Pandas will be one step further from extinction if conservationists can harness new findings about their anal odours to bring wild bears together. Matt Kaplan sniffs out a good idea

CUTE, cuddly and a nightmare to breed in captivity: the giant panda is an icon for the challenges that face conservationists. Ten years ago the bear's future looked grim. Wild populations were dwindling, captive animals hardly ever mated, and those that did frequently failed to raise their young. But in 1996, when all hope seemed lost, two pandas were sent from China to America. The tide was about to turn.

The bears did so well in their new home at San Diego Zoo that when representatives from the Wolong Panda Reserve in Sichuan Province, China, paid a visit, they barely recognised their animals. "I had been going out to Wolong for years and telling them that they needed to build bigger and better enclosures," says Donald Lindburg, a specialist in animal behaviour at the zoo. "They would nod their heads and do nothing." Finally here was incontrovertible evidence that pandas are remarkably sensitive to their environment.

That understanding, together with lessons about panda management learned at Wolong and San Diego, has led to huge progress. "Both sides were suddenly able to learn from each other and to apply what they were learning in their own ways," says Chinese wildlife conservation consultant Mabel Lam. The numbers testify to their success: a decade ago there were 25 pandas at Wolong, today there are 70.

But overcoming the challenge of captive breeding is only half of the story. Next the conservationists face the task of managing China's wild pandas, and it's a huge challenge. Humans have spread so far into traditional panda territories that groups of bears that once mixed are now isolated. "To maintain genetic diversity these populations have to be able to interbreed," says Lindburg. And on the face of it, human intervention looks doomed to failure here. The bears are extremely difficult to track down, living solitary lives on top of treacherously steep mountains in some of the densest and most remote forests on the planet. Even if you could find them and bring them together, the result would probably be a bloody battle between established bears and newcomers. But perhaps there's another way.

It is two decades since George Schaller, from the World Wildlife Fund, discovered that pandas mark trees in their territory using scents from their anal glands and urine. He realised that the bears were creating signposts for other pandas to follow, but what exactly were these odours saying? For years the enigma of panda scent communication lay unexplored. Then, with the arrival of the bears in San Diego in 1996, the case was reopened. The smelly messages have proved to be far more complex than anyone expected. And they may also be just the tool conservationists need to help wild pandas.

Working with zoo colleague Ron Swaisgood, Lindburg found that females advertise their sexual readiness by marking objects near ground-level with urine, while aggressive, sexually ready males show off their fitness by doing handstands to place anal secretions as high up as possible. By selectively exposing pandas to each other's odours, the researchers began to understand how the smells produced by one animal influence the behaviour of others.

Pandas of different ages respond to scents differently. Immature males actively avoid the scents of sexually active ones. Mature males track down females entering their reproductive cycle, and the scents these females produce make the males less aggressive and more libidinous. Likewise, male odours seem to boost female sexual motivation.

Next, Swaisgood and Lindburg enlisted the help of Lee Hagey, one of the zoo's analytical chemists. Using mass spectrometry, he found that male scents contained between 20 and 50 different compounds. Samples from females contained a whopping 150 to 200 compounds, which reflects the fact that they add urine and vaginal secretions to their scent deposits. "The nature of the chemicals was really complex," says Hagey. "We saw a series of short-chain fatty acids, typical of the products of bacterial metabolism."

He found that much of the smell is generated by bacteria that live in the scent gland. But the mixture also contains several aromatic anti-bacterial agents, including phenol -- probably secreted by the panda to control the bacteria living in the gland -- as well as other smelly compounds that seem to derive from plants in the bear's diet.

Even more intriguingly, each animal seems to produce a distinctive odour that acts as a sort of smelly name-tag that others can recognise. "To smell like a panda, you must have the complete basic set of chemicals present," says Hagey. "What distinguishes one from another is the relative proportion of the chemicals."

Being able to read panda scent opens up new possibilities for conservationists. One tactic that is already being used is to expose captive females to the smells of prospective mates to make them more receptive to subsequent sexual advances, thus boosting captive breeding success even further. But anal scents have even greater potential in the wild. "I could track every wild panda so long as I was given a steady supply of scent samples to plug into my computer," says Hagey.

Lindburg acknowledges that finding the scent trails will be difficult, but even so, it looks like the only viable option for the foreseeable future. That's because radio tracking, the approach most often used by conservationists, has been illegal in China since 1995, following an unsubstantiated claim that a panda was killed when its collar got hooked on a branch. Lu Xiaoping from the Convention on International Trade in Endangered Species believes that the law will only change if conservationists can prove that radio collars are safe and justified. And that's a tall order.

Tracking and identifying individual pandas in the wild would be just the beginning. Researchers at Wolong and San Diego Zoo are already considering plans to use scent marks to tackle the problem of genetic isolation. "If we use the scent of a female, we could potentially lead males from one area to another," says Swaisgood. "There is even the possibility that we could prepare the population he is moving into by placing his scent on trees in the area long before he arrives." A similar approach could also be used to introduce captive bears into the wild, with scent marks giving wild and captive animals a chance to "meet" before the actual release.

"We have a real challenge ahead of us," says Swaisgood. "There are around 20 isolated populations with fewer than 50 bears in each. With numbers like this we know management of genetic diversity is going to be really tricky." What's more, there is no guarantee that scent lures will work. "All of the scent analysis work that we have been doing is cutting-edge stuff," comments Hagey. "It hasn't been tried before." Perhaps the biggest problems are on the ground. At the moment there is nobody studying giant pandas in the wild, there are no research stations from which to conduct such research, and panda reserves are not much more than lines on maps.

Things are changing, though. "In recent years China has met reality face to face and there have been sweeping changes in their environmental policy," says Lindburg. A 10-year ban on logging began in 1997, and the government is also spending $30 billion on a reforestation project to expand protected areas. They are also developing plans for conservation corridors or "panda paths" between panda reserves. And the State Forestry Administration has started building protection stations to prevent deforestation and keep out poachers. There are even rumours from Beijing that field research stations in panda territory may soon get the green light.

Lindburg applauds China's efforts. "The Chinese receive a lot of criticism from America over conservation issues," he says, "but when I look at what they are doing for the panda I can only wish our government was doing so well." He adds that when it comes to panda conservation there are none of the usual political tensions between China and the US. "In fact, conservation relationships between us have done nothing but improve since the pandas arrived in 1996." There is hope for the panda yet.