This latest report comes from scientists who measured the Earth's magnetic field near the moon using instruments mounted on Lunar Prospector. Their results strengthen the theory that the moon has a metal core that is much smaller than cores of the inner planets of the solar system.

The new data agrees with Apollo mission seismic and sample-return evidence that suggests the moon is partly made of the same stuff as the Earth's upper crust, or mantle. And their findings agree with results released earlier this year by NASA Jet Propulsion Lab scientists who used Lunar Prospector to make a gravity map of the moon and who also conclude that the moon has a small, partially molten core.

Lon L. Hood of the University of Arizona in Tucson and collaborators detail the results of their magnetic measurements in the Aug. 1 issue of the American Geophysical Union's Geophysical Research Letters. Hood and his colleagues on the Lunar Prospector magnetometer team estimate that the moon's metal core is roughly 420 miles (680 km) across -- plus or minus 112 miles (180 km). This makes it only one to three percent of the moon's total mass. By contrast, the Earth's core constitutes one-third of its mass.

"We knew that the moon's core was small, but we didn't know it was this small," Hood said. "This really does add weight to the idea that the moon's origin is unique, unlike any other terrestrial body -- Earth, Venus, Mars or Mercury," Hood said. "The simplest hypothesis, and the most popular now, is that a Mars-sized object collided with Earth after Earth had differentiated into a core and mantle. The impact generated a vapor cloud, which was mostly composed of silicate, and that became the moon."

Alternate theories surmise that the Earth captured a fully formed moon, or that the moon formed in orbit around the Earth but that Earth somehow prevented metal from accreting to the moon. Those theories now seem less plausible, Hood said.

Collaborating with Hood on this research are David Mitchell and Robert P. Lin, of the University of California at Berkeley, Mario Acuna, of the Goddard Space Flight Center in Greenbelt, Md., and Alan B. Binder, of the Lunar Research Institute in Tucson. Binder is the principal investigator for the Lunar Prospector mission.

They measured changes in Earth's magnetic field as Lunar Prospector orbited the moon 21 times in April 1998. The measurements were made during a longer-than-usual unbroken passage of the moon within Earth's magnetic tail.

Electrical currents that flow through the moon's hot metallic core create magnetic fields that oppose Earth's magnetic field. The spacecraft magnetometer detected the slightly altered magnetic field, which the team analyzed to estimate limits on the size of the moon's core.

The Lunar Prospector magnetometer team has yet to analyze data that were taken this spring, Hood said. The sun is approaching a period of maximum solar activity, so it's possible there will be more "noise" that could mask some of the magnetic effects in the newer data.

While evidence for a very small lunar core is mounting, proof is not yet absolute, Hood added. If Japanese scientists succeed in deploying new seismometers on the moon, they may be able to provide absolute proof, he said.

Lunar Prospector, an 18-month, $63 million NASA mission ended July 31 with a controlled crash landing at the moon's south pole. The spacecraft was designed to study the moon's resources, gravitational and magnetic fields, composition and geologic history.