Replication analyses were undertaken with longitudinal data on adolescent males from the two companion studies: the Denver Youth Survey and the Rochester Youth Development Study. Steps two and higher of the Overt and the Covert Pathways were assessed across studies (i.e., physical fighting and violence in the Overt Pathway and property damage, moderately serious delinquency, and serious delinquency in the Covert Pathway). At each site, cumulative age of onset curves for steps in the pathways followed the expected pattern, with the less serious forms of problem behavior and delinquency occurring first and the more serious forms of delinquency occurring last. The best fit for the Covert Pathways was before age 15, and the Overt Pathway had a better fit across the three sites. The distinction between persisters and experimenters improved the fit, as persisters who advanced to the most serious forms of delinquency tended to proceed through the hypothesized steps in the stipulated order. Persisters and experimenters were also compared on the proportion who entered a pathway at an earlier compared to a later step. The sites varied in the percentage of boys in the Overt and Covert Pathways, respectively; the percentage of participants who persisted in the overt and covert pathway jointly ranged from 21 percent to 35 percent. Implications of the findings for interventions and policy are discussed. 4 figures, 4 tables, and 22 references