Asteroid Impacts In Earth History: Sleep Documentary

Full transcript

The limestone country around Chicxulub Pueblo, an inland community on the Yucatan Peninsula in Mexico, lies almost flat under a wide sky. No bowl, no rim, nothing at the surface announces that one of the largest asteroid impacts in Earth's history is recorded directly beneath a crater some two hundred kilometers across, buried about one kilometer down under younger sedimentary rock. It formed slightly over sixty-six million years ago when an asteroid roughly ten kilometers wide arrived. Nothing of that moment shows above the ground.

If nobody was there to see it, how do rocks, drill cores, and moon samples let us read the collisions that shaped Earth and its life? And how much of that reading is still uncertain? If these journeys through deep time help you rest, subscribe so the next ancient world can find you at bedtime. The route ahead follows the evidence rather than the drama. It begins where the impacting bodies came from in the crowded youth of the solar system and moves through the pitted record of the moon, the erased record of Earth, the long list of extinctions written into the fossil sequence, and finally to the buried Mexican crater itself and the patient, argumentative people who worked out what it was.

Settle in for a slow walk through that evidence. An impact event in the plainest definition is a collision between astronomical objects that produces measurable effects. Such collisions are not rare accidents in the life of a planetary system. They happen regularly, and the great majority involve small asteroids, comets, or meteoroids whose arrival changes almost nothing. A pebble of interplanetary rock meets an atmosphere, heats, and is gone. The record of that is a streak of light and at most, a scattering of dust settling for weeks afterwards.

What makes the rare large collision so consequential is speed. Nothing can strike Earth gently. The minimum possible impact speed is eleven point two kilometers per second, which is Earth's escape velocity, roughly forty thousand three hundred and twenty kilometers per hour. That floor exists because a body arriving from elsewhere has already been accelerated by Earth's gravity on the way in. Even an object drifting in from rest at an enormous distance gains that much before it reaches the ground.

Anything approaching with orbital motion of its own arrives faster still. Atmospheres soften the arrivals. Earth's envelope of nitrogen and oxygen shields the surface from most meteoroids, breaking them up and burning them away during entry. But the shield is thin compared with a mountain. Large bodies keep enough energy to reach the ground and to do substantial damage when they arrive. What they leave behind are impact craters and impact structures, and those features turn out to be the dominant landforms on solid bodies throughout the solar system.

Their sheer prevalence is the strongest empirical evidence available for how often and how hard this process operates. To read that evidence properly, it helps to know what is doing the striking and where it lives. Return then about four point five billion years to the solar nebula, the disk of gas and dust from which Earth and most other bodies in the solar system formed. In the region between the present orbits of Mars and Jupiter, small bodies called planetesimals were accumulating in the ordinary way.

They gathered dust, met one another gently, stuck, and grew. Nothing about that region marked it out as strange. The planetesimals there were thought to have evolved much like the rest of the objects in the nebula, growing steadily towards something larger. Then Jupiter approached its present mass, and the arithmetic of the region changed. Gravitational resonance is the mechanism. When a small body orbits in a simple ratio with a giant planet, it receives a repeated tug at the same point in its circuit, and those small nudges accumulate rather than cancel.

Excitation from orbital resonances with Jupiter stirred the belt planetesimals into steeper, faster-crossing orbits. Over ninety-nine percent of them were ejected from the region entirely. What survived was a thinned, agitated remnant. Accretion into a planet in that zone was prevented outright by the large gravitational perturbations of Jupiter, so the asteroids now there may simply be leftovers of the protoplanetary disk, a planet's worth of building material that was never permitted to finish.

The scale of that loss is difficult to hold in the mind. The primordial population of the main belt was probably about 200 times what it is today. Almost everything that was once there is somewhere else now, thrown outward into the dark or inward toward the Sun and the young terrestrial planets. Two independent lines of reasoning support that early sorting. The first is simulation. Models of the belt's history reproduce the observed population only when Jupiter's disruptive influence is included.

The second is a discontinuity in the asteroids themselves. Above a diameter of approximately 120 kilometers, or about 75 miles, the bodies show a different distribution of spin rates and different spectral properties than the bodies below that size. The reading of that break is that objects larger than roughly 120 kilometers accreted during the early era and survived from it, while the smaller ones are fragments produced by collisions between asteroids during or after the Jovian disruption.

The belt is therefore a mixture of originals and wreckage, and the wreckage vastly outnumbers the originals. A few bodies grew large enough to become worlds in their own right. Ceres and Vesta melted and differentiated, which means that heavy metallic elements sank inward to form a core, while lighter rocky minerals were left above in a mantle and crust. That process is the same one that gave Earth its iron heart, running at a much smaller scale and stopping much sooner. The popular picture of the asteroid belt needs correcting before going further.

In that picture, a spacecraft threads between tumbling boulders at close quarters, dodging constantly. The reality is close to the opposite. The main belt lies roughly two to four astronomical units from the Sun in a ring far wider than the distance between Earth and the Sun, and its contents are spread through an enormous volume. The belt is estimated to contain between 1.1 and 1.9 million bodies larger than one kilometer, along with millions of smaller ones, and it is still mostly empty.

Reaching any particular asteroid without aiming carefully would be improbable. A traveler crossing that region would most likely see nothing at all. The mass tells the same story. Adding every asteroid together gives roughly 2.39 times 10 to the 21st kilograms, which sounds enormous until it is compared with something familiar. That total is only about 3% of the mass of the Moon. All the material in the belt gathered up would not make a respectable moon, let alone a planet. Most of that modest total sits in very few objects.

Ceres alone accounts for something like thirty-nine to forty percent of the belt's mass. It measures about nine hundred and forty kilometers across by one account and about nine hundred and seventy- five kilometers by another. And the honest statement is that figure sits in a range rather than at a point. Ceres has a crust, a mantle, and a core, and may carry a surface layer of ice. It is a dwarf planet on a stable orbit, not a wanderer and not a threat. When the Dawn spacecraft arrived, it found a heavily cratered surface, but with fewer large craters than models had predicted.

The largest confirmed crater, Kerwan Basin, is two hundred and eighty- four kilometers across, while models had expected ten to fifteen craters larger than four hundred kilometers. The favored explanation is viscous relaxation, the slow flattening of a crust that behaves over long enough spans a little like something soft. Vesta and Pallas both measure just over five hundred kilometers across, and after Ceres, they are the next great pieces of the belt. Vesta holding about eleven percent of its mass and Pallas about eight point five percent.

Vesta is the differentiated one with a nickel-iron core, an olivine mantle, and a basaltic crust. A small rocky world built on the same plan as the inner planets. It formed inside the frost line and is devoid of water. Near its southern pole lies Rheasilvia, a crater so large that a set of compression fractures encircles the body around it, as though the whole asteroid rang and then held the shape of the ringing. Vesta also delivers itself to laboratories. It is the parent body of the HED meteorites, which make up five percent of all meteorites found on Earth.

That is a remarkable arrangement, a specific object, hundreds of millions of kilometers away, whose crust can be examined by hand in a museum drawer. Ceres offers no such courtesy, since no meteorites from Ceres have been found on Earth. Pallas keeps its own oddity. It rotates on its side with a rotation axis tilted at high angles to its orbital plane in the manner of Uranus. It is high in carbon and silicon, and its interior is described only as perhaps partially differentiated, which is a scientist's way of saying the question is open.

These bodies are not lurking. Most main belt asteroids follow slightly elliptical, stable orbits, revolving in the same direction as Earth and taking three to six years to complete a circuit. They have been doing so for a very long time, and the belt is a place of unhurried arrangement rather than menace. Closer study of the smaller bodies revealed something unexpected about how they are put together, and the clue came from clocks rather than cameras. Astronomers can measure how fast an asteroid spins by watching its brightness rise and fall as it turns.

Do that for enough bodies and a boundary appears. Very few asteroids larger than a hundred meters in diameter have a rotation period shorter than two point two hours. The boundary is not a coincidence of observation, but a consequence of physics. Spin a loose heap of rubble faster than that, and the inertial force at its surface exceeds the feeble gravity holding the surface material down. Loose grains and blocks simply drift away. The near absence of rapid spinners is therefore evidence that most bodies over a hundred meters across are not solid rock at all.

They are rubble piles, gravel and boulders held together by their own weak gravity, unable to survive a brisk rotation. Close-up visits deepened the picture. The asteroid two five three Mathilde is about fifty kilometers across and is saturated with craters whose diameters equal the asteroid's own radius. Impacts of that size should shatter a coherent body. A loosely bound heap, by contrast, absorbs the blow, compacts, and stays together, which is thought to be why Mathilde still exists to be photographed.

