Pinpointing when hunter-gatherers became farmers calls for clever science and detective work. Geoff Watts met sleuths piecing together the evidence.

The Neolithic revolution may not be as familiar as the French, the industrial, or the Russian revolutions, but it was no less influential. More a process than an event, it began well over 10,000 years ago and marked our ancestors' transition from a hunter-gather to an agrarian society.

The cornerstone of this revolution was the successful domestication of plants and animals, a transformation that presents a tough challenge to archaeologists. You can tell if our forebears used clay pots or stone axes because you can find them in their dwellings. With rather less certainty, you can tell if they ate meat or harvested grain because scatterings of animal bones and plant seeds attest to this. But how can you tell if the animals had been hunted or herded? How can you distinguish a crop that had been sown in a field from one that had been harvested in the wild?

You think like a detective, and seek the help of biology and physics, that's how. Two researchers who do just this are Melinda Zeder and Bruce Smith of the Smithsonian Institution's Archaeobiology Program in Washington DC, United States.

ÁñÁ«ÊÓƵ

The main role of physics has been in providing an accurate dating technique. Small-sample accelerator mass spectrometer radio carbon dating is now the gold standard - and offers an accuracy that lives up to the grandeur of its title. Having been dated, the material has to be interpreted. This is where biology comes in and, true to form, everything becomes a little less clear cut.

Seeds are the plant components most likely to survive the decay of time. They also provide the best clues. "Early farmers wanted certain crop improvements such as larger seeds, or more easily processed seeds with thinner seed coats," says Smith. "They would deliberately select them to use as seed stock in next year's planting."

ÁñÁ«ÊÓƵ

So here are the first clues to domestication: seed size and coat thickness. What else? "Those plants that tend to hold their seeds more firmly and don't disperse them are more likely to be there for their human harvesters." Likewise seeds that are clustered in one or a few easily harvested groups at the tip of the stalk. None of these characteristics has a selective advantage to wild plants - which is why they're good markers of human intervention.

Moving to animals, some of the evidence traditionally used to infer domestication has been far from watertight. The significance of finding the remains of an animal outside its normal geographical boundaries, for example, depends on how confident you can be about those boundaries.

But other markers offer more reliable evidence of man's hand. One is an overall reduction in size compared with the wild originals. "A premium on docility in managed animals may have led to human selection for small, more tractable beasts, especially males," says Zeder.

Zeder herself has been examining goats' horns. With domestication, large scimitar-shaped horns were replaced by the smaller, flattened structures familiar in the modern farmyard. "This is probably due to humans controlling breeding. Males no longer need to compete. They no longer need to attract females using this wonderful paraphernalia on top of their heads." In the absence of selective pressure to maintain the horns, generation by generation they shrink.

ÁñÁ«ÊÓƵ

The most subtle clues come from demographic profiling of the bones of animals found in human encampments. "Hunters and herders have a different approach to culling," Zeder explains. "Hunters want a short-term payoff. So they'll go for large adult males, or they'll drive a whole herd off the edge of a cliff, or they'll prey on lactating females and their suckling young." Each of these strategies will create a distinctive age and sex profile in the remains found at archaeological sites.

"Herders have a long-term investment strategy. They only need a few males for breeding. So most males will be killed young. They'll keep the females around for longer to breed and for milking purposes." In short, studying the age and gender of the bones in a kitchen midden can reveal much more than merely the type of meat being eaten.

The emerging picture of mankind's early efforts at domesticating animals and plants is far from complete. But archaeologists can now point to more than half a dozen areas where it seems to have happened independently. These range from Mexico and the southern central Andes through the Near East to northern China and the Yangtze River valley.