Willerslev, a well-known specialist in obtaining DNA from old bones and objects, saw potential biological samples throughout this hodgepodge of artifacts. Glancing at a small Egyptian cooking pot, he asked the tour leader, "Do you ever find any grain in these?" After studying a dinosaur skeleton that proved to be a cast, not actual bone, he said: "Too bad. There can be proteins on the teeth."

"I am always thinking, 'Is there something interesting to take DNA from?'" he said, glancing at the curators. "But they don't like it, because ..." Willerslev, who until recently traveled with a small power saw, made a back-and-forth slicing motion with his hand.

Willerslev was visiting Montreal to receive a science prize from the World Cultural Council -- one previously given to the string theorist Edward Witten and the astrophysicist Margaret Burbidge, for her work on quasars. Willerslev won it for "numerous breakthroughs in evolutionary genetics." These include recovering the first more or less complete genome of an ancient man, in 2010, and setting a record for the oldest genetic material ever retrieved: 2.4-million-year-old genes from a frozen mound in Greenland, which revealed that the Arctic desert was once a forest, complete with poplar, birch, and roaming mastodons.

These findings are only part of a wave of discoveries from what's being called an "ancient-DNA revolution," in which the same high-speed equipment used to study the DNA of living things is being turned on specimens from the past. At the Globe Institute, part of the University of Copenhagen, where Willerslev works, there's a freezer full of human molars and ear bones cut from skeletons previously unearthed by archaeologists. Another holds sediment cores drilled from lake bottoms, in which his group is finding traces of entire ecosystems that no longer exist.

"We're literally walking on DNA, both from the present and from the past."

The old genes have already revealed remarkable stories of human migrations around the globe. But researchers are hoping ancient DNA will be more than a telescope on the past -- they hope it will have concrete practical use in the present. Some have already started mining the DNA of our ancestors for clues to the origin of modern diseases, like diabetes and autoimmune conditions. Others aspire to use the old genetic data to modify organisms that exist today.

At Willerslev's center, for example, a grant of 500 million kroner ($69 million) from the foundation that owns the Danish drug company Novo Nordisk is underwriting a project whose aims include incorporating DNA variation from plants that lived in ancient climates into the genomes of food crops like barley, wheat, and rice. The plan is to redesign crops and even entire ecosystems to resist rising temperatures or unpredictable weather, and it is already underway -- last year, barley shoots bearing genetic information from plants that lived in Greenland 2 million years ago, when temperatures there were far higher than today, started springing up in experimental greenhouses.

Willerslev, who started out looking for genetic material in ice cores, is leaning into this possibility as the next frontier of ancient-DNA research, a way to turn it from historical curiosity to potential planet-saver. If nothing is done to help food crops adapt to climate change, "people will starve," he says. "But if we go back into the past in different climate regimes around the world, then we should be able to find genetic adaptations that are useful. It's nature's own response to a climate event. And can we get that? Yes, I believe we can."

In 1993, just a day before the release of the blockbuster Steven Spielberg film Jurassic Park, scientists claimed in a paper that they had extracted DNA from a 120-million-year-old weevil preserved in amber. The discovery seemed to bring the film's premise of a cloned T. rex closer to reality. "Sooner or later," a scientist said at the time, "we're going to find amber containing some biting insect that filled its stomach with blood from a dinosaur."

But those results turned out to be false -- likely the result of contamination by modern DNA. The problem is that modern DNA is much more abundant than what's left in an old tooth or sample of dirt. That's because the genetic molecule is constantly chomped on by microbes and broken up by water and radiation. Over time, the fragments get smaller and smaller, until most are so short that no one can tell whether they belonged to a person or a saber-toothed cat.