On a chilly January afternoon, Susannah Maidment stands on the shore of a London lake, staring down a pack of dinosaurs.

Maidment, a curator at the U.K.’s Natural History Museum, has come with me to tour Crystal Palace Park, which in 1854 included the world’s first public dinosaur showcase. The sculptures were a smash hit at their unveiling and sparked the dinomania that’s been with us ever since. More than a century before Steven Spielberg dazzled the world with Jurassic Park, the Crystal Palace dinosaurs drew two million visitors a year for three decades straight, and Charles Dickens name-dropped one in his novel Bleak House.

To grant us a detailed look at these 166-year-old monuments, Ellinor Michel and Sarah Jayne Slaughter, trustees with the nonprofit Friends of Crystal Palace Dinosaurs, guide us through a metal gate to the banks of the lake, where we don waders to make our crossing. I misjudge my first step and fall into the water, clambering onto the island’s shore, dripping wet and smelling of pond scum. “Welcome to Dinosaur Island!” Slaughter exclaims, grinning from ear to ear.

Tucked in among ferns and spongy beds of moss, the pale green sculptures are imposing, even imperious. The park’s two Iguanodon, a Cretaceous herbivore, resemble huge iguanas with nubs on their snouts—which scientists now understand were spikes on their thumbs. It’s tempting to dismiss the assemblage as outdated or the stuff of B movies. But Maidment sees the Crystal Palace dinosaurs for what they really are: the bleeding edge of scientific knowledge at the time, based on comparisons between living animals and the few fossils available to researchers.

Scientists still use this technique to re-create the fantastic beasts, filling in the soft gaps in time-worn fossils. Bones don’t preserve evidence of cheeks on ancient faces, Maidment says, as we pause between two of the statues, “but we reconstruct them as being there because it works: Animals today have cheeks.” The park’s sculptors used the same process, she says. “They were completely reasonable to reconstruct them like this from what they knew.”

In the nearly two centuries since, scientists have learned far more about dinosaurs than the builders of Crystal Palace Park ever could have dreamed. Now our understanding is seeing another revolution—one fueled by a wealth of fresh fossils and innovative research techniques. The resulting scientific bonanza is forcing us to rethink popular visions of these ancient animals.

For several years scientists have unveiled an average of about 50 new dinosaur species a year, a pace unthinkable decades ago. The updated menagerie ranges from pint-size fliers with bat wings to long-necked herbivores that were Earth’s biggest ever land animals. Medical scanners, particle accelerators, and chemical analyses are letting researchers virtually separate rock from bone and see fossils’ tiniest hidden features. From the colors of dinosaurs’ eggs and feathers to the shapes of their brains, our dino encyclopedia now includes unprecedented details on how these animals were born, grew up, and lived.

With these tools in hand, scientists today are not just overhauling our pop-culture notions about dinosaurs; in a sense they are bringing these remarkable creatures back to life. When it comes to dinosaur discovery, “I do genuinely think the golden age is right now,” says University of Edinburgh paleontologist Steve Brusatte.

It’s fitting that dinosaurs are so persistently captivating. For 150 million years they dominated landscapes across ancient Earth, and they lived on what are now all seven continents. Dinosaurs were hugely successful during their reign, adapting into a bevy of shapes and sizes.

Brusatte and others estimate that scientists have cataloged more than 1,100 species of extinct dinosaurs, and that’s just a subset of the species that once lived, because fossilization occurred in only a few environments. Their story continues to this day. When an asteroid slammed into Mexico’s Yucatán Peninsula 66 million years ago and wiped out three-quarters of life on Earth, one group of dinosaurs survived: the feathery creatures we now call birds. (Find out more on how the dinosaurs went extinct.)

Western science has formally studied dinosaurs only since the 1820s, but what we’ve learned reveals a lot about how land animals are affected by our ever changing planet.

As continents drifted apart and recombined—and as temperatures and sea levels rose and fell—dinosaurs persisted. What lessons can we take from their responses and resilience? Telling such an epic story requires a worldwide hunt for dinosaurs’ bones, and from Alaska to Zimbabwe, paleontologists are delivering as never before.

