Space / Solar System

The origin of the moon is a mystery as old as the tides. No other planet boasts such a singular and striking companion: Mercury and Venus have no moons at all, Mars has two tiny spud-shaped satellites, and the many moonlets of the gas giants are tiny in comparison with their mother planets. Pluto has, in Charon, a mate that rivals its relative size, but two or three Charons—or Pluto itself, for that matter—could fit inside the moon. Ours is bigger.

Although the moon has enthralled observers for aeons, planetary scientists have struggled to trace its ancestry. The manned lunar landings of the Apollo program were supposed to help by collecting hundreds of pounds of moon rocks. Three perfectly respectable theories for lunar origin had been proposed. All that was needed to choose among them was a chunk or two of the object in question. But the lunar samples may as well have been made of blue cheese. 

"Nobody knew what was going on with these rocks," says Jay Melosh, a planetary scientist at the University of Arizona in Tucson, who witnessed the crisis firsthand. "Some bits of data supported one hypothesis, other bits of data supported another. The evidence was ambiguous. The result was utter confusion." 

In the midst of this disappointment, a few stalwart scientists came up with yet another concept. Called the giant-impact theory, it proposed that the moon was a kind of divot nicked off when the ancient Earth collided with another celestial object fully half its size. The theory may have been far-fetched, but it was consistent with the Apollo findings. Its advocates said it might also explain certain dynamics of Earth-moon interactions.

Proving it seemed impossible. No trace of the impact would be found on Earth, because much of the crust folds under and recycles every 150 million years or so. And on closer scrutiny, a few troublesome facts seemed to belie the impact scenario. By the 1980s, the number of scientists investigating the origin of the moon had dwindled. Most thought the problem couldn't be solved. Others just lost interest.

Today the tides are turning again. In the past five years, a new generation of technology and talent has emerged to resuscitate the giant-impact theory. Faster computers have allowed theorists to construct more credible models of impact dynamics, and 21st-century geophysics has dispensed with some of the formerly troublesome "facts." Now scientists are invoking giant impacts to explain a constellation of planetary artifacts. Mercury's relatively large iron core, for example, suggests that an impact stripped away most of the planet's outer mantle. Venus may also be a fragment of that collision. A violent blow could have knocked Uranus into its odd axial tilt, which is about 90 degrees from upright. And Charon might have been born of Pluto's bombardment.

"Our thinking of the early solar system as a plodding and predictable place [has given] way to the notion of planet-size objects careening into one another in wild, stochastic ways," writes Robin Canup, a fledgling planetary scientist at the Southwest Research Institute in Boulder, Colorado.

With this shift in thinking, even the old guard has rediscovered its enthusiasm. "It's fair to say the giant-impact theory has more truth in it than any other theory," says Melosh.

Today the Apollo lunar samples are housed in a special NASA facility in Houston and loaned to investigators under strict conditions. At the University of Arizona, planetary scientist Tim Swindle keeps a few milligrams of moon dirt in a safe in a locked basement room that's checked nightly by security guards. Some of the soil traveled to Earth via Apollo 11; another sample was scooped up by robots in the Luna program, which the Soviet Union launched at about the same time.

"These samples were of course very valuable when they were first brought back," says Swindle, pulling a vial of dirt from the safe. Yet superficially, he says, there's not much difference between lunar rocks and "rocks out of the parking lot."

He sifts some gray brown lunar dust onto a microscope slide. Under magnification, there are hints of an unearthly provenance: Some of the dirt has been "weathered" by meteor impacts, with smaller grains and glassy surfaces produced by crushing and melting. In spectrometric tests, some of the dirt also shows signs of exposure to the solar wind—charged particles that stream from the sun and bombard the moon, which, unlike Earth, has neither an atmosphere nor a magnetic field to shield it.

But the mineral composition of moon rocks—including elements such as iron, silicon, magnesium, and manganese—resembles that of Earth rocks, says Swindle. And the distribution of oxygen isotopes—a feature that, like a local accent, tells scientists what part of the solar system a rock hails from—is also identical to that of terrestrial geology. "The rocks are fundamentally made of the same stuff as the rocks on Earth," Swindle says.

The similarities between moon and Earth rocks were compatible with each of the three now-defunct models of lunar origin. In the fission scenario, for example, a furiously spinning early Earth tosses some of its mass into orbit. The co-accretion model has the moon forming at the same time and place as Earth did, by condensing from an enormous disk of gas and dust surrounding the early sun. In the capture theory, Earth's gravitational field grabs a wandering moon from nearby space. If the moon calved off Earth or formed anywhere in the same neighborhood, it should look a lot like Earth.

But the Apollo samples differ in important ways from Earth rocks. For one thing, moon rocks are drier than anyone expected. It may seem obvious that rocks are dry. But on Earth, water molecules seep into spaces within the crystal lattices of minerals. Scientists recently estimated that hydrated minerals in Earth's lower mantle could store about five times as much water as the oceans hold. When researchers analyzed moon samples, they couldn't wring out a drop. "Nobody has found even a single molecule of water in moon rocks," says Melosh. The moon rocks are also depleted of volatile elements such as sodium and potassium that, like water, vaporize readily.

The Apollo missions brought more unwelcome tidings. Scientists have long known, for example, that the density of the moon is little more than half that of Earth. That difference had led some to propose that the moon lacks an iron core like the one in Earth's innards. Seismic readings by Apollo astronauts confirmed that the lunar core is small to nonexistent. If the moon was born at the same time and region of the solar system as Earth, why doesn't it, too, have the same kind of core?

Apollo revealed that the moon and Earth are like surprise bonbons: They look the same on the outside, but they have different fillings. None of the three leading models of lunar origin could reconcile this contradiction.