The much larger five eleven Davida, some three hundred kilometers across, shows a similarly angular profile, suggesting that it, too, is saturated with craters the size of its own radius. These are not smooth spheres. They are battered planetesimals and fragments of larger bodies, and only the very largest asteroids are roughly round. Their surfaces are soft in an unexpected way. Both Mathilde and 243 Ida, when observed up close, reveal a deep layer of regolith, the fine broken debris produced by ages of small impacts.

A footprint there would not press into rock. It would press into dust that has been stirred and re-stirred for billions of years. Sunlight and the solar wind work on that dust as well. Asteroids become darker and redder with age through space weathering. But the change is not gradual across the whole of solar system history. Most of the color change happens rapidly within the first hundred thousand years after a fresh surface is exposed. In practical terms, a bright new scar on an asteroid does not stay bright for long, which means a surface can look ancient while being, in geological terms, recently made.

Hygiea holds the most interesting unfinished argument among the large bodies. It is the largest carbonaceous asteroid lying relatively close to the plane of the ecliptic and carrying three to four percent of the belt's mass. Observations with the sphere imager on the very large telescope revealed that Hygiea has a nearly spherical shape, which is unusual for a body of its class. Roundness of that kind can be produced by hydrostatic equilibrium, in which a body's own gravity pulls it into a ball, or it can be inherited from a time when the body was in that state, or it can arise from a catastrophe.

Hygiea has no crater on its surface large enough to be the source of the family of fragments that shares its orbit. One reading of that absence is that Hygiea was completely disrupted by an impact and then re-coalesced, losing a little under two percent of its mass in the process and settling back into a sphere. The observations do not discriminate between these histories. The round shape is real, and the reason for it remains open. Collisions leave other signatures, too. Rubble piles, asteroid moons, binary pairs, and scattered asteroid families are all thought to result from impacts that disrupted a parent body.

As of October 2021, eighty-five near-Earth asteroids were known to have at least one moon, and three of those had two. The asteroid 3122 Florence, one of the largest potentially hazardous asteroids at four point five kilometers across, carries two small companions between one hundred and three hundred meters in size, found by radar imaging when it passed near Earth in 2017. So the belt is a slow, thinly spread archive of an early violence, and the same violence continues at a much reduced rate.

To measure that rate and to see what an unshielded planetary surface looks like after four billion years of it, the record has to be read somewhere that erases nothing. The record-keeping is only half the story because collisions did not only subtract. The planet on which all of this reading is done owes some of its own arrangement to arrivals from elsewhere. It is the third planet from the sun, and it remains the only astronomical object known to harbor life. Its early history included at least one event of a scale that no later impact has approached.

The formation of the Earth and Moon as a paired system has been hypothesized to be the result of a giant impact, a collision so large that the material thrown clear of the young Earth went into orbit rather than falling back and settled there into a companion. That companion now keeps a very steady distance. The Moon orbits at three hundred eighty- four thousand four hundred kilometers, which is two hundred and thirty-eight thousand eight hundred and fifty-five miles, and light crosses that gap in one point two eight seconds.

A signal sent from the surface of Earth reaches the Moon in rather less time than it takes to draw a slow breath. The Moon is roughly a quarter as wide as Earth, which makes it unusually large for a satellite of a rocky planet, and that oddity is one of the reasons the giant impact idea has held its place. A quiet caution belongs here. The idea is a hypothesis offered to explain a set of measurements, not a scene anyone witnessed. Other planets carry marks that have been read the same way.

Uranus and Venus both rotate backwards relative to the general spin of the solar system, and interplanetary impacts have been proposed to explain that reversal. A world turning the wrong way is a strange thing to account for by gentle processes. A single enormous blow is one candidate explanation, and it remains a proposal rather than a settled finding. Two further suggestions reach further into the personal. Impacts have been suggested as the origin of water on Earth, delivering it from bodies that formed in colder regions and carried ice.

Impacts have also been proposed as deliverers of the chemical building blocks for life, and the theory known as panspermia rests on exactly that premise. Both propositions should be held lightly. They are ideas under examination, offered because the sources of Earth's ocean and Earth's first chemistry are genuinely uncertain and not because anyone has traced a particular molecule to a particular arrival. What is clearer is the timing. The formation of the ocean and the first development of life both occurred within the first billion years of Earth's history.

Whatever the water's origin, it was here early, and something living was in it early too. Life then spread globally and began to alter the atmosphere and the surface of the planet itself, a slow chemical rewriting that led to the Great Oxidation Event two billion years ago. A world that may have been partly assembled and partly watered by collisions went on to change its own air. Reading any of that from Earth's own rocks is difficult because Earth is restless. The Moon is not. That is where the record was eventually opened, and it was opened by hand.

The landing sites for the last three crewed lunar missions were not chosen for scenery. Apollo fifteen, sixteen, and seventeen were sent to positions near the Imbrium, Nectaris, and Serenitatis basins, respectively, three of the great circular scars on the near side of the Moon. The reasoning was simple. Material thrown out of a basin lands around it, so a crew standing near the rim of one is standing on debris from the event that made it. The astronauts collected rock, bagged it, labeled it, and carried it home.

Among what came back were impact melt rocks. These are rocks that were heated past melting by the energy of a collision and then cooled again, and they carry a particular gift for anyone trying to date an event. Many radiometric clocks are reset by melting. The isotopic composition of a rock records, in effect, the moment it last solidified. Melt a rock in an impact, let it freeze, and the clock starts again at the instant of the impact. When those returned samples were dated, the results did something unexpected.

The rocks had last been molten during impact events falling within a rather narrow interval of time, not spread evenly across the Moon's whole existence as a steady rain of collisions would produce, but bunched. The majority of these impact melts are thought to have formed during collisions of asteroids or comets tens of kilometers across, the sort of arrivals that open impact craters hundreds of kilometers in diameter. A basin, in other words, and then another basin, and then another. In what looked like a short span, the ages clustered between about four point one and three point eight billion years ago.

Investigators drew the obvious inference from that clustering and postulated an intense bombardment, proposing a dramatic increase in the rate of impacts around three point nine billion years ago. They gave it a name with some weight to it, the Lunar Cataclysm, also known as the Late Heavy Bombardment. If the melts really did derive from Imbrium, Nectaris, and Serenitatis, then not only those three prominent basins but many others, placed in sequence on stratigraphic grounds, would have formed within the same brief window.

On Earth, that interval corresponds to the Neo-Hadean and Eo-Archean eras, a time when the ocean was young and life was either beginning or about to. Mercury, Venus, Earth, the Moon, and Mars are all held, under the hypothesis, to have been struck during the same episode. It should be said plainly that this is a hypothesis and nothing firmer. It was considered controversial when it was first proposed, and although it has gained wide credence since, definitive evidence remains elusive. The moon rocks are real, the dates are real, and the interpretation placed upon them is still being argued.

The case in favor has grown stronger in one particular way. Apollo sampled a small and deliberately chosen part of one hemisphere, which is exactly the sort of sampling that can produce a false pattern. Lunar meteorites offer a correction because they are thought to sample the lunar surface at random. Some fragment of the moon is knocked loose by a collision, wanders, and eventually falls to Earth, and there is no reason for such fragments to come preferentially from the places where astronauts happen to walk.

Many feldspathic lunar meteorites probably originated from the lunar far side, a region no crew ever visited. Impact melts within those meteorites have been dated, and consistent with the cataclysm hypothesis, none of their ages was found to be older than about three point nine billion years. Beyond the moon, the oldest surfaces in the inner solar system tell a compatible story. The cratering records of Mercury, of the Moon, and of the southern highlands of Mars all record a period of intense early bombardment around three point nine billion years ago.

Three separate worlds keeping three separate records, none of them capable of erasing much. The case against is not weak either. Some researchers doubt the heavy bombardment altogether. Their argument is statistical rather than geological. If a single very large impact scattered meltrock widely across the moon, then samples collected from several sites might all record that one event while appearing to record many, and the apparent clustering of ages would be an artifact of what got picked up rather than a signal of what happened.

A range of other evidence suggests a more extended period of lunar bombardment instead, lasting from approximately four point two billion years ago to about three point five billion years ago, which is a long, grinding decline rather than a spike. There is also a complication in the lunar meteorite ages themselves. Rather than clustering tightly at three point nine billion years, they span a range. Several hypotheses have been offered to explain the apparent spike in impactor flux, and no consensus yet exists among them.

