SKEPTICS IN BOTH the sports-medicine and genetics fields argue that such tests collect information on an irrelevant percentage of the nearly 25,000 genes identified and that the relationship between genetic tendencies and an individual's diet still remains poorly understood.

"It's a field in its infancy," says Arthur Leon, a professor in the School of Kinesiology at the University of Minnesota. "It's premature to design a diet around such a small sample of genes. There's very little clinical-trial evidence to support that kind of claim."

For different reasons, former Ironman Triathlon champion Dave Scott isn't convinced the science is valid either. Currently a professional coach based in Boulder, he doesn't advise his clients to get tested yet. "I'm not sure genes tell the whole story. It takes years of training to bring about an athlete's potential," says Scott. "Plus food will never make you go faster. It just allows you to sustain an intense workload longer, based on your training and mental profile."

Even the Department of Energy, which manages the results of the Genome Project, is alarmed, noting on its Web site that companies like Sciona aren't subject to regulation by the FDA.

Grimaldi admits that while he and his colleagues may not know everything, "the 19 genes give us an idea, and it's enough to make a useful start." And he's certain that Sciona and its competitors are on the forefront of something huge. "Even at 64K, the computer was a useful tool," he says, "and the growth in computing power was very rapid. It's the same with nutrigenomics. Soon we'll be able to test for hundreds of genes."

From a risk standpoint, USA Track & Field's Troxler sees none. "As long as you work with a nutritionist and are careful not to develop new nutrient deficiencies, then it's safe." The worst that can happen is that you're out a few hundred bucks for the test and you end up with a much improved whole-foods diet. That's not a bad deal on something Troxler calls "the future of sports nutrition."