A total of 20,117 offenders were studied. Arrestees were first divided into groups based on similar demographic information and criminal records prior to the time of a chosen arrest. Average termination rates were estimated for each group based on criminal activity after the chosen arrest using maximum likelihood techniques. The analysis indicates that termination rates generally decreased as the number of prior arrests increased for white, 17-29 year-old offenders (black offenders did not exhibit this variation). Black and white, 30-39 year-old offenders had a lower termination rate than offenders who were 17-29 years old or over 40 years old at the time of arrest. The study concludes that before findings on criminal-career termination rates can be used in developing crime-control policies, additional research is required. The study should be replicated in other jurisdictions using expanded information sources, refined offender attributes, and additional analyses to distinguish between the two competing explanations for a decrease in termination rates with increases in the number of prior arrests. 32-item bibliography and appended tables and equations