An increasing body of evidence shows that many 'environmental' measures are heritable, indicating genetic involvement in environmental exposure (or gene-environment correlation). In the present study we attempt to clarify why three such 'environmental' measures (maternal negativity, paternal negativity and negative life events) are consistently found to be heritable. Through multivariate genetic analysis of a sample of adolescent twins from the UK we show that the heritability of these putative environmental measures can be explained via their association with five behavioural phenotypes: oppositionality, delinquency, physical aggression, depression and anxiety. This is consistent with the notion that being genetically susceptible to certain behavioural difficulties could lead to exposure to certain life events, and this may account for the reported heritability of 'environmental' measures. Results are discussed in the context of possible active, evocative and passive gene-environment correlations.