Background: This study is the follow-up in early adolescence of children born to families with a history of dyslexia ( Gallagher, Frith, & Snowling, 2000). Methods: Fifty young people with a family history of dyslexia and 20 young people from control families were assessed at 12-13 years on a battery of tests of literacy and language skills, and they completed questionnaires tapping self-perception and print exposure. One parent from each family participated in an interview documenting family circumstances (including family literacy) and a range of environmental variables considered likely correlates of reading disability. They also rated their child's behavioural and emotional adjustment and their own health and well-being. Parental literacy levels were also measured. Results: Forty-two per cent of the "at-risk" sample had reading and spelling impairments. A significant proportion of the literacy-impaired group were affected by behavioural and emotional difficulties, although they were not low in terms of global self-esteem. The children in the at-risk subgroup who did not fulfill criteria for literacy impairment showed weak orthographic skills in adolescence and their reading was not fluent. There were no differences in the literacy levels or activities of the parents of impaired and unimpaired at-risk children, and no significant correlation between parent and child reading levels in the at-risk group. The impaired group read less than the other groups, their reading difficulties impacted learning at school and there was evidence that they also had an impact on family life and maternal well-being. Conclusions: The literacy difficulties of children at family-risk of dyslexia were longstanding and there was no evidence of catch-up in these skills between 8 and 13 years. The findings point to the role of gene-environment correlation in the determination of dyslexia.