The aim of this study was to examine environmental and gene – environment processes linking parenting (i.e., affect, control, responsiveness) and preschool children's behavioral adjustment difficulties (e.g., noncompliance, conduct problems) by using bivariate genetic analyses of parents' and observers' ratings. The sample included 120 identical and same-sex fraternal twin pairs (M age = 43 months). Data sources included in-home observations, interviews, and parents' reports. Observers' ratings of children's difficult behaviors included shared and nonshared environmental variance. In contrast, parents' ratings of children's conduct problems showed genetic and nonshared environmental variance. Observer-rated maternal behavior included shared and nonshared environmental variance, although maternal responsiveness also included child genetic variance. Parent self-reported negative and positive affect included shared and nonshared environment as well as child genetic variance. There was no evidence for gene – environment interaction or dominance. Higher levels of difficult behavior and conduct problems covaried with higher levels of maternal negative affect and control and lower levels of maternal positive affect and control. Shared environmental mediation of these correlations was found for observations, whereas genetic and nonshared environmental mediation was found for parents' ratings. In general, estimates of shared environmental variance and mediation were greatest for observational data, and estimates of child genetic variance and mediation were greatest for parent-rated data. The implications of this pattern of findings for genetic research on family processes are discussed.