The mantle, it turns out, tells a story of surprising sameness. The Chang'e-6 basalts carry chemical signatures of "depleted" mantle material — rock stripped of certain elements during earlier melting — much like the low-titanium basalts Apollo brought home. That means the mantle cumulates, the minerals that settled out as the Moon's magma ocean crystallized, are broadly similar under both hemispheres.

Mg-suite rocks backed that up. These form when accumulated mantle material partially melts and interacts with the crust, and their magnesium number (Mg#) — a standard measure of magnesium richness — acts as a fingerprint of the mantle they came from. The Chang'e-6 and Apollo Mg-suite samples show similar Mg# values.

In short: no compositional divide deep down.

What about the crust?

The crust is another matter entirely.

Chang'e-6 anorthosites show higher peak Mg# values than their Apollo counterparts, meaning the farside crust contains more magnesium-rich material. That matches what orbiting instruments had already suspected from a distance.

The Mg-suite rocks added a second signal. Farside samples contain extremely low concentrations of rare earth elements and lower thorium-to-samarium ratios (Th/Sm) than Apollo rocks. Either the farside magmas formed largely without KREEP-rich material, or the KREEP layer itself was laid down unevenly across the Moon.

The takeaway so far: one mantle, two very different crusts.

How could Earth's gravity build two faces?

The team's model ties these observations together with a sequence of events.

• The Moon's early magma ocean cooled and crystallized fairly evenly, producing the first mantle cumulates and an anorthositic crust, before the Moon became tidally locked to Earth.
• Once the Moon settled into an orbit keeping one hemisphere aimed at us, tidal forces went to work. Earth and the Moon were much closer together then, so the near side got noticeably more tidal heating than the far side.
• That temperature gap set off large-scale horizontal circulation in the molten interior. Hot magma flowed from the near side toward the far side, dragging magnesium-rich anorthosite material along with it.
• The dragged material piled up on the far side, thickening the crust and raising its magnesium content.
• The remaining melt flowed preferentially back toward the near side, concentrating KREEP and ilmenite-bearing cumulates (IBC) — dense, late-stage minerals — on the Earth-facing hemisphere. The far side ended up with only thin layers of them.

The model explains the central paradox of the new data: a chemically uniform mantle beneath two crusts that could hardly be more different.

The takeaway: the Moon's split personality may simply be a scar from standing too close to a young, hot Earth.

How solid is the conclusion?

Caution is warranted. The analysis rests on one mission's samples from one farside landing site, and the circulation scenario is a reconstruction of events more than four billion years old. Earlier competing ideas — a giant impact among them — remain on the table, and the study's claims about deep mantle uniformity rest on indirect chemical fingerprints rather than direct sampling.

Still, this is the first time scientists could put farside and nearside rocks side by side in the lab. By connecting Chang'e-6 samples with the Moon's early thermal history, the work gives researchers a concrete, testable picture of how Earth's closest neighbor got its two faces.

The bottom line: gravity, heat, and flowing rock — not luck — likely drew the Moon's asymmetry.