What's more, the guns and the owners that turn up for buybacks represent neither the kinds of weapons nor the types of people generally involved in gun crimes, said several researchers who have studied the programs. And some of those who participate in the buybacks are cashing in on spare weapons but keeping at least one at home--or they plan to use the proceeds to purchase another gun.

Gun buyback programs, in which local governments encourage residents to turn in firearms using modest cash payments or gift certificates as incentives, have become a recurring and highly visible feature of the American dialogue on violence. Just this April, when President Clinton announced a federal grant to assist the gun buyback program inWashington, he surrounded himself with a phalanx of police recruits and invoked the bloody chaos of a shooting three days earlier at the National Zoo.

Referring to the city's mayor and congressional delegate, Clinton declared, "When I called them, after that terrible tragedy at the zoo, and asked them what I could do to help, they said, `Well, why don't you help our gun buyback program?'"

The buyback programs have a potent political appeal at a time when gun violence is at the forefront of public concerns. On the one hand, they address gun-control advocates' desire to take weapons out of circulation. On the other, they generate minimal opposition from gun-rights defenders because nobody is forced to give up a weapon he wants to keep.

Among the largest buyback programs to date was one supervised last September by Cook County Sheriff's Department, which collected 5,347 guns in three weekends. The Chicago Housing Authority plans another gun buyback this September.

Still, independent follow-up studies of gun buybacks in Seattle, Sacramento, St. Louis and Boston found no evidence that the programs reduced gun crime. In Seattle, researchers also checked coroner's records and hospital admissions data for the six months following a buyback in 1992. They found no evidence of an effect on firearms-related deaths or injuries.

"The continuation of buyback programs is a triumph of wishful thinking over all the available evidence," said Garen Wintemute, director of the Violence Prevention Research Program at the University of California at Davis.

The benefits may be too subtle to detect, said Clinton administration officials, who this year plan to devote $15 million to assist local gun buyback programs. While they concede the programs do not often directly disarm criminals or recover the types of guns preferred by criminals, they nonetheless contend that eliminating any gun ultimately reduces the risk of death or injury.

"The first purpose of this is not trying to stop bad guys from robbing banks or bad guys from shooting each other. The first purpose is to get guns out of homes," said Lee Jones, a spokesman for the U.S.Department of Housing and Urban Development, which funds gun buybacks using money from an anti-drug program the department manages.

"We do think this can have a positive influence for reducing gun accidents and gun violence in the home. Or, for that matter, it prevents [the guns] from being stolen and used in crimes," Jones continued. But academic researchers--often divided by passionate differences over gun control--are in rare agreement in their conclusions. At a U.S. National Institute of Justice lecture delivered just weeks before Clinton's grant announcement, University of Pennsylvania professor Lawrence Sherman, who headed a wide-ranging assessment of crime prevention programs, called gun buybacks "the program that is best-known to be ineffective" in reducing firearms violence.

The numbers of weapons collected--typically no more than a few thousand guns, even in the most successful buyback--represent a tiny fraction of the nation's arsenal, with an estimated 220 million guns now in civilian hands and another 4.5 million newly manufactured guns added each year.

"At most, they take 1 [percent] to 2 percent of guns out of a [local] community, and he guns collected are among the least likely to be used for violence," Wintemute said. Guns used in crimes most often are modern, up-to-date, semi-automatic pistols, one weapon of choice being the 9 mm pistol used in the National Zoo shootings. The weapons turned in during buybacks overwhelmingly are older guns, such as revolvers, which in some cases don't even work. A Harvard University study of buyback programs in Boston in 1993 and 1994 found that nearly three-quarters of the guns recovered were made before 1968. In Seattle, one-quarter of the guns collected were inoperable.

Also, the gun owners who turn in their weapons tend to be middle-age or elderly. Street criminals tend to be adolescents and young adults.

In any case, surveys of the people who turn in their weapons frequently find they have additional guns at home they intend to keep: in Sacramento, 59 percent of participants said they did so, as did 62 percent of participants in St. Louis and 66 percent in Seattle.

Sometimes, people also use the money they receive from turning in an old gun- one that would command a low price on resale- to help pay for a higher-quality weapon. In St. Louis, 14 percent of buyback participants said they planned to purchase a new gun within the next year. Another 13 percent said they might.

Said Richard Rosenfeld, a criminology professor at the University of Missouri-St. Louis who conducted the study: "We found that the people most likely to be planning to buy another gun are the respondents at highest risk for gun violence. They tended to be the younger respondents, they tended to be the respondents more likely to have arrest records."

HUD officials point to signs of success following one project in Pittsburgh, an annual gun buyback campaign that began in December 1994. That program also includes a firearms safety education project and free distribution of child trigger locks to gun owners who would rather not give up their weapons.

There has been no formal evaluation of the project. But Dr. Matthew Masiello, a pediatrician who helped organize the Pittsburgh program, collected statistics showing a dramatic drop in the area's firearms deaths, which declined 39 percent from 1993 through 1996.The drop is much greater than the 16 percent decline in gun deaths nationally during the time period.

But even Masiello attributes the apparent success to the fact that the Pittsburgh program was "much more extensive" than simple gun buybacks. He cites as other important factors the trigger-lock distribution, firearms safety education, and a campaign to mobilize church groups and other community organizations to reduce gun violence. Other researchers are skeptical of any correlation with the buybacks at all.

Jacqueline Cohen, a criminologist at Carnegie Mellon University in Pittsburgh, said gang violence in 1993 nearly doubled gun homicides from the previous year and "on any basis, you would expect this unusual peak to be followed by a decline."

Cohen states otheer factors as probably more important in the decline in Pittsburgh's firearms deaths. Among them were an aggressive police campaign to combat gun crimes during that time period and a major federal prosecution that"basically decimated" Pittsburgh's LAW youth gang, which was heavily involved in gun violence. Also, large Northern urban centers similar to Pittsburgh generally experienced especially steep drops in gun crime from 1993 through 1996.

She said Pittsburgh is considering a program that would offer financial rewards for anyone who turned in another person for illegally carrying a gun in a public place. Such a program may be a more effective use of money than a buyback because "it's targeting the guns that are the cause of the problem," she said.

Rosenfeld, who has been hired as a consultant to evaluate the ongoing HUD-funded buybacks, said the concept of buying back guns may yet be proven an effective tool in reducing violence. "There is no evidence that they directly reduce gun violence in the form of gun assaults or gun homicides," he said.

But he theorized that programs more narrowly focused on public housing projects could potentially have a bigger impact, because they might lead to a bigger drop in the local gun supply.

Also, he said, the buyback programs may be used as a vehicle to foster closer long-term relationships between local police and residents to reduce crime, an effect that is difficult to measure but one that Rosenfeld believes has long-term benefits in controlling crime. But, countered Gary Kleck, a criminology professor at Florida State University, "It's not like we have infinite resources and can spend them on anything. We should focus on something that has some measurable effect."

Kleck argues that free distribution of trigger locks would be a much more cost-effective way of reducing accidental shootings.

In contrast to the typical $50 that buyback programs pay for a gun, he said, "A trigger lock will cost you $10 per gun. Not everyone will use them, but if you think about the type of people who participate in gun buyback programs, they're voluntary participants too."