The most popular of the mechanisms is the Nice model, which is popular among planetary scientists without being established. In that model, the giant planets underwent orbital migration, shifting their positions after formation. Migration of that kind disturbs everything held in resonance with the migrating planets. The model has objects scattered out of the asteroid belt, out of the Kuiper Belt, or out of both into eccentric orbits that carried them across the paths of the terrestrial planets.

A rearrangement in the outer solar system, in other words, delivering a bill to the inner one. One detail of that model leaves a possible fingerprint. In the Nice model, many Kuiper Belt objects are captured in the outer asteroid belt at distances greater than two point six astronomical units from the sun. Most were later ejected by Jupiter. Those that remained may be the D-type asteroids, dark bodies now sitting in the outer belt, whose origin would lie far beyond it. If that reading is right, part of the outer solar system is quietly parked among the asteroids and has been for four billion years.

None of this could be worked out from Earth's own surface, and the reason is written across the planet in water and moving rock. Earth is an ocean world. Water covers seventy point eight percent of the crust, and the remaining twenty-nine point two percent is land. The planet is rounded into an ellipsoid with a circumference of about forty thousand kilometers or twenty-four thousand nine hundred miles, and it is the densest planet in the solar system. Its average surface temperature is currently fourteen point seven six degrees Celsius, which is fifty-eight point five seven degrees Fahrenheit, a temperature at which water sits comfortably in all three of its states somewhere on the globe.

That mildness, that abundant liquid, and the shielding envelope of nitrogen and oxygen described earlier all conspire against the preservation of impact scars. The deeper problem is the crust itself. Earth's outer shell is broken into slowly moving tectonic plates, and those plates interact along their edges to produce mountain ranges, volcanoes, and earthquakes. The same machinery that builds the Himalaya and lights the volcanic arcs is a machine for destroying craters. A basin can be buried under sediment, folded and deformed until its circular outline is unrecognizable or carried on a plate to a trench and pushed down into the interior of the planet, where it ceases to exist as a feature at all.

Rain, rivers, ice, and wind finish whatever the plates leave behind. Earth does not so much lose its craters as recycle them. Bodies without that machinery keep everything. Impact craters are the dominant geographic feature on the moon, on Mercury, on Callisto and Ganymede, and on most small moons and asteroids. On those surfaces, nothing arrives to fill a hole, and nothing carries it away. So four billion years of arrivals simply accumulate, crater on crater, until later impacts overprint earlier ones.

Where surface geology is active, visible craters become scarce. Earth, Venus, Europa, Io, Titan, and Triton all belong to that second class, worlds busy enough to keep resurfacing themselves. The consequences for Earth's tally are severe. About one hundred and ninety terrestrial impact craters have been identified, ranging in diameter from a few tens of meters up to about three hundred kilometers, which is one hundred and ninety miles. Their ages run from the very recent, such as the Sikhote- Alin craters in Russia, whose creation was actually witnessed in 1947, back to more than two billion years.

Most are less than five hundred million years old, not because the bombardment began then, but because geological processes tend to obliterate anything older. They are found selectively in the stable interior regions of continents, the quietest places on a restless planet. Few undersea craters have been discovered at all, partly because surveying the sea floor is difficult, partly because the ocean bottom changes rapidly, and partly because plate tectonics subducts ocean floor into Earth's interior.

The rate of crater production since the early bombardment has been considerably lower, but it remains appreciable. On average, Earth experiences from one to three impacts large enough to produce a crater twenty kilometers in diameter every million years. Set that modeled rate against the one hundred and ninety craters actually cataloged, and the arithmetic indicates there should be far more relatively young craters on the planet than have so far been found. Somewhere out there, under farmland or forest or a kilometer of water, the missing ones are waiting or have already been unmade.

Finding them requires knowing what one looks like, which was not obvious for most of the history of geology. An impact crater is a depression in the surface of a solid astronomical body formed by the hypervelocity impact of a smaller object. Its usual signature is a raised rim surrounding a floor that lies lower in elevation than the surrounding terrain. Material is excavated from the center and piled around the edge, so the ground near the rim is literally the ground from the middle, turned over and thrown outward.

A volcanic crater is a different thing entirely, produced by explosion or by internal collapse, and it forms in a landscape that is otherwise volcanic. An impact crater can appear anywhere at all, in any rock, with no volcanic company whatsoever. Their shapes vary with circumstance. Impact craters are typically circular, which surprises people who expect an oblique arrival to gouge a long scar. The energy release at these speeds is closer to an explosion than to a scrape, and explosions are round.

Craters can nevertheless be elliptical or even irregular, where events such as landslides have altered them after the fact. The size range is extraordinary. At one end sit the microscopic craters visible on lunar rocks returned by the Apollo program, pits made by grains of dust traveling fast enough to matter. In the middle are simple bowl-shaped depressions. At the far end lie vast, complex, multi-ringed impact basins, structures with concentric rings of uplifted rock hundreds of kilometers across.

Meteor Crater is a well-known example of the small end on Earth, a clean bowl in the ground. Where surface processes have destroyed most of the original topography, the vocabulary changes. Geologists then speak of an impact structure or use the older and more evocative term astrobleme, which describes a feature that is no longer a whole but still carries the shocked and shattered rock of one. Much of Earth's record survives only in that condition. The older vocabulary encodes an assumption that the newer terms have quietly dropped.

Before the significance of impact cratering was widely recognized as a geological process, features that are now understood to be impact related were called crypto explosion structures or crypto volcanic structures. Both names contain a hidden verdict. A crypto volcanic structure is one whose explosive origin is concealed but presumed to lie below in the earth in some deep and unseen chamber. The possibility that the energy came from above, from outside the planet entirely was not in the vocabulary because it was not yet in the thinking.

Daniel M. Barringer, a mining engineer, was convinced already in 1903 that the crater he owned, the one now known as Meteor Crater, was of impact origin. Conviction is not the same as consensus, and the discipline took a long time to follow him. Barringer's certainty has to be understood in its context because the catalog he was arguing his way into barely existed at the time. A mining engineer reads ground for a living. He looks at a hillside and thinks about what was moved from where and by what force.

Barringer owned the Arizona bowl outright, walked it, and drew the conclusion that something had come down out of the sky and made it. In the first years of the twentieth century, that conclusion had almost no standing among geologists. The prevailing habit of mind looked downward for causes. Barringer looked up and then spent his money on the proposition. The catalog that eventually grew around that idea is smaller than most people expect. What the catalog actually represents is narrower than a count of arrivals.

Each confirmed structure passed through several successive filters before being assigned a name and a date. The ground had to be struck, and the structure had to survive in recognizable form. Someone had to examine it who was trained to read the diagnostic signs, which for most of geological history almost nobody was, and the broader community had to accept the identification before it entered the shared record. These filters operate unevenly across different environments. Stable continental interiors where the ground changes slowly and ancient rock surfaces remain exposed pass structures through all the filters far more readily than the dynamic margins of continents do.

The youngest entries in the catalog are also the easiest to identify because the original morphology has not yet been smoothed or buried. Older structures earned their place only because something unusual preserved them against the general trend. A thick cap of resistant rock, an unusually dry climate, or a chemistry that kept the shocked minerals legible in the record long after the landform itself had disappeared. Impact structures turn up selectively in the stable interior regions of continents.

The parts of the crust that have been sitting quietly for a very long time without being crumpled at a plate margin or drowned under new sediment. Those regions are also conveniently the places where geologists have mapped most thoroughly, where rock is exposed, where drilling records accumulate, and where a faintly circular arrangement of shattered stone is likely to be noticed by somebody. Preservation and attention overlap. The map of known craters is partly a map of where the evidence survives and partly a map of where people have gone looking.

The oceans expose that second effect plainly. Water covers most of the planet, so most arrivals land in it. Yet very few undersea craters have been found. Three obstacles work together. Surveying the sea floor is difficult and slow compared with walking across a field with a hammer. The ocean bottom changes quickly, sediment shifting and settling over anything that dents it. And the sea floor is not permanent. Pl Plate tectonics carries it steadily toward the trenches and pushes it down into the interior of the planet, taking whatever it holds with it.

Two-thirds of Earth's target surface is ocean floor, and the ocean floor is not a stable record. It is generated at mid-ocean ridges, carried across the basin by spreading, and consumed at subduction zones on a timescale of a few hundred million years at most. Any impact structure formed on that surface rides toward eventual destruction, and the catalog of known structures is therefore a severe undercount. Researchers who use the catalog treat it not as a survey result but as a lower bound, a number that establishes the minimum while the true total remains well out of reach.

The shortfall is not a reflection of careless searching or poor methodology. It reflects the fundamental asymmetry between the rate at which impacts arrive and the rate at which the surface they strike is preserved long enough to be studied. The missing structures are not lost through ignorance alone. They are lost because the planet is better at destroying evidence than the discipline is at finding it, and the planet has had far longer to practice. It means the same three things the geography already implied.

