This study, based on a sample of 172 children, examined the relation between average afternoon salivary cortisol levels measured at home at age 4.5 years and socioemotional adjustment a year and a half later, as reported by mothers, fathers, and teachers. Cortisol levels were hypothesized to be positively associated with withdrawal-type behaviors (e.g., internalizing, social wariness) and inversely related to approach-type behaviors, both negative and positive (e.g., externalizing, school engagement). Higher cortisol levels at age 4.5 predicted more internalizing behavior and social wariness as reported by teachers and mothers, although child gender moderated the relation between cortisol and mother report measures. An inverse relation was found between boys' cortisol levels and father report of externalizing behavior. A marginal inverse relation was found between child cortisol levels and teacher report of school engagement. Behavior assessed concurrently with cortisol collection did not account for the prospective relations observed,suggesting that cortisol adds uniquely to an understanding of behavioral development.