The two programs studies operated in the same four counties in central Florida, in the same juvenile courts, following the same juvenile law, and with essentially the same population. One program replaced the other as the result of the termination of a treatment-based pilot program (Project Diversion) and the start of a crime control or punishment-based pilot program (Juvenile Alternative Services Project). The research used multiple group discriminant analysis to examine official statistics for the 667 juvenile court dispositions for the four counties from January 1977 through September 1980. Results revealed that the degree of net widening was reduced when the punishment or crime control program replaced the treatment-based program. Findings indicate that incorporation the concept of punishment into diversion programs provides them with a limiting principle to counterbalance the expansive tendencies of treatment-focused programs, although it will not necessarily eliminate net widening. Tables, notes, and 43 references