One of the richest regions for new fossil finds is North Africa. Someone sweltering in the 105-degree heat of the Moroccan Sahara may find it tough to imagine that this landscape was once lush with waterways deep enough to host car-size fish. But National Geographic Explorer Nizar Ibrahim and his crew of paleontologists have returned to the region for years, chasing one of the weirdest dinosaurs ever found: a river monster called Spinosaurus aegyptiacus.

The first Spinosaurus fossils were discovered in Egypt in the 1910s but were destroyed in a World War II bombing raid in Germany. Still, surviving field notes, sketches, and photographs of the original fossils, along with a few isolated bones and teeth found later in the 20th century, hinted that this mysterious, sail-backed creature had some kind of aquatic lifestyle. Spinosaurus had conical teeth well adapted for nabbing fish, for example, so paleontologists surmised that it perhaps prowled the shallows and plucked fish out of the water, as herons or grizzly bears do. Ibrahim and his colleagues therefore made a huge splash in 2014 when they described a new partial skeleton of the animal found in Morocco, and used it to make the case that Spinosaurus spent much of its time swimming and feeding in the water.

To buttress the claims, Ibrahim’s team returned to the arid site in 2018 with the support of the National Geographic Society in hopes of finding more parts of the creature. The dig was brutal. To clear tons of rock, the crew bought the region’s only working jackhammer, which broke within minutes, forcing them to seek repairs from the seller of the defective tool. Several team members were hospitalized for exhaustion once they returned home. But fueled by Nutella and the promise of discovery, they started finding one vertebra after another from Spinosaurus’s tail, sometimes just minutes and inches apart. The diggers were so giddy over the trove of fossils, they drummed out musical beats with their rock hammers and broke into song.

Shaped like a paddle some 17 feet long, the unearthed appendage, unveiled earlier this year in the journal Nature, is the most extreme aquatic adaptation ever found in a large predatory dinosaur. It’s a hard-won revelation that stretches the outer bounds of how researchers thought dinosaurs moved through their environments. “This is going to become a symbol, an icon, of African paleontology,” Ibrahim tells me.

The story of Spinosaurus, with its desert vistas and historical intrigue, feels as if it could have been lifted from a movie script. But subsequent research on the fossil tail has shown just how different today’s study of dinosaurs can be.

As part of his work, Ibrahim traveled from Casablanca to Cambridge, Massachusetts, and the Harvard University lab of biologist George Lauder. By his own admission, Lauder is no paleontologist: He specializes in studying how aquatic animals move in water, using high-speed cameras and robots to figure out how they swim. To put Spinosaurus to the test, Lauder mounts an eight-inch-long, orange plastic cutout of the dinosaur’s tail to a metal rod attached to a $5,000 force transducer—part of a robotic “flapper” that dangles from the ceiling.

“It’s like a medieval torture device,” quips Harvard paleobiologist Stephanie Pierce, who designed and led the experiments, as Lauder lowers the robot into a flume.

Once submerged, the mounted tail springs to life, flapping back and forth and sending data from the apparatus to nearby computers. Pierce and Lauder’s results show that Spinosaurus’s tail could deliver more than eight times the forward thrust in water of the tails of related, landlubbing dinosaurs. A beast longer than Tyrannosaurus rex appears to have swum its way through rivers like a crocodile. “Where we started was, a dinosaur paleontologist gets in touch with another paleontologist, who gets in touch with a fish bioroboticist,” Pierce says. “If you want to do modern, cutting-edge research, it takes a team of people from very diverse backgrounds.”

These kinds of interdisciplinary lab experiments now define dinosaur research. Modern computers let scientists crunch through huge data sets of skeletal features and build dinosaur family trees. Close examinations of bone slices thinner than sheets of printer paper reveal, in detail, the length and timing of dinosaurs’ growth spurts. And with the same models used to forecast climate change, paleontologists can virtually sling an asteroid at Earth as it was 66 million years ago, to watch dinosaur habitats shrink in the resulting apocalyptic winter.

Few technologies have so profoundly altered our view of dinosaurs as medical CT scanning, which is now a standard in the paleo tool kit.

