The study used a longitudinal data set that consisted of a sample of 1,932 serious, chronic, male offenders released on parole from the California Youth Authority between July 1, 1981, and June 30, 1982, and followed up over a 10-year period. For arrests recorded between their 7th and 25th birthdays, age- specific and offense-specific rates were calculated for both the sample as a whole (incidence rates) and for active offenders only (lambda). The study found that after peaking in adolescence, the rate of arrest decreased with age for property and violent offenses; the rate of arrest increased with age for drug and "residual" offenses; and the arrest rate remained relatively constant with age, after reaching its peak, when all offenses were considered collectively. The authors argue that, overall, the analyses are more supportive of the claims of Hirschi and Gottfredson (that the relationship between age and crime is universally invariant for all offense types) than the claims of Blumstein and his colleagues. Finally, the authors infer from the analyses that the "net" thrown out in California to capture "violent" and "serious" career criminals through the "Three Strikes" law is likely to incarcerate a great number of offenders who would apparently present a relatively limited threat to society in the years to come. 4 tables, 6 figures, and 36 references