“In the grand scheme of things, plate tectonics is a young theory,” says Brian Savage, a seismologist at the University of Rhode Island. “The plate-tectonic theory is 50 or 60 years old. That’s not old. I always tell my students to compare it to evolution—that’s 150 years old, about as old as electricity and magnetism.”

In the half century since it found general acceptance among geologists, plate tectonics—the theory that continents drift and oceans open up across the surface of Earth over hundreds of millions of years—has become the common wisdom. Americans know why earthquakes happen and why Africa and South America seem to fit together. And geologists have learned much that has not yet made its way to the public consciousness: that oceanic plates are more dense than continental plates, for instance, such that they always subduct under continents.

But scholars of plate tectonics still can’t answer some pressing questions about the Earth system. They’re not certain of how the continents formed. They’re not sure if continents can be deformed or destroyed. And—until recently—they couldn’t confidently answer the question: How deep is a continental plate?

A new study, released last week in Science, may put that final debate to rest. Using seismological data taken from every continent in the world, the paper finds that continental plates begin between 80 and 120 miles below the surface.

In other words, you’re closer to the edge of space right now than you are to the bottom of a continental plate.

The finding will help scientists improve our understanding of the mechanics of the plates—not just their depth or composition, but how they move around Earth. The Earth is sliced up into many different kinds of layers. Elementary schoolers are familiar with the chemical bands: the crust, which terminates three to 40 miles below the surface; the silicate mantle, which descends 1,800 miles below the surface; and the inner and outer core, a white-hot sphere of iron and nickel that meet 3,959 miles from the surface.

But these familiar terms only describe what makes up the Earth’s layers, its chemistry. They don’t describe how the planet moves or how it reacts to heat. For that, geologists turn more specific words that describe Earth’s mechanical layers—lithosphere and asthenosphere.

“Imagine that you had a hot fudge brownie you had poked. It’s all liquid at first,” says Cin-Ty Lee, a geologist at Rice University who was not affiliated with the study. “Then it will cool from the top and it will generate that crusty hard layer. That top layer, the cold part, that’s the lithosphere. It’s the uppermost parts of the earth.”

The lithosphere, in other words, is the layer of the planet that makes up the tectonic plates. The warmer parts below constitute the asthenosphere. To a human observer, the asthenosphere would also appear as a kind of rock, but over millions of years it functions like a slow and ponderous liquid: upwelling, sliding around, mixing with itself.

In fact, both of these spheres would appear chemically similar, especially at their boundary. But the heat difference between them is stark. The interior of Earth is about 1,400 degrees Celsius (or 2,552 degrees Fahrenheit), while the surface is, on average, a balmy 25 degrees Celsius (77 degrees Fahrenheit).

“Where it transitions from 1,400 degrees to 25 degrees, that’s the lithosphere,” says Lee.

Geologists long ago identified the chemical layers of Earth—they can be predicted from surface research or directly observed via seismic stations or gravity-sensing satellites—but they have struggled to identify the boundary of the lithosphere. What’s more, scientists often arrived at conflicting results: While some estimated the lithosphere’s depth from diamonds found near volcanoes, other tried to sense it by measuring how seismic waves move through the inner Earth. The diamond-informed estimates said the lithosphere ended much closer to the surface, about 100 miles down, than the seismic estimates did.

The Science paper reconciles the two findings. It draws on more data, at high resolution, from every continent to estimate the edge of the lithosphere. “We used seismic waves generated by earthquakes with magnitude 5.5 or greater and recorded at stations all over the world from 1990 to 2015,” said Saikiran Tharimena, a research fellow at the University of Southampton. They also looked specifically at the middle of plates—the center of the Canadian shield, for instance—and not at their rocky and more fragmented coasts.

With this much larger store of data, modeled much more sensitively, researchers were able to see a large cutoff at the diamond-informed line between 80 and 130 miles below the surface.

That finding fleshes out our understanding of continental plates significantly. Conventional accounts of the continental-plate system describe vast islands of rock drifting on a hidden subterranean sea of magma. The truth is both more and less dramatic: Tharimena and his colleagues say that, below the lithospheric line, rock in the mantle is slightly melted by increased pressure. Rock at the top of the asthenosphere is perhaps 1 percent melted, overall—but that tiny amount of melt is enough to lubricate continental drift.

“Our results suggests that this lower boundary of the asthenosphere is quite sharp, i.e., there is an abrupt change in seismic velocity,” Tharimena told me in an an email. It is too sharp to be caused by heat or chemical change alone, they say.

“When you have a velocity that changes that much, it’s very hard to do, and one of the only ways to do it is to add a liquid phase,” said Lee. “This is the first paper to suggest a low-velocity zone below continents.”

He also praised the paper’s methodology and its “very interesting” finding.

“What [the paper] doesn’t answer is whether the presence of this liquid melt is causal or whether it’s just a byproduct” of history, he told me. In other words, the boundary may appear at roughly the 100-mile line today, but that may be a fluke of chemistry and not a permanent feature of Earth.

“Today, it appears that thickness of the lithosphere does intersect there, but it’s not obvious that has always been the case throughout Earth’s history,” he said.

Tharimena said the next step in the research is to look at the lithosphere more broadly using their high-resolution technique, especially at the places where oceans meet the continents. This will help give a global view of where the lithosphere ends and the more churning asthenosphere begins. And that will help answer a larger question: How did the continents form in the first place?

“It’s amazing we know what we know, but we don’t know a lot,” Savage told me, taking a survey of plate tectonics. Then he laughed. “Though probably the evolutionary biologists would say that, too.”