With mountain ranges topping 3,000 metres (10,000 feet) and deep craters, the farside of the Moon bears scant resemblance to the smoother surface and shallow lava-filled maria, or "seas", on the nearside visible from Earth.
Scientists have proffered many explanations for this split personality, also known as the "lunar dichotomy".
Some point to uneven tidal heating, the process by which energy from rotation and orbit deform a planet's outer crust.
Others argue that lopsided bombardment by asteroids and comets explain our Moon's Janus-faced exterior.
But a pair of researchers from the University of California at Santa Cruz, Martin Jutzi and Erik Asphaug, have proposed a new storyline for the sculpting of the lunar landscape, one that reaches back to the Moon's very origins.
Their findings, published in Nature, tie up loose ends that other theories cannot account for, they said.
Not long after Earth took shape more than four billion years ago it was likely struck a glancing blow by a Mars-sized body, an event called the giant-impact hypothesis.
The Moon is thought to be composed of the debris cast off by that collision, which probably created other, smaller, moon-like bodies as well.
As our solar system evolved toward its current configuration, none of these lesser orbs were likely to have survived very long -- unless they landed in a sweet spot called an Earth-Moon Trojan point.
At least one such mini-moon, about a third the diameter of the one we see today, could have been suspended between the gravitational pulls of the Moon and Earth for tens of millions of years, they calculated.
Eventually, however, it would have lost its moorings and crashed into the Moon, which was covered at that point by a magma ocean topped by a crystallised crust.
At high speed, planet-scale collisions create monstrous craters and vast amounts of vaporised debris, mostly melted by the intense heat.
But because the mini-moon, due to its position, would have been moving at a much slower speed -- about two-to-three kilometres per second -- the impact would have left a rim of mountains.
"According to our simulations, a large 'moon-to-Moon' size ratio and a subsonic impact velocity lead to an accretionary pile rather than a crater," Jutzi and Asphaug concluded.
This scenario would also help explain why the farside's crust is so much thicker, and why certain minerals are concentrated there, the study said.
"The current study demonstrates plausibility rather than proof," MIT researcher Maria Zuber cautioned in a commentary, also in Nature, noting that the origins of the farside highlands have been "a topic of speculation since the first global measurements of the Moon's shape."
Because the rubble that became the mountains would have crystallised millions of years earlier than the Moon, examining the age of the soil would be one way to confirm the study, but no such samples are available.
The new theory could also be bolstered -- or challenged -- by data expected next year from the NASA's Lunar Reconnaissance Orbiter mission, as well as high-resolution gravity mapping to be done by the Agency's Gravity Recovery and Interior Laboratory (GRAIL).