Some of the missing craters lie under ocean. Some lie under later rock and vegetation and ice, and some are sitting in plain sight in landscapes nobody has examined with impact in mind. The record is incomplete in a way that can be estimated but not yet filled. There is a further complication in treating the rate as a constant because it is not one. The cratering rate in the inner solar system fluctuates, and the reason lies back in the belt. When two sizable asteroids collide, they do not simply dent one another.

They produce a family of fragments sharing similar orbits, and gravitational nudges and slow orbital drift can send a portion of that family cascading inward over subsequent millions of years. A single collision far away therefore raises the local delivery rate long afterwards, the way a landslide upstream keeps sending material down a river for years after the slope has stopped moving. One such event has a name and an approximate date. The Baptistina family of asteroids formed in a collision about eighty million years ago, and that family is thought to have caused a large spike in the impact rate.

The hedging in that sentence is deliberate and belongs to the sources. The collision is inferred from the family, the spike is inferred from the collision, and the whole chain is described as something thought rather than something established. What the example demonstrates cleanly is the principle. The number of things crossing Earth's path is not fixed. It rises and falls according to accidents in a region hundreds of millions of kilometers away, and those accidents have their own timetable.

Whether the same fluctuation applies further out is not assumed. The rate of impact cratering in the outer solar system could be different from the rate in the inner solar system since the population of impacting bodies out there is different, moving at different speeds under the influence of different planets. The tidy picture of a single solar system cratering clock does not survive contact with the details. All of this concerns holes in rock. The reason so much effort has gone into counting and dating them is that at least one of these events is thought to have altered the history of life, and that raises a harder question about how the history of life is measured at all.

An extinction event, also called a mass extinction or a biotic crisis, is defined as a widespread and rapid decrease in the biodiversity of Earth. It is identified in the record by a sharp fall in the diversity and abundance of multicellular organisms. The underlying condition is arithmetic. Species are always being lost at some slow background rate, and new ones are always appearing through speciation. A mass extinction occurs when the rate of extinction rises sharply relative to both of those so that loss outruns replacement across many unrelated groups at once.

What that looks like in the field is a stack of sedimentary rock in which the fossils above are conspicuously fewer and conspicuously different than the fossils below. Counting such events turns out to be surprisingly contentious. Estimates of the number of major mass extinctions in the last five hundred and forty million years range from as few as five to more than twenty. The disagreement is not about whether the losses happened. It is about two prior questions that must be settled before counting can start.

What qualifies as major, and which data properly measure past diversity? Different answers to those questions produce different lists, and the lists have never fully converged. The most influential answer came in a landmark paper published in 1982 by Jack Sepkoski and David M. Raup. Working through compilations of the fossil record, they identified five particular geological intervals with excessive diversity loss. Their method was statistical rather than dramatic. Extinction rates through the Phanerozoic, the eon of abundant animal fossils, show a general trend of decline, and against that declining trend, five intervals stood out as outliers.

Those five became the Big Five, a phrase that has outlived a good deal of the reasoning that produced it. Later work applying more stringent statistical tests to a much larger body of data has confirmed the general picture while softening the outline. Multicellular animal life in the current Phanerozoic Eon has experienced at least five major mass extinctions and many minor ones. But the Big Five cannot be defined as cleanly as the name suggests. They appear now to represent the largest or some of the largest points on a relatively smooth continuum of extinction events rather than five isolated catastrophes standing apart from a calm background.

The line between a major extinction and a minor one is drawn by the analyst rather than by nature. The original method also had a specific blind spot. Sepkoski and Raup initially tracked absolute extinction rather than proportional extinction, counting how many groups vanished rather than what fraction of the standing diversity vanished. In an interval when total diversity was low, an event can remove most of what exists and still produce a small absolute number. Several events in the Cambrian and early Ordovician meet or exceed the Big Five in proportional severity.

Yet overall diversity remained rather low until the great Ordovician biodiversity event filled the seas with far more kinds of animal. Measured in absolute terms, those early crises simply did not register, and the biodiversity estimates of 1982 passed over them. Older still and larger by some readings than anything in the animal record is a crisis with no shells or skeletons in it whatsoever. Early in the Proterozoic eon, life itself altered the atmosphere, and the resulting Great Oxidation Event, sometimes called the Oxygen Catastrophe, is presumed to have caused a far more extensive mass extinction of microbial life than any of the big five.

Free oxygen was a poison to much of the biosphere that produced it. There is also a suspected extinction event at the end of the Ediacaran just before the Cambrian explosion, whose magnitude is unknown and which may have ushered in the Phanerozoic. The deep history of loss extends well below the level where fossils are easy to count. Of the five great intervals that Sepkoski and Raup marked out, the first in time sits at the close of the Ordovician between four hundred and forty-five and four hundred and forty- four million years ago.

The world then was a marine world. The animals that mattered lived in shallow seas and on the sea floor, and the continents were largely bare of the vegetation that would later cover them. There were no dinosaurs anywhere on Earth, nor anything remotely like them, and there would not be for hundreds of millions of years. What existed were reefs, shells, and swimming and crawling invertebrates in seas that had recently become unusually rich through the great Ordovician biodiversification event.

The extinction that followed was probably not a single blow, but two, spaced around the Ordovician and Silurian transition. The first pulse corresponds to a cooling event and to a large glaciation in the southern part of the world. Ice locks up water, sea level falls, and the shallow shelf seas where most of that diversity lived shrink or drain entirely. The second pulse coincided with the thaw when the ice released its hold and the shelves flooded again. Both directions of change proved lethal, which is a pattern worth holding on to.

Stability was the habitat. The transitions in either direction were the crisis. Together, the two pulses removed twenty-seven percent of all families, fifty-seven percent of all genera, and eighty-five percent of all species. Those three figures rise as the categories narrow, which is what a severe extinction looks like statistically. Whole families often survived through a single surviving genus. By the percentage of genera lost, many scientists rank the late Ordovician event as the second largest of the five major extinctions.

The cause, however, is under active argument, and the argument has moved recently. The reading of cooling and glaciation as the driver rests on numerous studies, particularly for the first pulse. Then studies in May 2020 suggested something close to the opposite for the underlying trigger, proposing global warming related to volcanism combined with anoxia, in which seawater loses its dissolved oxygen. That proposal sits at odds with the earlier work rather than replacing it. More recent suggestions have offered a route that accommodates both, with the deposition of volcanic ash driving reductions in atmospheric carbon dioxide, which in turn produced the glaciation and the anoxia that the rocks record.

No asteroid appears anywhere in that account. There is no iridium layer, no shocked mineral horizon, no crater of the right age proposed as the agent. The second largest of the great extinctions, by many counts, is explained entirely by processes belonging to Earth itself, and even so, the explanation is not settled. The same holds for the next interval, higher in the column of rock and roughly seventy-three million years later in time. The Late Devonian mass extinction is dated in these sources to three hundred and seventy-two million years ago, and the Late Devonian was an epoch of high diversity loss concentrated into two major extinction events rather than one.

The setting again is quiet water. Reefs stood in warm, shallow seas built by organisms rather different from the corals of later ages, and the sea floor communities around them were dense and long established. Scientists have linked both of the Devonian extinctions to anoxic events, intervals in which seawater is deprived of oxygen. An anoxic ocean is not a violent place to look at. The water darkens, the sediment beneath turns black with unoxidized organic matter, and the animals that need dissolved oxygen simply fail to persist.

The record of it is a change in the color and chemistry of mud laid down without any drama at all, above which the reef builders are diminished. Here, the account has to stop, and the reason should be stated plainly rather than disguised. The available sources describe only two of the big five in any detail, the Late Ordovician and the Late Devonian, and the excerpt covering the Devonian breaks off partway through. They contain nothing usable about the end-Permian event or the end-Triassic event, two intervals of enormous importance to the history of life.

Filling that silence with confident narration would be inventing rather than reporting, so the silence stays. A record that admits its gaps is worth more than one that papers over them, and the same principle governs everything already said about crater counts and lunar dates. What can be drawn from the two events that are documented is a comparison that matters for everything ahead. Both were severe. Both reshaped marine life. Both are explained, so far as they are explained at all, by climate, ocean chemistry, glaciation, volcanism, and the interactions among them.

Neither has an impact attached to it. Of all the great losses recorded in the fossil sequence, exactly one carries a confirmed collision as its cause, and the evidence for that single case had to be assembled from a thin layer of clay, an unremarkable metal, and a buried structure that nobody had been looking for. The body that arrived carries a peculiar distinction. It is one of the best studied objects in the history of the solar system, and not a macroscopic fragment of it has ever been recovered.

