Villarrica volcano in ChileReuters

Geologists tell a pretty broad-brush narrative of Earth’s 4.5 billion-year history. For its first half-billion years, the newly formed planet was a seething ball of lava constantly pelted by giant space rocks, including a Mars-sized object that sheared off a chunk that became the moon. Things calmed down when the Late Heavy Bombardment tapered off some 3.8 billion years ago, but volcanoes ensured Earth’s atmosphere remained a toxic stew of gases with almost no oxygen to speak of. It stayed that way for another billion years, when single-celled bacteria filled the oceans. Around 2.5 billion years ago, at the end of the Archean era, algae figured out how to make energy from sunlight, and the Great Oxygenation Event gave Earth its lungs. Complex life took its time, finally exploding in the Cambrian era some 500 million years ago. Evolution moved a lot faster after that, resulting in dinosaurs, then mammals, then us.

It’s a great story, and scientists have been telling it for decades, but tiny fossilized space pebbles from Australia may upend it entirely, giving us a new narrative about Earth’s adolescence. These pebbles rained down on our planet’s surface 2.7 billion years ago. As they passed through the upper atmosphere, they melted and rusted, making new crystal shapes and minerals that only form where there is plenty of oxygen. A new paper describing the space pebbles will be published today in the journal Nature. It suggests the atmosphere’s upper reaches were surprisingly rich in oxygen during the Archean, when Earth’s surface had practically none.

“If they’re right, a lot of people have had misconceptions, or have been wrong,” says Kevin Zahnle, a planetary scientist at NASA’s Ames Research Center. Moreover, if the research holds up, geologists will have a new mystery on their hands: How did all that oxygen get there, and why didn’t it reach the ground?

The paper’s authors—Andy Tomkins and Lara Bowlt from Monash University in Victoria, Australia—set out to find the world’s oldest fossil micrometeorites, originally hoping to compare how many of them fell during the Archean with how many fall to Earth today. The duo spent a week driving through the Pilbara Craton in northwestern Australia, which harbors some of the oldest rocks on Earth—up to 3.8 billion years old. They hiked up hilly outcrops and excavated blocks of limestone and dolomite, camping in the bush and cooking by fire at night.

They selected solid blocks of grayish rock with fine, straight lines embedded within them, which they knew were signs of slow sediment buildup. The rocks contained micrometeorites, chips from asteroids that formed when the Earth did. The chips fell on the shores of an ancient alkaline lake and would eventually be buried under the sediment, where they lay for 2.7 billion years.

Back at the lab, Tomkins and his team carved away the rocks’ outer layers and dissolved the limestone with acid so they could get at the micrometeorites. All 60 of the microscopic pebbles they found were cosmic spherules, meaning they were material that completely melted during their trip through the upper atmosphere, around 46 to 56 miles up. They started out as sand-sized grains of iron and nickel alloy and melted down to a few hairs’ width in diameter. As they melted, they formed iron oxides—rust, basically—including exotic forms that don’t occur on Earth’s surface. One in particular, called wustite, only forms at high temperatures when oxygen is added to metal.

“The survival of wustite is almost a smoking gun” that they really are meteorites, says Zahnle, who wasn’t involved in the work.

When Tomkins saw the minerals, he realized he’d not only found the oldest micrometeorites ever seen, but that they could serve as a sample of the early upper atmosphere. “Several other groups have found them before us, but always in comparatively young rocks—nobody has thought to use them to look at ancient atmospheric chemistry before,” he says.

He set about figuring out how the atmosphere’s upper reaches could fill with oxygen without photosynthetic life. Sunlight makes the most sense. Its ultraviolet radiation splits larger molecules, like carbon dioxide, water or sulfur dioxide from volcanoes, into simpler gases—including oxygen. Zahnle favors water as the source, but no one knows for sure, he says. Whatever the source, it doesn’t explain how the upper atmosphere could hold abundant oxygen while the lower atmosphere suffocated. Rocks from Australia and South Africa—the most ancient geologic source texts on the planet—offer pretty solid evidence that there was no oxygen at the ground at that time.

Tomkins acknowledges that atmospheric modelers will have to test these ideas, and that many questions remain. Meanwhile, other geologists are also trying to tease out what the ancient atmosphere could have looked like. One of them, Roger Buick at the University of Washington in Seattle, published a study this week arguing the Archean atmosphere was less than half as dense as today’s.

Buick, who like Tomkins is Australian, says he knows of no other spot on Earth with better-preserved ancient rocks than the Pilbara. They are time capsules of early Earth, early life, and now the early atmosphere.

“Old rocks are usually buggered up. Like old people, they’ve suffered the vicissitudes of life; they’re not fresh and pristine and beautiful. They’re usually a mess. But these ones aren’t, and are very well preserved,” he says. “It’s just that Australia is a really boring place, and nothing has happened there for billions of years, geologically speaking.”

Next, Tomkins wants to extract micrometeorites from a wider range of ages, from about 3.45 to about 2.22 billion years ago, to investigate how the upper atmosphere changed over that time span. It would include the Great Oxygenation Event, when cyanobacteria started pumping voluminous oxygen into the oceans and air. It would also be great if the Curiosity rover team found micrometeorites on Mars, too, so scientists could scrutinize that planet’s ancient atmosphere, Tomkins says.

As it happens, his work has implications for the hunt for life around other planets. The presence of oxygen in an exoplanet’s atmosphere has long been considered a strong hint that life is present, because of the way plants fill Earth’s atmosphere with oxygen. These new findings suggest it won’t be that simple.

Zahnle says scientists will debate the finer points of Tomkins’ work, but the presence of such old micrometeorites is astonishing—both in their survival, and the tale they have lived to tell.

“If you’re interested in Earth’s history, the arrival of oxygen is the big event, after the origin of life,” he says.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.