But one evening last August, Horz sat down to read National Geographic's 'Return to Mars' issue -- the volume that showcased 3-D pictures from Pathfinder's landing site -- and he saw a field strewn with rocks that looked like they had been busted up by impacts from small but fast-moving objects. The rocks looked to Horz like they had been broken by small meteorites, a possibility that teams studying Mars surface geology had not suggested.

"I looked at it and it was very obvious to me. I saw the crater (on the rock named Stimpy) right away, and I saw these numbers of rocks that clearly broke in place on the surface," Horz said. Some small pieces looked like they came off of larger pieces not far away, one or two of them with the tops knocked off, he said. "It's exactly what you see in the lab, and that's so difficult to make with other processes."

Horz learned to recognize the breaks created by high-speed collisions during years in the lab shooting stony projectiles at rock samples. He did that work for NASA during the 1980s. The research attempted to simulate the affects of small-meteorite impacts on lunar rocks -- tests that were needed to help geologists interpret the geology of the moon's surface.

Geologists have long thought small-scale cratering would not affect the martian surface because, just like on Earth, the atmosphere would protect the planet. Everything but the largest meteoroids would be burnt up by friction in the atmosphere or slowed down so much from their cosmic velocity that they would not create craters, many believed. This is what happens with all but the largest meteorites falling to Earth.

In contrast, the surface of Earth's atmosphereless moon is thoroughly churned by meteorite impacts. It is being bombarded continually by meteorites from the microscopic to the mammoth smashing into its surface.

Until the August evening looking at Pathfinder photos, Horz had long sided with influential studies that had concluded that small-scale cratering would not be a factor in martian geology. One paper published in 1970 that particularly infuenced Horz had determined that craters less than 165 feet (50 meters) in diameter would not be created on Mars, he said.

Driven by the images of the Pathfinder site, though, Horz decided to reinvestigate those conclusions. He enlisted the help of a group at Lockheed Martin's advanced flight systems section to test just what size particles could survive entry into Mars' relatively-thin carbon dioxide atmosphere. (The pressure on Mars' surface is less than 2 percent that of Earth's atmosphere at sea level.)

The Lockheed group uses detailed computer calculations to model how errant satellites and spacecraft debris break up when reentering Earth's atmosphere. Engineers use such calculations when designing spacecraft and in determining the hazards that failed space projects pose to humans.

Horz asked the Lockheed group to change some of their calculation codes to model the affect Mars' atmosphere would have on entering meteorites. It ran calculations for meteorites of many different sizes and compositions entering the martian atmosphere over a range of angles at different speeds.

The group's leader, aerospace engineer William Rochelle, summed up the results. "Small fragments, and lots of them, can actually make it through the martian atmosphere," he said.

For example, calculations determined that stony meteoroids larger than about three inches in diameter could survive entry (although they would lose some mass to vaporization in the atmosphere) and smash into the surface at speeds of about 1,400 mph (2 kilometers per second).

This result gave Horz confidence that small meteorite collisions were indeed responsible for much of the jagged breaks seen at the Pathfinder landing site. The results of Rochelle's calculations and Horz's argument were published in the Sept. 24 issue of the journal Science.

Despite Horz's and Rochelle's work, geologists on the Mars Pathfinder science teams are still skeptical that small meteorite impacts caused the site's broken rocks.

Dan Britt, a geologist at the University of Tennessee at Knoxville, who was a member of the imaging science team for the Mars Pathfinder mission, wonders why meteorite impacts are not noticed in the martian soil. The red martian ground, which appears to be a sort of desert hardpan, is smooth in Pathfinder pictures.

"If there's so many jagged rocks, there should be some craters gouged out in the hardpan also, and we don't see that," Britt said, adding that the small impact solution is unnecessarily complicated.

"The easy explanation is, the stuff is ejected from a crater. It's not very hard to make jagged edges if you blow things up with big impacts," Britt said.

Just over a mile (2.2 kilometers) away from the Pathfinder site is a large crater that could be the origin of some of the busted rocks, he said.

While Horz's explanation is interesting, numerous other explanations exist, said Tim Parker, a geologist at NASA's Jet Propulsion Laboratory in Pasadena, Calif. Parker worked on the team that analyzed Mars geology through Pathfinder data, and he has exhaustively studied almost every Pathfinder image.

Speaking of Horz's small-meteorite interpretation, Parker said, "It's certainly reasonable, but I don't think it's a foregone conclusion because of the fact that you're in a catastrophic flood channel. You can get that kind of scar just from banging boulders together. It happens on Earth."

"Go to the mountains and go find yourself an active stream," Britt agreed. "Look at the rocks around there. Even when they're rounded, they have jagged edges because they whack against other boulders."

Most likely, all these processes are occurring together, Parker said.

"That's the frustrating aspect of doing surface science remotely. The geologic history of a landscape is almost always much more complex than a simple description from one discipline can fathom," he said.