“We’ve been able to pull all of these extinct bones into a computer, where we can do things with them,” says Ohio University paleontologist Lawrence Witmer. “We can reconstruct missing bits … and do crash tests and run simulations and better understand how these animals actually functioned.”

Scanning also eliminates a past trade-off: whether to sacrifice a fossil’s soft-tissue imprints to whittle down to bones. Stories abound of dinosaur skin impressions being ground to dust during preparation. Now, researchers virtually cleave bone from rock. “It does make you wonder, what things have we overlooked or bulldozed through?” says Mark Witton, a paleoartist at the U.K.’s University of Portsmouth.

The field’s modern sense of caution has yielded an avalanche of discoveries. Recently Witmer used CT scans to show that major groups of dinosaurs evolved distinct cranial air-conditioning systems to keep their brains from overheating. Armored dinosaurs, such as the ankylosaur Euoplocephalus, relied on their nasal passages, which evolved into ducts shaped like crazy straws to shed heat as the animal breathed, cooling the blood destined for the brain. By contrast, large predators such as T. rex vented excess heat with large snout sinuses. Like blacksmiths working bellows, the dinosaurs flexed their jaws to force air in and out of the chambers, causing moisture to evaporate and wick away heat, like sweat on a summer day.

CT scans also can give us a sense of how dinosaurs moved and changed as they grew. Using x-ray videos and computer animations of alligators and birds, the University of South Florida’s Ryan Carney built 3D models that revealed in 2016 that the feathered dinosaur Archaeopteryx could flap its wings in a way that enabled self-powered flight. And to understand how the Patagonian herbivore Mussaurus grew up, Argentine researcher Alejandro Otero assembled scans of the dinosaur’s bones in a computer to simulate its stance at different ages. Just like human babies, Mussaurus hatchlings walked on all fours and then matured into walking more upright on their two hind limbs.

The deeper paleontologists can look into each new bit of bone, the more they can unravel precious details about the past—and that means they’ve had to seriously scale up their tools.

In the northwestern corner of Grenoble, France, on a triangular spit of land where two rivers meet, a gray ring half a mile around rises out of the smog. It’s as if aliens had touched down in the Alps for some skiing and a spot of fondue. The eerie structure is the European Synchrotron Radiation Facility, which in recent years has become a mecca for paleontologists, thanks to staff researcher Paul Tafforeau.

The ESRF is a particle accelerator that flings electrons around at nearly the speed of light. As the electron beam makes its laps, magnets along the circular track bend the particle stream. The disruption makes the particles give off some of the world’s most intense x-rays, which researchers often use to study new materials and medicines. Tafforeau specializes in using the x-rays to look inside fossils that typical CT scanners can’t make sense of, at resolutions those scanners can’t reach.

As we tour the accelerator’s steel and concrete innards, I ask Tafforeau just how discerning the machine can get. He gestures to a display case, with 3D-printed examples of past fossils he’s x-rayed. Portions of one of them, a burrow more than 250 million years old, were scanned finely enough to resolve details as narrow as a human red blood cell. When conditions are just right, Tafforeau’s scans can show features less than a hundredth of that width. Such is the power of a magnifying glass the size of a football stadium.

With great power, though, comes great responsibility. To illustrate the importance of safety to new students, Tafforeau uses an unfiltered beam to light objects on fire and roast coffee beans. “Most of the beams we are using to scan fossils would kill you in just a few seconds,” he says.

The ESRF’s intensity has worked wonders for Dennis Voeten of Sweden’s Uppsala University, who used it to virtually slice through Archaeopteryx fossils and trace out the cross sections of the bones in exacting detail. Because bones must withstand the strain of flight, their geometric structure can speak to the animals’ flying styles. Though Archaeopteryx’s anatomy didn’t allow for a fully birdlike flap, the cross sections of its wing bones most closely resemble those in living pheasants, which fly in short bursts. It’s a striking hint at how the 150-million-year-old creature—an iconic snapshot of dinosaurs’ evolution into birds—navigated the Jurassic island chains it may once have called home.

