Mighty forces beyond the solar system billions of years ago might have shaped much of the land beneath our feet today. A study recently published in the journal Geology proposes that Earth’s continental crust went through major growth spurts when the early solar system surfed through the four major spiral arms of the Milky Way. The galactic passages triggered a rain of comets on Earth, and their giant impacts built colossal amounts of new crust.

“It blows my mind that the continental crust we’re standing on—where the majority of Earth’s biomass is—has a potentially fundamental connection to the cosmos,” says Chris Kirkland, a geologist at Curtin University in Australia and lead author of the paper.

Kirkland and his colleagues matched the dates of crust growth to the timing of galactic movements—both of which occurred every 170 million to 200 million years—and tied that timing to comet impacts via trace crystals and tiny glass beads in the ground that preserved details of the collisions.  But the theory has proved controversial. It has garnered support from other geologists but has also been criticized for ignoring simpler terrestrial explanations. Astronomers, too, have picked holes in the theory’s celestial arguments.

Although the study covers the scale of the Milky Way, the work began by analyzing miniscule crystals only 100 microns wide. Those crystals have settled in places where Earth’s earliest continental history is preserved: ancient, stable blocks of crust in Greenland and in Australia. But they once resided within the vast, seething ocean of magma deep underground before they journeyed upward—trapped within a melted blob that spewed onto the surface and hardened into new crust. Despite the harrowing journey, the crystals kept a precise record of their birth. Confined within them are elements, such as uranium, that decay in a clockwork fashion, allowing scientists to use them to track age. Other elements in the crystals mark the composition of the parental magma.

Using these records, Kirkland’s team deduced that during a period stretching roughly 3.8 billion to 2.8 billion years ago, magma punctured the surface and built new crust every 170 million to 200 million years. That pattern matches the rate at which Earth passes through the galaxy’s spiral arms. But why would this galactic scale be imprinted on Earth?

Kirkland and his colleagues argue that as the solar system moves into a spiral arm, the enhanced stellar density in the arm dislodges comets from the Oort cloud—the vast reservoir of icy bodies that could extend as far as about 1.6 light-years beyond the sun—sending some on an inward trajectory toward Earth The bombarding comets excavated huge volumes of Earth’s surface, creating tremendous heat but releasing pressure below. “The rocky bit on top acts a bit like a champagne cork or a valve on a pressure cooker,” says Tim Johnson, a geologist at Curtin University and a co-author of the paper. When the pressure dropped, the change lowered the mantle’s melting point, causing the rock to form massive quantities of magma. The buoyant magma then rose to the surface, where it solidified into basalt and looked much like the dark rocks in Hawaii or Iceland, though on colossal scales. The basalt was so thick (likely tens of kilometers) that the base was blazing hot, forcing it to melt once again and form granite—a pale-colored rock that is so buoyant that it would float on top of everything else.

Evidence for the theory can be found not just within the crystals but also in areas called spherule beds. When a comet strikes Earth, the ultrahot sediments that were blasted into the atmosphere fuse into molten rain, then solidify again on the ground as a layer of tiny glass beads known as spherules. The team notes that two spherule beds, which occurred around 3.45 billion and 3.25 billion years ago, match times of enhanced crust production (derived from the crystals) and when the solar system moved into a spiral arm.

“It’s a brilliant insight,” says Brendan Murphy, a geologist at St. Francis Xavier University in Nova Scotia, who studies global tectonics and was not involved in the study. “It’s certainly got me thinking about the early Earth in a way I hadn’t thought about it before.”

Other scientists, however, are skeptical. “To me, there’s not enough evidence to go to such an extraordinary explanation,” says Ross Mitchell, a geologist at the Chinese Academy of Sciences, who was not involved in the study. He worries that only two spherule beds match the crystal data, and one of those beds is not located within Australia—meaning that the timing might not match after all. Mitchell says that makes the paper shaky as well. “It’s like trying to build a house on the Mississippi Delta,” he says. “Maybe it lasts for a month when people read your paper, but a year or two from now, that foundation might crumble.”

Mitchell has another objection: scientists should not entertain extraordinary ideas until they have “ruled out the simple boring ideas,” he says. And for the cycle that Kirkland’s team attribute to galactic forces, he has a boring explanation in mind: Regular movements of the plates that make up Earth’s crust. The 170-million- to 200-million-year cycle is exactly the amount of time it takes for a slab of crust to travel from the surface down to the core-mantle boundary in the well-known process of subduction. Then, because other material moves up as slabs sink, it is likely that plumes of melted mantle rise and create new crust during this same time period. “If we already have ready-made explanations from Earth’s own cycles, do we really need to go to the outer reaches of the galaxy?” Mitchell asks.

But Kirkland has a comeback. He argues that subduction had not yet begun at this point in Earth’s early history: the planet was simply too hot and the mantle was moving too vigorously for crust to sink all the way down to the core-mantle boundary. And Murphy agrees, noting that the origin of subduction is a contentious topic within geology and that many researchers argue that it got a later start. Bombardments, on the other hand, were certainly happening early on. “When we look at our celestial neighbors, we know that the early history of the solar system was a shooting gallery,” he says.

The claim that the shooting gallery consisted of comets, however, is another controversial aspect of the theory. Comets, ejected from the outer reaches of the solar system, arrive with a lot of energy that would lead to a great deal of melt and new crust. But scientists do not see much evidence for cometary impacts in the inner solar system. Instead asteroids—which carry less energy—are the more typical incoming objects. “So if this did happen, it’s a process that left almost no geologic trace,” says Margaret Landis, a planetary scientist at the University of Colorado Boulder, who was not involved in the study.

Bill Bottke, a planetary scientist at the Southwest Research Institute, agrees that comet showers are quite rare. And when they do happen, the comets are very unlikely to hit Earth. Although it is true that comet showers are triggered by passing stars that perturb the inner Oort cloud, such a star has to get relatively close to our sun for the showers to occur, and these events happen extremely infrequently—even during a spiral arm passage. “In science, it’s worth being ambitious but not ambitious to the point where you push the data beyond what’s reliable,” Landis says.

Johnson says he has received a number of dismissive e-mails from geologists and is not surprised by the criticism. He has also gotten messages of support. “These sorts of things are always going to upset some people who prefer the status quo,” he says. “But that’s great—that’s exactly how science moves forward.” Murphy agrees that the paper will encourage others to look at the data in a completely different light and to test it further. “It certainly has my head spinning,” he says. “I’m sure that’s true of anyone who’s reading this paper. And irrespective of whether it proves to be right or wrong, that’s a major contribution.”