Indices of neuroticism and major psychopathology for late adolescents are predicted by parental conflict, but they are better predicted when the gender of the adolescent and the gender of the dominant parent are also considered. More neurotic daughters come from high conflict families with dominant fathers, whereas more neurotic sons come from high conflict families with dominant mothers. The triple interaction of conflict, parental dominance, and gender of the child adds significantly to the prediction of adolescent neuroticism and major psychopathology but not to the prediction of social nonconformity. Social nonconformity was significantly predicted only by parental conflict.In the first half of the 20th century, psychodynamic, conditioning, social learning, and communication theories of psychopathology provided plausible alternatives to earlier constitutional models. Despite the opinion of reviewers that little empirical evidence supported these social-environmental explanations of neurosis and psychosis (Frank, 1965;Jacob, 1975), they remain popular with clinicians. In a recent review that reexamined many of the earlier studies and 15 later reports, Doane (1978) arrived at a more optimistic conclusion regarding the empirical evidence. Doane concluded that several dimensions discriminated families of disturbed from those of normal individuals: patterns of conflict, family coalition patterns, flexibility versus rigidity, family effectiveness and efficiency, and deviant styles of communication. This recent evidence suggests that we are just beginning to develop models of sufficient complexity to cope with the intricacies of family influences.