BACKGROUND: Twin studies consistently point to limited genetic influence on attachment security in the infancy period, but no study has examined whether this remains the case in later development. This study presents the findings from a twin study examining the relative importance of genetic and environmental influences on attachment in adolescence. METHODS: The sample included 551 twin pairs aged 15 years recruited from the larger Twins Early Development Study (TEDS). Attachment was assessed using a semistructured interview, the Child Attachment Interview. RESULTS: We found robust associations between MZ twins' scores for Coherence and their overall security of attachment (r = .42, p < .001; kappa = .26, p < .001), but substantially lower associations for DZ twins (r = .20, p = .001; kappa = .09, p = .20), suggesting genetic influence on adolescent attachment (and substantial nonshared environment). Model-fitting analyses confirmed this impression, indicating approximately 40% heritability of attachment and negligible influence of the shared environment. CONCLUSIONS: The results suggest that genes may play an important role in adolescent attachment and point to the potentially distinct aetiological mechanisms involved in individual differences in attachment beyond early childhood.