By Louis Jacobson
Special to The Washington Post
Monday, April 26, 1999; Page A09

ITHACA, N.Y.—Earth's crust consists of large, slow-moving plates, like an orange whose rind is still attached to the juicier core but is split into several chunks.

Some plates move apart at a slow but generally constant speed. At these ridges, molten rock flows into the void between the separating plates. When it cools, the new rock forms a seal between plates, at least until the plates have moved far enough apart for new rock to form.

These borders typically form an unusual shape: They curve gracefully along the ocean floor, following the border between two plates. But every so often, they will shift to the right or the left, precisely perpendicular to the direction of the border between the plates. The forces that create such "transform faults" have long puzzled researchers.

Eberhard Bodenschatz, a Cornell University physicist, thinks he is gaining important new insights into this process using heated-wax models to approximate the sea floor ridges where Earth's plates spread apart. Bodenschatz has found striking correlations between the wax's behavior and actual patterns of sea floor spreading.

"We just don't understand these processes yet," says University of South Florida geophysicist Sarah Tebbens. "If we could understand them in wax, that could go a long way to understanding them in real life."

Some geophysicists, however, reject the notion that Bodenschatz's wax models can encapsulate physical processes taking place on scales many orders of magnitude greater. Geophysicists are more comfortable finding answers by carefully monitoring and measuring actual underwater environments.

"There had been an earlier attempt at wax modeling, and this one looks like an improvement on that," says Christopher Scholz, a geophysicist at Columbia University's Lamont-Doherty Earth Observatory. "But even though it looks cool, we know it's not really the same thing as real life, so I'm not sure how much importance to give it. I'm not sure whether it really drives us in a new direction."

In 1972, physicists Douglas Oldenburg and James Brune successfully modeled this process with wax. In 1994, Bodenschatz tried to replicate their experiment, hoping to tweak it and find out more. He filled a three-foot-long tank with melted wax at about 77 degrees Celsius and directed a stream of colder, 25-degree Celsius air toward the center of it. Once the cold air solidified a top layer of wax, Bodenschatz cut a straight line through it with a knife. Then, two parallel arms in the wax tank began to spread apart from each other at a constant, but very slow, rate.

What Bodenschatz discovered was very different from what Oldenburg and Brune had found. Instead of getting miniature transform faults, Bodenschatz's experiments produced zig-zag patterns that were completely unfamiliar to him. Nothing Bodenschatz could do--not even having his undergraduate researchers set up the experiment outside in the frigid upstate New York winter--changed the pattern.

Bodenschatz's puzzlement continued until 1997, when Tebbens informed him that she had seen similar zig-zags in semi-cooled molten magma in volcanic "lava lakes." "It was both disappointing and exciting," Bodenschatz says. "I had discovered something new that no one had seen, but I also wanted to see the transform faults."

On a hunch, Bodenschatz decided to call Shell Oil, the company that had manufactured the wax used by both himself and Oldenburg and Brune. A Shell official told him it was possible that the wax had changed in composition since Oldenburg and Brune's day, because the wells from which the wax's petrochemical components had been drawn had aged, possibly changing the relative proportions of hydrocarbons in the final mix, or perhaps the regularity of their structure.

So the company sent Bodenschatz a new supply of wax from a different source. When Bodenschatz tried it, he found that it did indeed form the transform faults he'd sought. Better yet, the formations varied in shape exactly as they do on the sea floor.

When pulled apart at the slowest speed--10 micrometers per second--the wax formed a valley. (In fact, if the experiment went on long enough, the process was so strong that a wax-free void would appear at the bottom of the valley and grow wider with passing time.) If he tripled the speed, it formed an essentially flat plane. If he tripled the speed again, it formed a slight vertical bump.

These ratios neatly approximated figures measured from below the oceans. The slowest real-life sea floor separations--such as those in the mid-Atlantic ridge, which spreads at only 21 millimeters a year--form a valley. At ridges that move apart three times faster, scientists have observed flat planes. And at speeds seven times faster than the slowest, slight upward peaks have been found, such as those at the east Pacific rise.

Moreover, Bodenschatz found numerous cases in which "microplates" formed at the boundaries between ridges. These microplates are often found in real life, too, such as one off Easter Island. Both the wax and real-life models spin in place because of the motions of the two much larger plates they are sandwiched between.

These findings are forcing some rethinking of old ideas. Some scientists had argued that the perpendicular angles that exemplify transform faults stemmed from their need to adjust to the curvature of Earth. But the fact that similar structures emerged in a small, flat box suggests that may not be enough to explain it.

To be sure, Bodenschatz acknowledges some drawbacks that limit the applicability of his experiments. Unlike the real world, his experiments don't have a layer of ocean water lying on top of them. In addition, Scholz points out that real-life ridges do not have as much molten rock underlying them as Bodenschatz's experiment does. In real life, the undergirding tends to be solid or partially melted rock, Scholz says.

And while Bodenschatz's wax does have a relative density that approximates the rock in question, it is a fundamental property of the universe that gravity does not scale. As a result, critics say that Bodenschatz's models are so much smaller than real transform faults that it is an inaccurate leap to assume that the same properties hold at the bottom of the ocean.

Indeed, even the experiments so far completed do not explain how transform faults emerge; instead, they point to a model that might be used in the future to answer that still-unsolved question. Bodenschatz would like to quantify such factors as wax thickness and temperature gradients, then compare them to real-life data. If he is successful, the results could help create computer models that finally describe the ocean-floor processes.

"If I have a model that can do a lot of the things the Earth does, even if it's not to scale, then some truth will be hiding in it," Bodenschatz says. "If I can make the model work, I can learn something from the Earth, like a biologist using a mouse to approximate a human."

Wax Tectonics

A Cornell University physicist is using heated-wax models to approximate sea-floor ridges where the Earth's tectonic plates spread apart in the hope of gaining important new insights into how this mysterious process works.

The heated-wax models have produced striking correlations between the wax's behavior and actual patterns of sea-floor spreading produced when rifts on the ocean bottom shift.

SOURCES: Cornell University; Heezen & Tharp