Everything known about its size comes from the size and shape of the wound and from the debris it scattered, which is why the sources do not agree on a figure. One account gives a diameter of about ten kilometers, six miles. The international panel of scientists that reviewed the case in March twenty ten gave a range of ten to fifteen kilometers, six to nine miles. The impactor most likely fell somewhere in that band, and ten to fifteen kilometers is the responsible reading of the crater evidence.

Scale comparisons help more than the raw number does. Mount Everest stands just under nine kilometers from sea level to summit. The object was therefore at least the height of the greatest mountain on the planet, and possibly half as much again, traveling as a single mass through empty space. Set beside larger things, however, it shrinks at once. Phobos, the inner moon of Mars, measures twenty-two kilometers across, which makes the Chicxulub impactor a modest object even by the standards of small bodies.

Nothing about it was exceptional except its trajectory. Speed did the rest. The floor of eleven point two kilometers per second already described applies to every arrival at Earth, and a body on an independent orbit arrives faster than the floor. The energy released has been put at the equivalent of one hundred million megatons of TNT, which is four point two times ten to the twenty-third joules. The comparison offered in the sources is with the atomic bombs dropped on Hiroshima and Nagasaki, and the collision released over a billion times their energy.

Those figures do not need any raising of the voice. They are simply what happens when a mountain-sized mass is not slowed by anything at all. The date has been sharpened considerably in recent decades, and the sharpening is a small lesson in how such numbers behave. The event is described as occurring slightly over sixty-six million years ago. Paul Renne of the Berkeley Geochronology Center reported a figure of sixty-six million and thirty-eight thousand years ago, plus or minus eleven thousand years, obtained by argon-argon dating, a radiometric method applied to material whose isotopic clock was reset at the moment of the event.

An uncertainty of eleven thousand years on sixty-six million is roughly one part in six thousand. That is a remarkable grip on a moment that no instrument witnessed and no observer recorded. Anyone who learned about this event some time ago may carry a different number. The impact was commonly cited as having happened about sixty-five million years ago, and that figure was standard until Renne and colleagues published their revision in twenty thirteen and moved the accepted value to sixty-six million.

The older number is not a myth or an error of principle. It was the best available answer, and it was replaced by a better one, which is the ordinary way such things go. Work on the identity of the impactor has been narrower and quieter. In nineteen ninety-eight, Shukolyukov published isotopic evidence in Science bearing on the Cretaceous tertiary impactor and its type. Isotopes serve as a fingerprint of origin, since different classes of body carry different proportions of certain elements, and debris preserved in the boundary rocks retains that signature.

The question being asked was not how large the object was, nor how fast, but what family it belonged to. Answering the larger questions required going down into the structure itself, and that took until twenty sixteen. The drilling project of that year targeted the peak ring, the inner ring of uplifted rock that forms within the largest impact structures. Rings of that kind exist elsewhere in the solar system in abundance, and on Earth almost nowhere, since the surface here is not in the business of preserving them.

The Chicxulub peak ring is the only one intact and directly accessible for scientific research. Its two hundred kilometers of buried structure place it among the very largest on the planet, in company with the Sudbury and Vredefort structures, both of which are far older. A drill core comes up as a series of stone cylinders laid out in order of depth like a bookshelf read from the bottom. What the drillers pulled from the peak ring was granite. Granite forms deep within the earth, and it is emphatically not typical seafloor rock, which is what the shallow tropical setting of the Yucatan should have provided.

The granite had been shocked and partly melted, and the reading of it is that this deep material was lifted to the surface in minutes. Rock that belongs kilometers down had been raised, rearranged, and set again within the span of a short walk. Above and around that uplifted rock lay sand. Sand in a marine sedimentary sequence is a record of moving water, and the deposits recovered from those cores record colossal seawater movement directly after the impact. The ocean did not stay where it had been.

The evidence of that is not a description in any chronicle, but a change of grain size in a cylinder of rock read by people leaning over a bench. Most telling of all was something the cores did not contain. The target rocks of the Yucatan should have included gypsum, a common sulfate bearing mineral, and gypsum is nearly absent from the recovered material. An absence in a drill core is ordinarily a disappointment. Here it was the point. The inference is that the gypsum was vaporized outright and dispersed into the atmosphere as an aerosol sulfur compounds lofted high above the weather and spread around the world.

Rock that should have been there had left the ground and joined the sky. The conclusion drawn from that absence was stated in careful terms, and the exact language should be quoted directly. The cores confirmed the presence of a probable link between the impact and global longer term effects on climate and the food chain. Not a proven chain of cause and effect from crater to dying forest. A probable link established by a missing mineral three decades after the argument began. Widely accepted is the phrase the sources use for where that argument now stands.

It is now widely accepted that the devastation and climate disruption resulting from the impact was the primary cause of the Cretaceous Paleogene extinction event. The magnitude of that event is given as the loss of about seventy-five percent of plant and animal species on Earth, including all non-avian dinosaurs. Another account puts it at roughly seventy percent of all species besides the dinosaurs, and both figures rest on the same fossil record, with the true proportion sitting somewhere in that neighborhood.

Somewhere around three quarters of the living world was removed. The fourth quarter carried everything forward. Rene went further than the date alone. He posits that the mass extinction of the dinosaurs occurred within thirty-three thousand years of the impact date, which sounds like a long time in a human sense, and is a fraction of a percent of the interval separating that moment from this one. The word posits belongs in the sentence. It is a position argued from dating, not a stopwatch reading.

What happened in the hours and days afterwards is the part everyone wants, and it is the part the sources decline to supply. They offer the sand deposits, the vaporized sulfate, and the phrase devastation and climate disruption. An hour by hour account with skies described and temperatures named would be composition rather than evidence, and nothing of the kind will be attempted here. The rocks are specific about some things and silent about others, and the silence has been earned. The other half of the consequence is easier to overlook because it is measured in what followed rather than what stopped.

The same event is believed to have accelerated the evolution of mammals, which were present throughout the Cretaceous in modest roles and inherited a great deal of vacated space. That acceleration led in time to mammalian dominance, and mammalian dominance eventually set in place the conditions for the rise of humans. No hominid stood anywhere on Earth when the asteroid arrived, nor for tens of millions of years afterwards. Every human being who has ever lived is downstream of a collision, which is a strange inheritance to hold and a quiet one.

All of this rests at the bottom on a layer of clay. At Gubbio in Italy, within an ordinary sequence of sedimentary rock, there runs a thin band that marks the boundary between the Cretaceous and the Paleogene. It is clay, not volcanic ash, and the distinction matters because ash would point to an eruption. A comparable band appears at Caravaca in Spain and again near Raton in New Mexico, which serves as the type locality of the iridium anomaly, the reference section against which other occurrences are compared.

The same thin layer of clay has been traced worldwide, sitting at exactly the same position in the sequence wherever the rocks of that age survive. To look at, the layer offers nothing. It is a narrow stripe of fine sediment, a few fingers wide in a cliff face, distinguishable from what lies above and below mostly by texture and color. The iridium that made it famous is invisible. It does not glint. It forms no metallic film, and no amount of staring reveals it. Its presence is known only by chemical analysis, in which the concentration comes out around a hundred times greater than normal for Earth's crust, and at Gubbio as much as one hundred and sixty times the background level.

Different sections give different multiples, and reports from around the world describe spikes of varying size, so the anomaly is a family of measurements rather than a single value. Two other things hide in the same clay. Microscopic grains of shocked quartz occur within it, quartz whose internal structure has been deformed by pressures that ordinary geology does not supply. In places there are small weathered glass beads thought to be tektites, droplets of melted rock thrown clear and chilled in flight.

These are not features anyone notices while walking past. They are found by dissolving samples, sieving residues, and putting the leftovers under a microscope. Around the buried Mexican structure, the same signatures recur, with shocked quartz and tektites in the surrounding areas and instruments registering a gravity anomaly, a subtle variation in the pull of the ground that betrays a change in the rock below. The record is not perfectly tidy, and one publication makes the untidiness explicit.

In nineteen eighty-nine, Graup and Spettel described the mineralogy and phase chemistry of an iridium-enriched layer from the Latengebirge in the Bavarian Alps that sits below the boundary rather than at it. An iridium enrichment in the wrong stratigraphic position complicates any account in which one layer everywhere means one event. The literature accommodated it as literature does by publishing it and arguing about it, and it stands as a reminder that boundary sections are real rock with real irregularities rather than a diagram.

