As if the complaints from tribal-rights groups weren't enough, the herculean effort to unlock the rainbow of human diversity also has taken some lumps from closer quarters.

Some anthropologists assert that the diversity project's reliance on building family trees with DNA markers is a crude way to measure human history, much less capture so nuanced and fluid a notion as race.

A better metaphor for tracking our biological identity, they argue, is to trace a braided river that splits and rejoins and splits again, often erasing its genetic channel over the meandering course of eons. In other words, they say it probably can't be done.

"I'm sorry, but I think this project is half-baked," said Jonathan Marks, a biological anthropologist at Yale and one of the most vocal scientific critics of the sampling effort. "The leaders are first-rate geneticists who happen to have a folk knowledge of anthropology. They're still asking questions like, 'How close are the Irish related to the French?' These are things most people stopped asking in the 19th Century. What's French anyway? That's a cultural precept, not a biological one.

"Frankly, I think they're naive," Marks added tartly. "It reminds me of a bunch of teenagers trying to build a cyclotron in their back yard."

Undaunted, project supporters reply that any confusion about "Frenchness" or "Irishness" is exactly what a global DNA survey will clear up once and for all by neatly unbraiding the knotty rope of culture and biology.

The trump card in the debate is that the gene pools that hold many of the answers are evaporating.

"You can nit-pick about methodology for years, but in the end, it comes down to having this one opportunity to take a genetic snapshot of the world," Weiss said. "Do we do it or don't we?

"We're scientists, not politicians," he said, invoking a mantra repeated by all bringers of knowledge, from the first who harnessed fire to the splitters of the atom. "But at least we can lay the biological information out on the table. People can see it out there."

Watching a graduate student isolate the DNA from the Yanomamo samples and drip it into small capsules, Weiss observed wistfully that the primordial chain of molecules that makes us all so fundamentally alike and yet profoundly unique is transparent and without color--that in its purest form, the stuff of life is always absolutely clear.