A photo of the Burgess Shale

The Burgess Shale, a storied rock formation that offered some of the earliest glimpses of half-a-billion-year-old creatures, crops out high above Tokumm Creek.


Read our COVID-19 research and news.

Some of Earth’s first animals—including a mysterious, alien-looking creature—are spilling out of Canadian rocks

KOOTENAY NATIONAL PARK IN CANADA—The drumming of the jackhammer deepens. Then, a block of shale butterflies open, exposing to crisp mountain air a surface that hasn't seen sunlight in half a billion years. "Woo!" says paleontologist Cédric Aria of the Nanjing Institute of Geology and Palaeontology in China, bracing the top slab of rock upright.

Its underside bears charcoal-colored smudges that look vaguely like horseshoe crabs or the Millennium Falcon from Star Wars. "It's a spaceship landing area here," says expedition leader Jean-Bernard Caron, curator of invertebrate paleontology at the Royal Ontario Museum (ROM) in Toronto, Canada.

Those "spaceships" are carapaces, molted onto a long-vanished ocean floor by a species new to science. This field season they've been spilling out of the rocks here, where Caron's team has spent the past few years unearthing groundbreaking animal fossils from the Cambrian period, the coming-out party for animal life on Earth. During the Cambrian, which began about 540 million years ago, nearly all modern animal groups—as diverse as mollusks and chordates—leapt into the fossil record. Those early marine animals exhibited a dazzling array of body plans, as though evolution needed to indulge a creative streak before buckling down. For more than a century, scientists have struggled to make heads or tails—sometimes literally—of those specimens, figure out how they relate to life today, and understand what fueled the evolutionary explosion.

Gingerly, Aria and Caron place the top piece of their slab aside. Space is hard to come by in the quarry, perched on a ledge the size of a small bedroom at an altitude of 2500 meters, far above Tokumm Creek. For years, an equally forbidding site about 40 kilometers northwest of this valley offered the clearest window on the Cambrian. There, in 1909, U.S. paleontologist Charles Doolittle Walcott discovered the Burgess Shale, a fossil formation that preserves not only hard shells, but also soft features such as the legs, eyes, and guts of Cambrian creepy-crawlies.

But in recent years, Caron has shown that the richest fossil-bearing rock extends many kilometers beyond Walcott's site. This summer's excavation marks his latest visit to this long Cambrian tapestry. Each new stop has offered striking views of unfamiliar animals, many already described in high-profile papers: the little fish relative Metaspriggina, a vertebrate ancestor that Caron now speculates clustered in schools; the pincered Tokummia; and the ice cream cone–shaped fossils called hyoliths, which Caron's Ph.D. student Joseph Moysiuk last year linked to shelled animals called brachiopods, some of which persist today.

Other sites around the world are also opening new vistas of the Cambrian. Scientists can now explore the animal explosion with a highlight reel of specimens, along with results from new imaging technologies and genetic and developmental studies of living organisms. "There have been a host of new discoveries," says paleontologist Doug Erwin of the Smithsonian Institution's National Museum of Natural History in Washington, D.C. Researchers may be closer than ever to fitting these strange creatures into their proper places in the tree of life—and understanding the "explosion" that birthed them.

Jean-Bernard Caron shows off the "mothership," an enigmatic Cambrian life form his team found in the Canadian Rockies this summer.


Each new find brings the simple joy of unearthing and imagining a seemingly alien creature. On a break, Caron cautiously shows off this year's crown jewel, found about a week earlier. It's an intact, hand-size carapace with a center spine, like a Prussian spiked helmet frozen in ancient rock. Another undescribed species, it seems to be related to the spaceships. Caron's team calls it the mothership.

He's nervous just holding it. Burgess Shale fossils are so valuable that Parks Canada keeps the exact locations of Caron's sites secret, monitors them with cameras, and prosecutes fossil poachers. ROM once insured a Burgess Shale specimen for half a million Canadian dollars when it went on loan, he says—and that was an animal known through multiple fossils. This is one of a kind.

"It's going to be iconic," Caron says. "It's the most extraordinary fossil I've ever found."

For years, Caron suspected Walcott's site might be rivaled elsewhere in the Rocky Mountains. The breakthrough came in 2012, near an area called Marble Canyon, where a 2003 wildfire had burned off the trees. While crossing an avalanche chute filled with broken tiles of rock, his reconnaissance party found itself surrounded by impressions of soft-bodied creatures, many with unfamiliar shapes. "It was clear that nobody had ever been walking over this pile of rocks before with this purpose in mind," says Bob Gaines, a geochemist from Pomona College in Claremont, California, who has joined Caron's expeditions since the beginning.

They returned to excavate in 2014. At least one in five of the animals they found at Marble Canyon belongs to species new to science, the team concluded. Now, they've moved on to other sites along the valley.

