It's a cool, clear morning, with just a few wisps of cirrus streaking the sky, when Francesca Smith settles into a spot just above Ross's quarry, sitting cross-legged on the crackled ground. Across the road, along the ridgeline, are three petroleum pump jacks, their heads slowly rising and dipping as underground reservoirs empty, then fill with oil.

An associate professor at Northwestern University, the bubbly, brown-haired geochemist arrived only yesterday, having driven herself and two young assistants out from Illinois. Wing hands her a slab of mudstone. "Look," he says, pointing to the merest fragment of a leaf splayed across the surface. "That's a little piece of organic matter — it's probably part of the cuticle." At that moment, the wind picks up, and the paper-thin specimen peels back from the rock, threatening to fly away. "Emergency wrap!" Smith shouts, hastily enfolding her prize in an envelope of foam.

For a moment, Smith contemplates the pump jacks, visual metaphors linking the prehistoric world she and Wing are exploring to our world's present and future. The carbon released during the PETM, she notes, is also thought to have come from organic sources, just like the carbon we pump into the atmosphere every time we turn on a light or drive a car. "The only difference," Smith reflects, "is that we're doing it much, much faster."

During the Paleocene-Eocene Thermal Maximum, scientists estimate that a massive amount of carbon — 4 to 5 trillion metric tons, perhaps — flooded into the atmosphere. That's about 10 times more carbon than humans have pumped out since 1751, and the rough equivalent of how much carbon remains stored in fossil fuels.

From a climatological perspective, it makes sense that the infusion of that much carbon would jack up temperatures. After all, carbon combines with oxygen to form carbon dioxide, which, next to water vapor, is the most abundant of the planet's greenhouse gases. As their name suggests, these gases (which also include methane and nitrous oxide) behave rather like the transparent panes of a greenhouse: They allow the sun's rays to stream in but trap a good deal of the heat the earth beams back in response.

In general, this is a good thing; it helps create what some call the Goldilocks effect — the fact that, to human beings and other creatures, Earth's temperature seems "just right," neither hellishly hot like Venus nor bitterly cold like Mars. That's not to say that our planet's pane of greenhouse gases never varies in thickness. Ancient air bubbles trapped in Antarctica's ice show that levels of carbon dioxide declined during past Ice Ages and rose during warm interglacials such as our own.

It's not clear how much carbon dioxide there was in the atmosphere on the eve of the PETM, but scientists think levels may have reached somewhere between 500 and 750 parts per million. This compares to 380 parts per million at present, 280 parts per million in pre-industrial times, and 180 during past glacial high stands. As a result, the late Paleocene was already quite warm, about as warm as many climatologists project our world could become by the start of the next century.

The large amount of carbon dioxide in the pre-PETM atmosphere almost certainly came from a sustained spate of volcanic eruptions. (During the Paleocene, volcanoes were particularly active.) As a result, the atmosphere's load of carbon dioxide gradually rose. Then, around 55.5 million years ago, carbon dioxide levels shot up very sharply, perhaps to 1,800 or more parts per million, leaving behind a distinctive geochemical signature.

The signature, Smith explains, takes the form of a dramatic shift in the ratio between two stable forms of carbon, heavier carbon-13 and lighter carbon-12. It's this shift that scientists first picked up in the calcareous shells of marine organisms, then found in the teeth of terrestrial mammals. Last year, Smith and Wing showed that, in the Big Horn Basin, the shift is captured by leaf waxes as well. "And there is only one way we know of to shift the ratio as much as it shifted," Smith says, "and that's to add a lot more light carbon."

The richest concentrations of light carbon are found in organic materials, including fossil fuels like coal (which forms from deeply buried plants) and methane gas (primarily a byproduct of microbial decomposition). While scientists are still not sure what triggered the massive release of light carbon at the start of the PETM, they do have a number of possible culprits, including fires that raged though forests, dried-up peat bogs and even underground coal seams, and effusively erupting volcanoes whose magma intruded into organic-rich sediments, cooking out the carbon.

Of all the scenarios so far floated, perhaps the most provocative invokes the dissolution of methane hydrates on the seafloor. These are ice-like solids in which water molecules form crystalline cages that entrap molecules of gas; they form, and remain stable, within specific ranges of temperature and pressure. At the end of the Paleocene, Rice University earth scientist Gerald Dickens has suggested, a jolt of warmth from an unknown source pushed these strange solids to the point that the methane gas inside them started burbling up through the ocean and into the atmosphere.

Methane is much shorter-lived than carbon dioxide, but it's also a more effective greenhouse gas. And as methane breaks down, the carbon it contains recombines with oxygen to form carbon dioxide, which can circulate through the climate system for thousands of years.

After serious study, many experts have concluded that not enough methane was locked up in hydrate form to have single-handedly caused the PETM. That does not constitute an absolution, however. A big release of seafloor methane could still have been part of a sustained chain reaction whereby an initial rise in carbon caused enough warming to trigger the release of additional carbon that caused still more warming, and so on. Might the carbon we are so heedlessly pumping out today spark a similar sequence of events? 

As scientists try to imagine the consequences of our greenhouse gas emissions, they invariably return to this question. The earth abounds with "traps" for carbon — not just seafloor hydrates, but also terrestrial forests, marine plankton and frozen Arctic soils. Some of these our own warming climate may already be springing open. Scientists from the University of Alaska, Fairbanks recently calculated that the permafrost of the far North sequesters 100 billion tons of carbon in its top three feet alone; as the permafrost thaws, that carbon will progressively leak into the atmosphere.

The release of carbon from some of these traps will be offset by the uptake of carbon in others. But at some point, scientists worry, the release of carbon may so far outweigh its absorption that the situation will cascade out of control. By the time we realize we're in serious trouble, in other words, it may be too late to do much about it.