A retrospective analysis of adopted and non adopted children attending a children's mental health centre between 1959 and 1973 indicated a higher than expected number of males and a significantly higher number of professional fathers than expected in the adopted group. A sub-group of adopted children identified as having learning problems by reports of the parents and teachers had a different pattern of test scores on the Wechsler Intelligence Scale for Children from a non adopted group, who were also selected for learning difficulties. On the parent's and clinician's ratings, the adopted group had more parent-identified hyperactivity symptoms and clinicians' diagnoses of hyperactivity. Treatment data indicated that they used more stimulant medication; this was particularly significant for the early adoptees. The early adoptees were considered to be more hyperactive by their parents than the late adoptees. The late adoptees were more impaired on psychometric testing than those placed early, before six months of age. These findings are discussed in relation to possible explanations for the higher than expected prevalence of adoptees seen in clinical populations.