes a body; a single leg bone becomes the animal it carried; a two-inch piece of shell becomes the egg, and the egg becomes the fifty-foot armored giant that laid it. You choose what stands beside them: a person five feet seven, a quarter, a door, a school bus, or a blue whale a hundred feet long. Each yardstick carries one true number and nothing else, and stands just past the animal so the pair reads as a pair.
The discipline is in the line itself, and this is the part that makes it a document rather than a picture. Solid means measured. Dashed means estimated. Dotted means reconstructed. The page classifies every span from the record's own words, never from what would look impressive, so the ruler tells you the size of a thing and, in the same stroke, how well that size is known. A height bracket stands beside the animal under the same grammar. A tooth gets a crown-and-root bracket and a human molar at true scale next to it, so the comparison is a tooth against a tooth. Casts are drawn at full size. Reduced-scale replicas, like the tenth-scale skull, live in the registry instead, because a full-size ruler would tell the wrong size.
The animals are drawn as projections, in bone-white line work with blue as emitted light, breathing, scanlines, the occasional data glitch, because they are reconstructions and should look like reconstructions. The reference objects are flat slate diagrams, so the eye sorts them instantly: the glowing thing is inferred, the flat thing is a fact. And the specimen at the end does not glow at all. It is the only thing in the whole sequence that is actually in the collection.
My name is Julio. I am a father of three daughters, and TerraMar began with my second.
Like many children, she fell in love with dinosaurs. But over time I realized it wasn't simply a childhood fascination. She saw them with a sense of wonder that was impossible to ignore. As her father, I found myself drawn into that same curiosity. One fossil became another, then another, and before long we weren't simply collecting prehistoric objects. We were learning the story of our planet together.
Along the way I discovered something else. My wife shared that same fascination, though I hadn't fully appreciated it when we first met. As a family, we found ourselves spending weekends visiting natural history museums, exploring prehistoric life, and asking bigger questions about where we came from and how our world came to be. Now all three of our daughters also share a similar sense of wonder.
Looking back, that path makes perfect sense.
Long before TerraMar existed, I felt most at home outdoors. I grew up hiking, camping, exploring caves, climbing mountains, rappelling, whitewater rafting, hunting, fishing, and eventually scuba diving. My parents raised my brother, my sister, and me with a deep appreciation for the natural world, encouraging us to experience it firsthand rather than simply observe it. Photography came much later, not as the beginning of that journey but as a way to preserve and share the places, wildlife, landscapes, and moments that had already shaped my life. Looking back, TerraMar feels like a natural continuation of that same path. Prehistoric life, meaning dinosaurs, ancient marine reptiles, fossils, minerals, and the deep history of our planet, simply became another way of exploring the world with the same sense of curiosity and wonder that first drew me outdoors as a child.
As the collection grew, I noticed something that surprised me.
Too often, collections become invisible. A fossil is purchased, placed on a shelf, and eventually forgotten. Its story disappears long before the object does.
I wanted something different.
I wanted a place where every specimen could be understood, where every object had a reason for being here, and where my children could one day discover not only what we collected, but why we chose to preserve it.
That idea became TerraMar.
What began as a simple registry slowly evolved into something much larger: a living archive of Earth's history intertwined with our own family's journey. Every entry documents more than an object. It captures a place, a moment in deep time, the evidence behind it, and often the personal story that brought it into the collection.
One realization changed the way I looked at the entire collection.
When I arranged everything chronologically, I discovered that the collection held more than three billion years of Earth's history.
Holding that span of time in your hands is difficult to comprehend. It transforms a shelf of fossils into a timeline of life itself.
TerraMar was designed with longevity in mind. But it is not simple, and it was never meant to be.
Behind these pages are countless hours of reading, checking, and challenging what I thought I knew before anything was allowed onto the site. The sky above the Moon is drawn from a real star catalog, 5,044 stars placed where they truly sit. The continents drift on the same plate model geologists use. Every object is measured, and every entry says how its age is known and what has been restored.
The real work has been taking all of that and making it feel effortless: deep science, translated into something a child can scroll and a scientist can check.
The science itself never stands still. New finds change old answers, and some of what is written here will one day be revised. Each entry is the best understanding available when it was written, and when the science moves, the entry moves with it.
It is built on open web technologies and kept entirely in its own home, depending on no outside service, so it can be preserved, moved, and maintained for decades to come. My hope is that whether someone opens these pages next year or fifty years from now, they will still be able to explore this collection and understand the stories it preserves.
I don't know what TerraMar will look like in the future.
Perhaps my daughters will add to it.
Perhaps it will remain exactly as it is today.
Either outcome is perfectly fine.
Because TerraMar was never built simply to collect fossils.
It was built to preserve curiosity.
To encourage questions.
To celebrate discovery.
And to remind our family that the greatest stories on Earth began long before we did, and will continue long after us.