By analyzing data from a full census of almost 22,000 inmates, as well as a randomly selected sample of 1,000 pre-adjudicated inmates, the study obtained limited offense and demographic information on the entire census, as well as detailed data on the prior records and post-release criminal behavior for the sample of 1,000 inmates. Data used in this report were extracted from automated information systems maintained in Los Angeles County and the hand-coded abstractions of official records. The study found that on January 15, 1996, there were 21,758 inmates in custody of the Los Angeles Jails; 11,967 were pre-adjudicated, and 9,971 were adjudicated. On the day of the census, 87 percent of all jail inmates were male and 13 percent were female; the average age was 32 years. The study determined that the Los Angeles jails were occupied almost entirely by persons charged or convicted of current felony offenses, combined with extensive criminal records. Thus, the jails now housed only the "worst of the worst," and competition for scarce jail beds seemed to be increasing. The study concluded that few inmates currently housed in Los Angeles Jails were good candidates for intermediate-sanctions programs; and jail capacity should be expanded to permit lengthier incarceration for the more dangerous pre-adjudicated and post-adjudicated inmates. There were few in the system who should be selected for diversion programs, and those given intermediate sanctions should receive more surveillance and treatment under enhanced conditions. For long-range planning, the study recommends that Los Angeles County develop an ongoing ability to simulate various criminal justice policies and the effects they would have on criminal justice agencies. Such a sentencing simulation or "population projection" model could take into account changes in demographics, crime rates, police arrest practices, length of sentences imposed, and projected time served. 12 figures, 19 tables, and 31 references