The idea that a collision ended the age of dinosaurs did not begin with the clay. In a nineteen fifty-three publication, the geologists Alan O. Kelly and Frank Dahill analyzed global geological evidence and suggested that one or more giant asteroids had struck Earth, causing an angular shift in the planet's axis, global floods, firestorms, atmospheric occlusion, and the extinction of the dinosaurs. Some of that has aged better than the rest. What the proposal lacked was any strong confirming evidence, and it joined other earlier speculations about impact events that had the same difficulty.

An idea without a measurement attached to it tends to stay where it was put. The measurement came from a father and a son. In the late nineteen seventies, the geologist Walter Alvarez was working on the Gubbio sequence, and his father, Luis Walter Alvarez, was a Nobel Prize-winning physicist. Their reasoning was almost embarrassingly simple in outline. Iridium is much more abundant in meteorites than in Earth's crust. If an unusual quantity of iridium appears in a thin layer laid down worldwide at a single moment, an extraterrestrial source becomes the economical explanation.

The mechanism they proposed was that the impactor vaporized on arrival, its material spread through the atmosphere, and it settled out across the surface of the planet among the other debris, producing the iridium-enriched clay. They did not do the chemistry alone. Frank Asaro and Helen Vaughn Michelle, chemists at the University of California, Berkeley, performed the analyses that turned an idea into a number, and the four of them published in Science in June 1980. Almost simultaneously and independently, the same explanation was arrived at by the Dutch paleontologist Jan Smit, who with Jan Hertogen had published iridium findings from Caravaca in Spain in Nature in May 1980.

Two teams, two countries, two sections of rock, the same conclusion within weeks of each other. That kind of convergence is rare, and in retrospect, it is one of the strongest features of the case. Reception was another matter. No consensus existed at the time on what had caused either the extinction or the boundary layer, and the rival explanations were serious ones. A nearby supernova was proposed, which would also deliver unusual elements. Climate change was proposed, so was a geomagnetic reversal, a flip of the planet's magnetic field.

Against that field of candidates, an asteroid was one hypothesis among several, distinguished chiefly by a chemical anomaly whose interpretation was itself contested. The sharpest objection came from paleontologists, and it came from the fossils rather than the chemistry. Many rejected the impact hypothesis on the basis of what became known as the three-meter problem. In the sections examined, dinosaur fossils were scarce in the last few meters of rock beneath the boundary, which suggested to those researchers that the animals had already been dwindling before the layer was laid down.

A gradual die-off does not require a sudden cause. The dispute was therefore not obstinacy, but a genuine conflict between two kinds of evidence, one chemical and global, the other biological and local, and the two did not agree. Argument at that intensity produces paper. Reports of similar iridium spikes at the boundary arrived from sections across the globe, and interest in the cause of the extinction became one of the largest questions in the earth sciences. Over two thousand papers were published on the topic during the nineteen eighties.

Through all of that, one item remained conspicuously missing. There were no known impact craters of the right age and the right size anywhere on Earth. The hypothesis had a chemistry, a worldwide layer, and no hole. The hole was found by people who were not looking for a hole. That is the shape of this part of the story, and it took a decade for the two halves to meet. Through the 1980s, the search ran in one direction. A worldwide layer implied a source. The source implied a structure of considerable size, and the structure had to be somewhere.

Nothing in the catalog matched. The known craters of roughly the right dimensions were far too old, and the craters of roughly the right age were far too small. That absence is what spurred the hunt, and the hunt was conducted by people who believed a crater existed because the chemistry required one. Meanwhile, in the late 1970s, two geophysicists named Antonio Camargo and Glen Penfield were flying and mapping the Yucatan Peninsula on behalf of the petroleum industry. Oil exploration is a matter of reading buried shapes without seeing them.

Instruments record the faint variations in the pull of the ground and in its magnetism, and from those variations, a picture of the rock below is assembled. What emerged from the Yucatan surveys was not the sort of structure petroleum work is looking for. It was a vast buried arc, symmetrical, far larger than any geological feature the region should contain. Penfield thought it was a crater. He could not prove it. The samples that would have settled the question were not available to him, and the material he could reach did not carry the diagnostic evidence he needed.

Without that evidence, a buried circle remains a curiosity in a survey report. He gave up the search, and the structure stayed in the files, unconnected to the argument that was consuming the earth sciences a discipline away. The connection was finally made through contact with Alan R. Hildebrand in 1990. Through that contact, Penfield obtained samples, and the samples indicated an impact origin. Ten years of searching for a crater and rather more than ten years of a crater sitting unclaimed in the records of an oil survey resolved at last into the same object.

There is a lesson in that about how discoveries actually arrive. The material was already collected. What was missing was a conversation between two groups of specialists who had no obvious reason to speak. The formal verdict came two decades later. In March 2010, an international panel reviewed twenty years of accumulated literature on the extinction. The team numbered forty-one scientists, and in reaching its conclusion, it also ruled out competing explanations, including massive volcanism, which had been a serious contender for a long while.

The panel endorsed the impact hypothesis and named the buried Mexican structure as the cause. Even so, the language in the sources does not claim more than the evidence establishes. The site is described as now considered the likely impact site and as the place where evidence indicates the asteroid fell. That is the language of a strong inference rather than a witnessed fact, which is exactly what it is. A great deal of chemistry, drilling, dating, and gravity data all point at one place on the Gulf Coast of Mexico, and the people closest to the evidence still write the sentence with a hedge in it.

There is one further observation about that whole enterprise. Every person in it knew what an asteroid was. That knowledge is younger than most of the buildings in a European city, and it begins on a winter night in Sicily. Before 1801, the solar system, as understood, consisted of the Sun, the planets and their moons, and comets. There was no category for a small rocky body in an independent orbit because no such body had ever been seen. What existed instead was a suspicion about an empty space, and the suspicion came from arithmetic.

In 1772, the German astronomer Johann Elert Bode published a numerical progression citing the earlier work of Johann Daniel Titius. The scheme is now called the Titius-Bode Law, and it is now discredited, but at the time it looked uncanny. Set Saturn's distance from the Sun at one hundred parts. Then Mercury falls at four, Venus at four plus three, which is seven, Earth at four plus six, which is ten, and Mars at four plus twelve, which is sixteen. The next term is four plus twenty-four, which is twenty-eight, and at twenty-eight there was nothing.

Beyond the gap, the sequence resumes obediently, with Jupiter at four plus forty-eight, or fifty-two, and Saturn at one hundred. Bode found the vacancy intolerable and said so in language that belongs to his century rather than to modern journals. "Can one believe," he asked, "that the founder of the universe had left this space empty?" Certainly not. The formula placed the missing world at an orbital radius near two point eight astronomical units, roughly four hundred twenty million kilometers from the Sun.

Then William Herschel discovered Uranus, sitting close to the distance the progression predicted for a planet beyond Saturn, and a numerical curiosity acquired the standing of a law. Astronomers began to take the gap seriously as a place with something in it. In 1800, a group led by Franz Xaver von Zach, editor of the German astronomical journal Monatliche Correspondenz, decided to organize a hunt. Requests went out by letter to twenty-four experienced astronomers, whom von Zach dubbed the Celestial Police, asking each to search a portion of the sky methodically for the expected planet.

The whole apparatus of the search was paper and post. A coordinated observing campaign across Europe moved at the speed of a courier. One of the astronomers selected was Giuseppe Piazzi, a Catholic priest working at the Academy of Palermo in Sicily. His invitation had not yet reached him on the night of one January 1801, when he was at his instrument doing something entirely routine. He was checking positions against a star catalog, looking for the eighty-seventh star of the catalog of the zodiacal stars of Mr.

Lacaille, and he found that it was preceded by another. His own account of that first sighting is beautifully undramatic. "The light," he wrote, "was a little faint and of the color of Jupiter, but similar to many others which generally are reckoned of the eighth magnitude." Therefore, he had no doubt of its being any other than a fixed star. Only the following nights unsettled him. On the evening of the third, he recorded, his suspicion was converted into certainty, being assured it was not a fixed star.

Nevertheless, before making it known, he waited until the evening of the fourth, when he had the satisfaction to see it had moved at the same rate as on the preceding days. Three nights of watching a faint point shift its place and a deliberate pause before announcing anything. Piazzi observed the object twenty-four times in all. The final observation came on eleven February 1801, when illness interrupted his work and ended the sequence. On twenty-four January, he had written to just two colleagues, his compatriot Barnaba Oriani in Milan and Bode in Berlin.

