Data were obtained from the Rochester Youth Development Study, a multiwave panel study of juvenile delinquency and drug use. Affective ties to parents were assessed using Likert-type items from the Child's Attitude Toward Parent Scale. Parental supervision was measured by two items, how often parents knew where their children were and whether parents knew who their children were with. Results showed juvenile delinquency and parental supervision were reciprocally related, whereas affective ties appeared to be a consequence rather than a cause of juvenile delinquency, at least by middle adolescence. In general, interrelationships among variables were more complex than those suggested by earlier unidirectional theories and emphasized the importance of interactional perspectives in understanding links between parenting and adolescent behavior. Adolescent perceptions of parental supervision had indirect effects on affective ties to parents, primarily because lower levels of perceived parental supervision contributed to juvenile delinquency and led indirectly through juvenile delinquency to weaker affective ties. Findings lend support to the notion that lack of parental supervision contributes to the detachment of adolescents from parental influence and support, at least in early adolescence. Conversely, adolescents who are watched feel they are important to their parents and this belief reinforces the quality of the adolescent-parent relationship. Appendixes list analysis variables and juvenile delinquency items by offense type. 93 references, 16 notes, 2 tables, and 2 figures