Conservationists may be miscalculating the numbers of the threatened animals such as elephants, say African and American researchers. The error occurs because of a flaw in the way they estimate animal numbers from the piles of dung the creatures leave behind. The mistake could lead researchers to think that there are twice as many elephants as there really are in some regions, according to Andrew Plumptre of the Wildlife conservation Society (WCS) in New York. Biologist Katy Payne of Cornell University in Ithaca, New York, agrees. 'We really need to know elephant numbers and the evidence that we have is quite indirect,' says Payne, who electronically tracks elephants. Counting elephants from aeroplanes is impossible in the vast, equatorial rainforests of Central Africa. So researchers often estimate elephant numbers by counting dung piles in a given area. They also need to know the rate at which dung decays. Because it's extremely difficult to determine these rates, however, researchers tallying elephants in one region tend to rely on standard decay rate established elsewhere. But researchers at the WCS have found that this decay rate varies from region to region depending on the climate and environment. Using the wrong values can lead to the census astray, says Plumptre. He and his colleague Anthony Chifu Nchanji studied decaying elephant dung in the forests of Banyang-Mbo Wildlife Sanctuary in southwest Cameroon. They found that the dung decayed between 55 and 65 percent more slowly than dung in the rainforests of neighboring Gabon. If researchers use decay rates from Gabon to count elephants in Cameroon, they would probably find more elephants than are actually around. This could mean estimates in Cameroon at least twice as high as those derived from decay rates calculated locally, says Plumptre. 'However accurate your dung density estimate, the decay rate can severely affect the result.' Plumptre also says that the dung-pile census should be carried out over a region similar in size to an elephant's natural range. The usual technique of monitoring only small, protected areas distorts numbers because elephants move in and out of these regions, he says. 'If the elephant population increases within the protected area, you can not determine whether it is a real increase or whether it is due to elephants moving in because they are being poached outside.' Plumptre says that similar problems may also plague other animal census studies that rely on indirect evidence such as nests, tracks or burrows. The word 'threatened' in the first sentence of the first paragraph could be best replaced by______.