How Cambrian species are related to today's animals has been debated since the fossils first came to light. Walcott classified his oddities within known groups, noting that some Burgess Shale fossils, such as the brachiopods, persisted after the Cambrian or even into the present. So, for example, he concluded almost all the creatures resembling today's arthropods were crustaceans.

But later paleontologists had other ideas. Harvard University's Stephen Jay Gould perhaps best captured the charisma of Cambrian life in his 1989 book Wonderful Life: The Burgess Shale and the Nature of History, in which he lavished attention on the "weird wonders" excavated from Walcott's city block–size quarry. Gould argued that oddballs such as the aptly named Hallucigenia, a worm with legs and hard spines, seem unrelated to later animals. He slotted the unusual forms into their own phyla and argued that they were evolution's forgotten experiments, later cast aside by contingencies of fate.

Contemporary paleontologists have settled on yet another way to understand them. Consider the arthropods, arguably Earth's most successful animals. In a family tree, the spray of recent branches that end in living arthropods—spiders, insects, crustaceans—constitutes a "crown" group. But some animals in the Burgess Shale probably come from earlier "stems" that branched off before the crown arthropods. These branches of the tree don't have surviving descendants, like a childless great-uncle grinning out from a family photo. In that view, many of Gould's weird wonders are stem group organisms, related to the ancestors of current creatures although not ancestors themselves. Newer fossils from the Canadian Rockies help support that view. Caron argued in 2015, for example, that his specimens of Hallucigenia have features suggesting the animal belongs on a stem group of the velvet worms, creatures that still crawl around in tropical forests spitting slime.

All in the family

A partial schematic of a proposed family tree of arthropods shows the complex relations among living and extinct groups. Some extinct Cambrian creatures (red) may belong to "stem" groups that branched off the arthropod tree before the common ancestor of living groups like arachnids and insects.

Hallucigenia Velvet worms Trilobites and relatives Radiodontans Centipedes, millipedes Horseshoe crabs Arachnids Crustaceans Insects 0 541millionyears ago

Similar analysis awaits the spaceships. At first glance, Caron's team thinks they are a new species or group of radiodontans, stem arthropods that also include Anomalocaris, the Cambrian's charismatic apex predator—a clawed, fearsome-jawed swimmer half a meter long. Filling out the branches of that stem group gives a "step-by-step view of how an arthropod built its body" through evolutionary time, says paleontologist Allison Daley at the University of Lausanne in Switzerland.

Throughout much of Cambrian paleontology, that's the game—a high-stakes, sometimes contentious race to find diagnostic body parts on known or new fossils, make arguments about what taxonomic groups they belong to, and maybe revise evolutionary history in the process.

In the past few years, paleontologists have approached the problem with an array of new techniques. Those include scanning electron microscopes, which can discern a specimen's chemical makeup as well as image it, and computerized tomography (CT) scans, which can penetrate fossils without scraping away material. Those tools have also illuminated a startling series of internal features: fossilized Cambrian brains. Beginning in 2011, paleontologist Xiaoya Ma, now at the University of Exeter in the United Kingdom, published a string of papers tracing nervous tissue in exceptionally preserved Chinese fossils. Those nervous system architectures offer a parallel way to sort animals into evolutionary groups, beyond the usual anatomical structures, and other teams have presented their own compelling specimens.

In fossils of the shrimplike Chengjiangocaris kunmingensis from southwest China, for example, "we have this structure that looks almost like a pearl necklace," running almost head to tail, says Javier Ortega-Hernández, an incoming professor at Harvard. His team, led by Jie Yang at Yunnan University in Kunming, China, argued in 2016 that the necklace is a nerve cord studded with smaller clusters of neurons, themselves sprouting tiny nerve fibers. Living arthropods no longer have those fibers. But today's velvet worms and priapulid worms do, implying kinship between long-vanished stem arthropods and those groups.

Critics argue that paleontologists such as Ma and Ortega-Hernández overinterpret some fossils, spotting nervous tissues that aren't there. Many of those structures, the critics say, might just be "halos," biofilms formed when microbes broke down internal parts like muscles or guts after death. But other researchers are convinced. "If you look at the best-preserved nervous systems, there's no doubt" that the features are real, says Graham Budd, a paleontologist at Uppsala University in Sweden and an architect of the current stem-and-crown concept.

Bold claims that use anatomy to revise family trees engender similar controversy throughout the field. One argument that Hallucigenia fits with the velvet worms, for example, depends on the exact shape of its claws. But other teams counter that the claws aren't diagnostic of ancestry.

The uncertainties leave paleontologists ever hungry for newer, better specimens. "When there is a debate, you bring a new fossil and say, ‘Look, this is the feature we see,’" Caron says, warming up in a tent perched high above Tokumm Creek. "Without fossils, it's speculation."