Tell a New Yorker that rats can save lives and he might consider it an impossible trick for the beady-eyed sewer dwellers.
But for the past 10 years, a group of Belgian researchers based in Tanzania have been training a species of giant African rats to sniff out land mines and unexploded ordnance.
"When people see they can use these animals for humanitarian purposes, it changes their perception," said Bart Weetjens, whose nonprofit group, Apopo, has pioneered the use of the African or Gambian giant pouched rat in mine detection. "People find it most fascinating."
By teaching local residents how to handle the rats — a food source for some Africans — the group hopes to develop a cheap, reliable, indigenous resource for de-mining, an expensive and dangerous process that typically operates in unstable, war-ravaged regions.
"The whole philosophy of the project is to provide local people with tools and techniques so they can manage to deal with the problem themselves," Mr. Weetjens said.
Mr. Weetjens and Apopo hope the creatures, the largest rats by size in the world, will join hand-held metal detectors, armored vehicles and dogs in the effort to detect and remove land mines, which kill or maim as many as 20,000 people a year, according to the International Campaign to Ban Land Mines.
The campaign, a grouping of about 1,400 nongovernmental organizations from 90 countries, issued a survey last year that listed more than two dozen African countries dealing with buried anti-personnel mines and mine contamination. The problem was especially acute in Angola, Mozambique, Sudan, Ethiopia and Eritrea.
Angola — along with Afghanistan, Cambodia and Colombia — is thought to have one of the world's most severe problems, with an estimated 7 million land mines hidden in unmarked fields four years after the end of a long civil war, the government said.
With its headquarters at the Sokoine University of Agriculture in Morogoro, Tanzania, Apopo has 50 rats working in Mozambique and plans to expand to Angola within two years if it can secure enough funding. Thirteen other African countries have expressed an interest in the program, Mr. Weetjens said.
"They see it as an African technology," he said.
In the 3½ years that Apopo has been using fully trained rats in Mozambique, the rodents have cleared about 100 acres of land and found 23 mines. "A few animals ... died from old age or food poisoning," Mr. Weetjens said, but none has been destroyed in an accident.
In the laboratory, young rats — which have long been adopted as pets in Africa and the United States — are first domesticated and taught to associate food with a clicking sound. Then, using the food and clicking as a reward, they are presented with three holes and trained to identify the one releasing the chemical vapors similar to those found in explosives.
In the field, the rat is leashed and made to run in straight lines across a 1,000-square-foot area during 30-minute shifts. Every time it stops to investigate the ground, the handler marks the spot on a paper grid and feeds the rat on the assumption that it has found a mine.
Metal detectors are used to double-check the rats' work before they extract or explode the mines, Mr. Weetjens said.
The process takes advantage of the pouched rat's powerful nose. In the wild, the nocturnal rats use their sense of smell to hunt for nuts, fruits and vegetables in the dark, said Ron Verhagen, a biologist who has studied rodents for 30 years and helped Mr. Weetjens start Apopo.
The bomb-sniffers reproduce rapidly, generating a steady stream of new candidates for the program. But because they are nocturnal, the rats can work for only a few hours in the morning, before the heat becomes too intense.
The de-mining community has given the program mixed reviews, questioning whether animals work as reliably and consistently as man-made devices.
"Everybody is trying to look for a magic bullet," said Andrew Lyons, vice president of the Halo Trust USA, part of an international organization that removes war debris. "Dogs and rats might be cheaper on a square-meter basis, but the constraint is they actually have to work. And in a humanitarian context, they have to work 100 percent" of the time.
The main advantage of using animals is that they focus on the scent generated by a buried explosive, said Perry Baltimore, director of the Marshall Legacy Institute, which specializes in detection dogs. They are not distracted by the countless pieces of metal that usually litter old battlefields, a problem for mechanical metal detectors.
Apopo handlers run many rats over each row in the minefield, in case the first one misses a mine. Although they are not perfect, Mr. Weetjens said, every rat working in the field must first pass a test requiring it to find every mine in a designated area without a single mistake.
At least one organization, Handicap International, is "watching [the rats] with considerable interest," said Wendy Batson, the organization's U.S. director.
During a recent trip to a Handicap International program in Mozambique, Ms. Batson made a short side trip to see the rats in action.
"I thought it was really intriguing," she said. "There's no question Apopo makes a good case for rats."
Handicap International uses dog teams in a limited capacity, such as in Kosovo where they will check an area after people have cleared it with metal detectors. Ms. Batson said she thinks rats could play a similar role but only after Apopo produces enough "rigorous test data."
Rats have several advantages over dogs in Africa, Mr. Weetjens said. Because they are indigenous, rats are much less susceptible to the tropical diseases that can kill foreign-born dogs. While the weight of a dog has on rare occasions detonated a mine, it is virtually impossible for a 5-pound rat to do so.
Mr. Weetjens estimates that the training and deployment of each rat costs $2,000 to $3,000, which includes the handler's pay. The cost of acquiring, training and delivering a certified detection dog to a site costs about $20,000, Mr. Baltimore said.
"There's no easy, quick fix to any of this," Ms. Batson said. "You're always looking for a balance between safety, speed and cost efficiency. That's why, of course, there's always a lot of interest in new techniques that might make it easier and safer."