The quest for an effective "nonlethal" chemical agent like the one that killed more than 100 hostages in Moscow last weekend has tantalized U.S. military and law enforcement officials for years.

But even though the government has undertaken several research projects into incapacitating gases and aerosols since the mid-1990s, the effort has proceeded slowly in the face of thus-far insurmountable technical hurdles and concern about violating the 1993 Chemical Weapons Convention.

A Pentagon spokesman this week issued a statement saying "the U.S. military is not currently involved in any programs or research related to the development or procurement of incapacitating agents," did not plan any such research and has not stockpiled any agents.

But as recently as May 2000, the Defense Department paid $69,931 to a Michigan-based firm to begin a multiphase project "to demonstrate the feasibility of innovative, safe and reliable chemical immobilizing agents."

The first phase of the project was to include animal tests, and the second phase was to include "human volunteer studies." Officials at the Bel Air, Md., office of OptiMetrics, Inc., the contractor, did not respond to telephone inquiries seeking information about the project.

Also in 2000, the Pentagon-funded Applied Research Laboratory at Pennsylvania State University issued a report on incapacitating agents that concluded their development is "both achievable and desirable."

"There was no hard research done, and there has been none done here" on such agents, said Andrew Mazzara, director of the laboratory's Institute for Emerging Defense Technologies. He characterized the study as a review of existing literature on the subject.

Still, Mazzara, a retired Marine colonel who ran the Pentagon's Joint Non-Lethal Weapons Directorate before joining the laboratory in 1999, suggested that "what we saw in Russia almost cries out for more rather than less research into this."

His views clashed sharply with those of Edward Hammond, director of the Austin-based Sunshine Project, a leading opponent of U.S. ventures into nonlethal technology:

"Using chemical weapons, including incapacitating chemical weapons, is a slippery slope," Hammond said. "We've gone down it before, but it seems like we're going down it again."

Next week, the National Academy of Science is scheduled to release "An Assessment of Non-lethal Weapons Science and Technology," which will, in part, evaluate the utility of incapacitating agents. The report was commissioned by the Marine Corps' Joint Non-Lethal Weapons Directorate and the Office of Naval Research.

Advocates and opponents of incapacitating agents agree that the idea is a noble one -- a gas or aerosol that would gently yet immediately render large numbers of people harmlessly unconscious, instantly terminating a hostage crisis or a riot without gunfire, billy clubs or needless violence.

In practice, however, as the Moscow theater debacle showed last weekend, implementing such a remedy is fraught with dangers. When it comes to using these disabling "calmatives," as they known, the margin of error is so narrow as to be nonexistent.

"There is no such thing as a knockout drug," said Alan P. Zelicoff, senior scientist in the Center for National Security and Arms Control, at Sandia National Laboratory. "I can put you down with morphine; I can put you down with valium, I can put you down with barbiturates. But in all cases, I have a high risk of hurting you very badly."

Zelicoff said the opioid drug fentanyl, acknowledged by Russian authorities as the basis of the aerosol pumped into the Moscow theater, has an extremely low "therapeutic index" -- the difference between rendering a person unconscious and hurting or killing the person is very small. Anesthesia, Zelicoff said, is "controlled death."

This problem -- that anesthesia, relaxants or anti-pain analgesics are highly individualized at high doses -- has never been overcome. Many experts agreed that knocking out a heterogeneous population of several hundred people of all ages, all sizes, both sexes and with some of them sitting close to the vents and others far away, is simply not possible with current technology and should never even be attempted.

"The whole idea of nonlethal chemical warfare agents is a myth," said Elisa Harris, a senior research scholar at the University of Maryland and a former Clinton administration National Security Council official. "Anyone who tries to suggest otherwise is ignoring the evidence."

Discussion of this dilemma pervades government studies of incapacitants. C. Parker Ferguson, a key researcher for the Army in the mid-1990s, acknowledged that "it's a very complex situation -- it's hard enough to use them in the operating room without compounding the problem with larger groups."

Still, the difficulties have not stopped researchers. Ferguson, now working as an independent consultant and contract researcher, is listed as principal investigator for the OptiMetrics contract, which dismisses "previous approaches" to the problem as "deficient in one or more technical aspects." Ferguson said he was no longer connected with the project, and would not describe the results to date.

Opponents of incapacitants suggest not only that the research is a waste of time, but also that the use of the agents undermines the Chemical Weapons Convention in many respects.

"You just know our people are saying, 'What the hell are the Russians up to?' " said the University of Maryland's Harris. "Incidents like that could engender greater efforts not only on our part, but in other countries."

Still, noted Ted Prociv, a deputy assistant to the secretary of defense for chemical and biological matters in the Clinton administration, the stakes could be huge in a world where the United States is involved in "police actions" like those in Haiti, Somalia and Bosnia, where large numbers of civilians were involved.

"These rogue countries think nothing of drawing you into a situation where you're surrounded by noncombatants and where you can't kill anybody," said Prociv, president and CEO of the Springfield engineering firm Versar. "You have to have something besides billy clubs and machine guns."