Kimi Chapelle of South Africa’s University of the Witwatersrand has used the facility to look inside the world’s oldest known dinosaur eggs, which belong to the southern African herbivore Massospondylus. The x-rays let her reconstruct the embryonic skulls inside, down to tiny teeth the dinosaur would have either shed or reabsorbed before hatching. Modern gecko embryos also have these prototeeth, despite the fact that geckos’ and dinosaurs’ last common ancestors lived more than a quarter billion years ago. Thanks in part to geckos, Chapelle figured out that the Massospondylus embryos made it three-fifths of the way through their incubation before dying—an intimate glimpse into lives cut short more than 200 million years ago. “That makes them feel much more real,” she tells me.

Each spring Beijing’s Institute of Vertebrate Paleontology and Paleoanthropology welcomes its own symbol of life’s ephemeral nature, as a blanket of cherry and plum blossoms unfurls across the Chinese capital. To Jingmai O’Connor, the scene is impossibly charming: Gargoyles sculpted in the likenesses of ancient fish, dinosaurs, and saber-toothed cats look out from the main building onto gaggles of laughing schoolchildren. “It’s like paleontology Disneyland, almost,” the IVPP researcher says.

On the inside, though, the IVPP is more time machine than theme park. Since the 1990s, farmers, researchers, and fossil dealers in northeastern China’s Liaoning Province have brought in hundreds of fossils that have upended our understanding of how dinosaurs looked and behaved. Many preserve traces of feathers that confirm plumage first evolved before dinosaurs ever flew. Some fossils also reveal, in dramatic fashion, that dinosaurs other than birds’ closest ancestors also tried defying gravity.

Few dinosaurs better reflect the constantly shifting picture than the scansoriopterygids (SCAN-soar-ee-OP-tuh-RIH-jidz), an obscure group of Jurassic dinosaurs with a mouthful of a name. Some scientists once thought that the crow-size animals used their four-inch-long fingers to nab bugs, like modern aye-ayes do. But in 2015, IVPP researchers unveiled a bizarre member of the group that turned out to be a lost dead end in the origins of flight. Unlike any other dinosaur ever found, Yi qi had batlike membranous wings that it supported with its long outer fingers and bony wrist spurs. “That’s what the story is: One very important specimen … really just kind of knocked everything that we thought we knew upside down,” O’Connor says.

China’s fossils, and others from equally remarkable sites around the world, preserve vestiges of all kinds of tissue. In 2014 researchers announced they had found an Edmontosaurus regalis, a type of hadrosaur, in western Canada that has a cockscomb of mummified flesh, like what you’d see on a rooster. This is a structure nobody knew the dinosaur had, despite knowing of the species for nearly a century. Dinosaurs’ bones had shown that the creatures used exaggerated body parts to woo mates and jockey for social status, just like modern animals, or to find their kin. With Edmontosaurus and other dinosaurs bearing soft tissues, paleontologists are seeing hints of these displays’ true splendor.

In a few cases, researchers can even infer some of the animals’ original chemistries. In 2008 scientists led by paleontologist Jakob Vinther, now of the U.K.’s University of Bristol, figured out that melanosomes, tiny subcellular sacs filled with the pigment melanin, could fossilize. The discovery opened the door to a field once thought impossible: figuring out the colors of extinct dinosaurs’ skin and feathers, based on the shapes, sizes, and arrangements of their melanosomes.

These reconstructions come with caveats: Living animals employ other pigments besides melanin, and some extinct dinosaurs probably did too. Even so, recent finds have been astonishing. The feathered dinosaur Anchiornis, which lived in what is now China, had a reddish crest; the early ceratopsian Psittacosaurus had red-brown skin that contributed to an early form of dino camouflage. In 2018 an international team reported that the feathers of Caihong, a dinosaur that lived in the same region as Yi qi, once shimmered with all the colors of the rainbow.

Even more of life’s molecules may last through deep time. In the 2000s, North Carolina State University paleontologist Mary Schweitzer made waves when she found that some dinosaur fossils, including T. rex specimens, contained preserved cells, blood vessels, and maybe even vestiges of proteins. Ever since, Schweitzer and a growing cohort of scientists have asked how such substances could persist—and what we could learn from them.