He reported the object as a comet and then quietly undercut his own classification in the same letter. "Since its movement is so slow and rather uniform," he wrote, "it has occurred to him several times that it might be something better than a comet." In April, he sent the complete set of observations to Oriani, to Bode, and to the French astronomer Jérôme Lalande, and the information was printed in the September issue of the Monatliche Korrespondenz. By then, the object was effectively gone.

Its apparent position had shifted, mostly because Earth had moved along its own orbit, and it now lay too close to the glare of the Sun for anyone to confirm what Piazzi had seen. A discovery announced is not a discovery kept. Without a predicted path, the faint point would have to be found again by luck. The path was supplied by a mathematician of 24. Carl Friedrich Gauss developed an efficient method of orbit determination, worked the problem in a few weeks by hand from Piazzi's scattered positions, and sent the resulting prediction to von Zach.

On 31 December 1801, von Zach and a fellow celestial policeman, Heinrich Olbers, pointed their instruments where the calculation said to point and found the object near the predicted position. Almost exactly a year after Piazzi first noticed it, the thing was recovered, and it has never been lost since. Its distance from the Sun came out at 2.8 astronomical units, which fitted the progression almost perfectly. Piazzi named it Ceres Ferdinandia in honor of the patron goddess of Sicily and of King Ferdinand of Bourbon.

For a short while, Europe had a new planet, filling a gap that a discredited formula had insisted upon. The status did not last because Ceres turned out to have company. Von Zach's group went on to find Pallas, then Juno, then Vesta, the last of them in 1807. Each behaved like the first. With the equipment of the time, they appeared as points of light, indistinguishable from stars, showing little or no planetary disk, and the only thing separating them from the background was their apparent motion from night to night.

A planet that cannot be resolved into a disk sits awkwardly in the category of planets. Herschel proposed a word for the awkwardness. He coined it from the Greek asteroeides, meaning star-like or star-shaped, derived from aster, a star or planet, and the new bodies became asteroids. The term did not immediately settle the taxonomy. Through the early second half of the 19th century, the words asteroid and planet, not always qualified as minor, were still used interchangeably, and nobody was in a hurry to resolve the ambiguity.

Then the discoveries stopped. After Vesta in 1807, the sky produced nothing new of that kind for 38 years. The professional interest wandered elsewhere, and the gap was eventually closed by an amateur. Karl Ludwig Hencke began searching in 1830 and kept at it for 15 years. He was looking for Vesta when he found something else, a body later named 5 Astraea, and Gauss was given the honor of naming it, which closed a small circle rather neatly. After that, the field filled up quickly. Other astronomers joined the search, and by the end of 1851, fifteen asteroids were known.

That was enough to break the notation. Each new body had been given its own iconic symbol in the manner of the planets, and the symbols had multiplied into roughly two dozen by 1852, often occurring in several competing variants. The Royal Astronomical Society concluded that discoveries were arriving too fast for such a system to survive. Johann Franz Encke made the change in the edition of the Berliner Astronomisches Jahrbuch prepared for 1854. In place of individual glyphs, he introduced a disc or circle, a traditional symbol for a star used generically for any asteroid and carrying a number that recorded its order of discovery.

The next body found, 16 Psyche in 1852, was the first to be designated that way at the time of its discovery, although it was granted an iconic symbol as well. Later that same year, when de Gasparis discovered the twentieth, Benjamin Valz gave it both a name and its rank in the sequence, and 20 Massalia became the first asteroid never assigned a symbol of its own. The practice died out entirely after thirty-seven Fides in 1855. Prestige clung to the hunt for a while. In 1868, when James Craig Watson discovered the hundredth asteroid, the French Academy of Sciences marked the occasion by engraving the faces of the three most successful hunters of the era, Carl Theodor Robert Luther, John Russell Hind, and Hermann Goldschmidt, on a commemorative medallion.

The decisive change came in 1891 when Max Wolf turned photography on the problem. A long exposure holds the stars as points, while a moving body draws itself as a short streak across the plate. So the search no longer required a person comparing positions at the eyepiece night after night. It required a person examining a plate. Wolf alone discovered two hundred and forty-eight asteroids, beginning with 323 Brucia, at a time when barely more than three hundred were known in total. One observer with a camera nearly doubled the catalog.

The work did not bring much glory with it. Most astronomers could not be bothered with these objects, which cluttered plates taken for other purposes and yielded little of interest individually. Some called them the vermin of the skies, a phrase variously attributed to Eduard Suess and to Edmund Weiss. A century after Piazzi, only a few thousand had been identified, numbered, and named. The method that grew out of Wolf's plates stayed recognizably the same for a long time, and it rewarded patience above all.

A region of sky was photographed with a wide-field telescope or an astrograph in pairs, the two exposures typically taken about an hour apart. Several such pairs might be gathered over a series of days. The pair was then viewed under a stereoscope. Because the two plates were taken from slightly different moments, anything orbiting the Sun had shifted position between them, and the stereoscopic view converted that shift into apparent depth. The moving object seemed to float slightly above the fixed background of stars.

A body hidden among thousands of identical points announced itself simply by hovering. Once identified, the object's location was measured precisely with a digitizing microscope against the known positions of surrounding stars. Even then, nothing had been discovered. What the observer possessed was an apparition, a single sighting of something moving, and it received a provisional designation composed of the year, a letter for the half month in which it was seen, and a further letter and number giving its sequence in the form 1998 FJ74.

The last step is administrative and decisive. Locations and times are sent to the Minor Planet Center, where computer programs test whether the new apparition ties earlier apparitions together into a single coherent orbit. If it does, the object receives a catalog number, and the observer of the first apparition with a calculated orbit is declared the discoverer and granted the right to name it, subject to approval by the International Astronomical Union. An object is not real to the catalog until its path is known.

The numbers that system now holds would have astonished the celestial police. As of May twenty twenty-five, the Minor Planet Center held data on one thousand four hundred and sixty thousand three hundred and fifty-six minor planets, of which about eight hundred and twenty-six thousand eight hundred and sixty-four had accumulated enough information to receive numbered designations. As of April twenty twenty-two, twenty-eight thousand seven hundred and seventy-two near-Earth asteroids were known, and eight hundred and seventy-eight of those measured a kilometer or more across.

Those near-Earth objects fall into families defined purely by orbital geometry. The Atiras, also called Apoheles, stay strictly inside Earth's orbit, with their farthest point from the Sun below zero point nine eight three astronomical units. The Atens have a semi-major axis under one astronomical unit and cross Earth's path. The Apollos have a semi-major axis above one astronomical unit and also cross it. The Amors stay strictly outside, with their closest approach to the Sun between one point zero one seven and one point three astronomical units.

Anything crossing the orbital path of Earth is an Earth-crosser. Most of them did not begin there. Gravitational interactions with Jupiter drive them out of the belt and inward, and a small number are extinct comets, bodies that have exhausted their surface volatiles and now behave like rock. Their orbits are further perturbed by the gravity of other bodies and by the Yarkovsky effect, a gentle push arising from the way a rotating body absorbs sunlight on one side and radiates the heat away on another.

Warmth applied over millions of years moves a mountain. For all that precision, the central word remains loose. Asteroid has never been officially defined. Since two thousand and six, the IAU has preferred the term small solar system body, noting that minor planet may still be used, but that the newer phrase is generally favored. The category that began as a word for a point of light that would not resolve into a disk has never been given a firm boundary, and the objects have carried on regardless.

Some of them have been watched arriving. Throughout recorded history, hundreds of Earth impacts and exploding bolides have been reported, and a number of them caused deaths, injuries, property damage, or other significant local consequences. The scale is entirely different from anything in the deep past discussed so far, and the two should not be blurred together. The best known of the modern events occurred in nineteen oh eight over a very sparsely populated part of Siberia in Russia and is called the Tunguska event.

Its remoteness is a large part of why it is remembered as a scientific puzzle rather than as a catastrophe. The Chelyabinsk meteor of twenty thirteen was different in that respect. It is the only known incident in modern times to result in numerous injuries, and it remains the largest recorded object to have encountered Earth since Tunguska. Elsewhere in the solar system, the process has been observed directly. In July nineteen ninety-four, Comet Shoemaker-Levy Nine broke apart and collided with Jupiter, and that sequence provided the first direct observation of a collision between two solar system objects.

Telescopes were pointed at a predicted impact, and the impact obliged. In twenty thirteen, the reach extended beyond the solar system entirely when NASA's Spitzer Space Telescope detected what appeared to be a massive terrestrial planet impact around the star ID8 in the star cluster NGC 2547, a detection later confirmed by ground observations. Spacecraft have meanwhile been going out to the bodies themselves. The first close-up observation of an asteroid was made by the Galileo spacecraft on its way elsewhere.

