More than half of the adult male felons released on probation and monitored by California researchers were convicted of new crimes within 40 months, and a third of those went right back on probation, according to a landmark federally-funded study.

The $185,000 report by the Rand Corp., the first systematic look at the effectiveness of probation for adult felons, concluded that the majority of those kept out of prison remained "a serious threat to the public."

It recommended much more intensive surveillance if probation is to remain a popular alternative to sending felons to the nation's overcrowded prisons.

The study, funded by the National Institute of Justice, tracked 1,672 California men on probation over 40 months and found that 65 percent were rearrested, 51 percent convicted and 34 percent incarcerated.

About 51 percent of the new crimes were property offenses, such as burglary or theft, and 24 percent were violent offenses, such as murder, rape, assault or robbery, "the crimes that society considers most threatening," the 112-page report said.

Titled "Granting Felons Probation: Public Risks and Alternatives," the report, released today, fuels debate over the growing use of probation, the system of sending convicted criminals back to their communities rather than to prison on the condition they stay out of trouble and check in regularly with their local probation department.

The report noted that 60 percent to 80 percent of all criminal convictions in the country end in probation sentences, although the majority of convicted murderers, rapists and robbers are sent to prison.

"Prisons are crowded and probation is being increasingly used to catch the overflow," said Joan Petersilia, a Rand criminologist who was the study's principal author.

Other probation experts said the study generally confirmed the results of earlier, more limited tracking of probationers but said California's problem may have been exacerbated by state budget limits.

Barry Nidorf, chief probation officer in Los Angeles County, said individual case loads in his office had doubled since 1981. He has a staff of 1,200 to supervise 55,000 adults and 18,000 juveniles, allowing many to assume they can violate the conditions of their probation undetected.

Alan M. Schuman, social services director for the District of Columbia's Superior Court and vice president of the National Association of Probation Executives, said budget limits in California had left probation officers "just barely able to keep names straight" in many cases.

He said he expects the rate of new crimes by felons on probation in the District would be lower than in California's Alameda and Los Angeles counties, the targets of the Rand study.

Rand's Petersilia said it is possible that recidivism rates are higher in the two counties than in places with less severe funding problems. However, she suggested that many other areas are heading toward similar budget woes.

Petersilia said her group has received funding to compare the rearrest rates of the targeted probationers with similar felons who were released after serving time in prison. Previous studies suggest that the recidivism rate among probationers will be the lower one.

Given the enormous expense of building more prisons -- 31 states are under court orders or consent decrees to reduce overcrowding -- and the apparent failure of probation for most adult felons, the Rand report recommended a compromise -- "intensive surveillance" systems that keep much closer watch on the lives of convicted felons at large.

At least eight states have experimented with such programs. "They are very hot right now," said Schuman, who added that he would like to try the idea in the District.

Mostly designed for nonviolent offenders, they usually require work or community service, face-to-face contact with probation officers several times a week and "house arrest" restrictions.

A New Jersey program, the Rand group reported, requires participants to be home from 10 p.m. to 6 a.m., report in daily, maintain employment and pay restitution to their victims. Preliminary data indicates that one of 226 persons was returned to prison for a serious offense in the first 14 months of the program.

Some jurisdictions have tried fitting probationers with special bracelets that allow electronic sensors to track them. Civil libertarians have expressed doubts about this method, but Petersilia said in an interview, "I don't have the problem with it that some people have. It makes a lot of sense to me . . . especially when the alternative is to be confined in prison."

Her report, however, admitted that intensive surveillance would be costly -- perhaps $2,000 to $5,000 a year per offender compared with $300 per year for probation. The report suggested charging the offender for part of the cost.

It costs $14,000 a year to keep one adult in a California prison, the report said.