At her lab in New Haven, Connecticut, Yale Ph.D. candidate Jasmina Wiemann shows me how she grinds up a small piece of Allosaurus bone for analysis. She transfers the dust into a tube and invites me to add an acid solution, which fizzes and turns a deep brown. “It reminds me always a little bit of Coca-Cola,” she says. Under a microscope, the gunk left behind includes spongy mahogany chunks shot through with black squiggles. I can’t believe what I’m seeing. The brown schmutz was once protein-rich tissue. The squiggles? The outlines of bone cells that lived more than 145 million years ago in a toothy Jurassic predator as long as a school bus. After millions of years, heat and pressure often will transform these kinds of microscopic remains through chemical reactions. Despite their altered state, the materials hold invaluable clues to dinosaur behavior. In a 2018 study Wiemann showed that when certain dinosaur eggshells are struck with a laser, the light that scatters back reveals degraded protoporphyrin and biliverdin, compounds that give modern bird eggs color and speckling.

Based on such analysis, the calcified eggs of the Velociraptor relative Deinonychus had a bluish hue, suggesting that like modern birds with similarly colorful eggs, the dinosaur had open-air nests and brooded its young. By contrast, fossilized Protoceratops embryos found in Mongolia and Mussaurus embryos from Patagonia are surrounded by what was once leathery eggshell, according to a study Wiemann co-authored this year. The find suggests not only that these dinosaurs buried their nests like modern sea turtles do, but also that the first dinosaurs’ eggs were similarly soft. That adds a twist to dinosaurs’ evolutionary story, since it implies hard eggshells—which are found throughout the group Dinosauria—must not have a common origin. Instead, the trait evolved at least three times.

More than anything else, scientific advances show us that dinosaurs weren’t the one-note menaces we sometimes see in popular culture. They courted each other with elaborate displays and squabbled for social status. They suffered broken bones and infections. They snapped after bugs and nibbled on ferns. Their days were as rich and varied, frenzied and humdrum, as those of the birds outside our windows. Even the biggest, baddest T. rex sometimes took a nap.

It’s a realization that hits me as I walk through the lab of Yale assistant professor Bhart-Anjan Bhullar, whose cramped office is just down the hall from Wiemann’s. Bhullar might work in a geology department, but he’s taken only three geology classes in his life. He studies fossils and the embryos of living animals to try to unlock how ancient dinosaurs begot birds.

If anyone could genetically tweak a bird and make a chickensaurus, it’s probably Bhullar. In one 2012 study he found that bird skulls are riffs on ancient dinosaurs’ juvenile skulls, developmentally speaking: Young dinosaur skulls had thinner bones and more flexibility, which birds leveraged to evolve beaks. Parts of the old tool kit survived. Bhullar also has shown that blocking the beak’s key molecular pathways can give chicken embryos Archaeopteryx-like snouts.

Across the avian body plan, Bhullar has found other striking examples of how bird embryos essentially summarize their own evolutionary history. He shows me a microscope image of a quail embryo’s forelimb, which looks exactly like the arm of a raptor dinosaur, down to its tiny dinosaurian hand. “That’s Deinonychus! Look at that!” Bhullar exclaims as he points to his laptop. Only closer to hatching is this ancestral form overwritten to become the familiar avian wing.

Long after I leave Yale I can’t get that little quail claw out of my head. After years of reporting on extinct dinosaurs, I’ve grown dangerously accustomed to thinking about them in the past tense. But they are still with us, as ghosts within the eggs of their avian descendants.

The links between past and present get clearer in London, as our time on Dinosaur Island draws to a close. While the dinosaurs’ world ended in a flash, the Crystal Palace dinosaurs face a slower, more creeping threat. The sculptures have been added to the U.K.’s registry of at-risk heritage, but a lack of upkeep has let cracks splinter through many of their discolored exteriors. In May part of the face broke off the island’s Megalosaurus, damage caused either by deterioration or possibly by vandals. Conservation efforts are being planned, led by Friends of Crystal Palace Dinosaurs.

With the need for renewal all around us, I ask Maidment how today’s scientists would build their version of Crystal Palace Park. Maidment offers an elegant answer: She’d fill it with birds. “Dinosaurs are the most diverse terrestrial vertebrates alive today, you know,” she says, as a flock of gulls glides overhead and splashes down into the waters beyond. “They never stopped.”