NEAR Shoemaker studied Eros. Dawn visited both Vesta and Ceres. The Japanese missions Hayabusa and Hayabusa2 went further, studying Itokawa and Ryugu respectively, and returning physical samples of each. And OSIRIS-REx collected a sample at Bennu in twenty twenty and delivered it to Earth in twenty twenty-three. Lucy, launched in twenty twenty-one, is tasked with studying ten different asteroids, two in the main belt and eight among the Jupiter Trojans. Psyche, launched in October twenty twenty-three, is going to the metallic body of the same name.

Tianwen-2, launched in May twenty twenty-five, to explore the co- orbital near-Earth asteroid four six nine twenty-one nine Kamo'oalewa and the active asteroid three one one P PanSTARRS and to bring back regolith from the first of them. One mission was sent to push rather than to look. NASA's Double Asteroid Redirection Test, launched in twenty twenty-one, was crashed deliberately into the small, non-threatening asteroid Dimorphos in September twenty twenty-two as a test of whether the orbit of such a body can be altered by impact.

The European Space Agency's Hera, launched in October twenty twenty-four, is intended to study the results of that collision in detail. A species that spent most of its history unable to recognize a crater has begun, tentatively, to make one on purpose. The reasoning behind that effort has been stated plainly by the people who advocate it. In April twenty eighteen, the B612 Foundation put it as a matter of certainty and timing together, saying that it is one hundred percent certain Earth will be hit by a devastating asteroid, but that it is not one hundred percent certain when.

That same year, the physicist Stephen Hawking, in his final book, Brief Answers to the Big Questions, considered an asteroid collision the biggest threat to the planet. Neither statement asks for alarm. Both describe a process that has been running since the solar system had planets in it, and that is now, for the first time, being counted. None of the ancient arrivals in this account had a witness. No one stood on a ridge and watched a basin open on the moon. No one saw the Yucatan sky, and no one was there to see anything at all for tens of millions of years afterwards.

Everything that has been established about those events was established after the fact, from material that stayed behind. The residue is a strange inheritance to work with, a metal that cannot be seen, hiding in a stripe of pale sediment, a cylinder of stone that belongs kilometers beneath the ground and was found lying much nearer the top. Melted rock, bagged and labeled by gloved hands, carried across a quarter of a million miles and set down on a laboratory bench. Not one of those things is a collision.

Each is a leftover, and leftovers are the entire archive. Building a science on leftovers demands a particular temperament. It requires people willing to spend a working life on a question they can never settle by looking, and willing to be corrected by a sample that has not been recovered yet. It also requires a habit that outsiders sometimes mistake for weakness, which is the open declaration of what remains unproven. That habit runs through every part of the record. An early era of intense bombardment across the inner solar system is still named as a hypothesis, held widely and defended vigorously and not yet closed.

The number of great losses in the fossil sequence refuses to settle, and the people counting them disagree about what should be counted. Even after a drilling ship put a bit through a kilometer of rock and pulled up exactly the evidence the argument needed, the connection between that impact and the long climatic consequences was written down as probable rather than certain. A less careful discipline would have rounded all of it up. Probable would have become proven, hypothesis would have become fact, and the range of species loss estimates would have hardened into a single confident figure repeated everywhere.

Nothing would have been gained by that except a false sense of completion, and something important would have been lost. The hedges are load-bearing. They mark exactly where the evidence stops and where the next question begins, and a reader who trusts the hedges can trust everything standing beside them. There is a quiet dignity in that arrangement. The people who assembled this picture were not, for the most part, dramatic figures. They were a priest checking a star catalog on a winter night, a mining engineer arguing with his own profession, oil geophysicists reading buried shapes for a living, chemists running samples in a Berkeley laboratory, and crews on a drilling platform bringing up stone in the dark.

What they produced together is an account of collisions nobody saw, assembled slowly enough that most of them did not live to see it finished. The picture they left points back to one place on the Gulf Coast of Mexico and to one moment slightly over sixty-six million years ago. Set the pieces of it down gently now. A buried structure under the limestone country of the Yucatán, found because instruments felt the faint pull of what lies beneath. A thin line of clay traced from Italy to Spain to New Mexico and onward around the world, no wider than a few fingers in a cliff face.

An anomaly of a rare metal within that line, measured rather than seen. Lunar melt rocks whose isotopic clocks were reset by collisions, carried home in the last years of the Apollo landings. And around all of it, three decades of argument between people who read fossils and people who read chemistry, neither group wrong to insist on what their own evidence showed. None of that machinery existed on the day itself. The world that received the impact had no notion of asteroids, no vocabulary for craters, and no interest in either.

Return to it briefly and then let it go. The place that would become the Yucatán Peninsula lay under warm, shallow sea. It had lain that way for a very long time, and the sea was doing what such seas do, which is to make rock out of water and time. Fine pale sediment settled through sunlit water and came to rest on what had settled the year before and the century before and the age before. Layer upon patient layer, a platform of light-colored stone thickened beneath that water, holding within it the minerals that shallow tropical seas leave behind.

Far below the platform, unreachable and irrelevant to everything living above it, sat granite. Light reached the sea floor there. Currents moved across it and shifted the finer material into ripples. Things grew, fed, and died, and their remains joined the sediment and became part of the platform in their turn. Beyond the shore, on ground that would eventually be mapped by geophysicists in aircraft, the great land animals of the Cretaceous went about ordinary lives, unaware of being the last of their kind and entirely unaware of anything beyond the sky.

Nothing in that scene was waiting. The object crossing toward it was following an orbit, as it had for four and a half billion years, obeying the same arithmetic that keeps every other small body in its path. It carried no intention and no meaning. It was rock of a size that is unremarkable among rocks in space, moving at a speed that is ordinary among speeds in space, and it happened to intersect a planet. Afterwards, the sea returned. That is the part the cores record most plainly, water coming back over disturbed ground and laying sand where sand had no business being.

Then the ordinary work resumed. Sediment fell again through sunlit water. Fine material settled on the wreckage and buried it, and kept settling, and kept burying until the newest rock stood a full kilometer above the ring of shattered stone beneath. Sixty-six million years of quiet accumulation closed over the whole event as smoothly as water closing over a dropped stone, which is why the country there is flat today. The limestone kept its own counsel. Farmland and scrub and low trees, a wide sky, an inland village that lent the structure its name, and no rim, no bowl, no scar of any kind to suggest what the ground remembers.

The clay line closed over as well. In an Italian cliff, in a Spanish section, in the rocks near Ratón, the sediment simply continued upward after the boundary, recording a world with fewer kinds of animals in it and a great many empty roles. The survivors filled those roles over the following millions of years, and the small furred creatures that had lived modestly through the Cretaceous found the world unusually accommodating. Everything that has read this record is descended from that quarter of life that carried on.

The belt those bodies came from is still turning. Something over a million cataloged objects follow their slightly elliptical paths between Mars and Jupiter, three to six years to a circuit, spread so thinly through so much space that the whole region would seem empty to anyone crossing it. Ceres holds its stable orbit and its possible ice. Vesta carries its enormous southern crater and goes on quietly delivering fragments of its crust to museum drawers. Somewhere among them, family by scattered family, are the fragments of collisions that have not yet arrived anywhere and may never arrive at all.

Earth, meanwhile, keeps doing what it has always done to the evidence. Plates move, rain falls, sediment buries, and craters are unmade at roughly the rate they are made. Beneath cultivated ground and dense woodland and the sediment-covered ocean bed, structures that no catalog has yet named are being worn away by the same processes that removed everything before them. The count of one hundred and ninety will grow slowly and will never be complete. None of that requires any vigilance from a listener.

The orbits are indifferent, the timescales are enormous, and the counting is now being done carefully by people whose whole profession is patience. So let the last of it settle. The buried ring under the pale stone, two hundred kilometers of it, holding its shape in the dark where no light has reached it since before mammals were interesting. The thin gray line in the cliffs, which needs no more than a single candle to read by and which will hold its position in the sequence whether anyone reads it or not. The lunar rocks in their drawers, still keeping the times they were last melted.

The asteroid itself, gone entirely, dispersed into a layer of clay and a scattering of glass beads and an anomaly in a chemical assay, which is the only grave it has. Deep time takes all of it back without effort. The Cretaceous sea closes, the limestone thickens. The question of what happened there is answered as fully as leftovers permit, and the parts that stay open have been marked honestly, and the whole slow argument comes to rest. There is nothing left to keep watch over now. Let the ancient sea go dark above the platform.

Let the sediment fall and blow that candle out. Rest well, sleep deeply, and let the ancient Earth fade softly into dreams.