Background: Autistic spectrum disorders (ASD) are thought to be caused by a combination of genetic and environmental risk factors, but identifying specific environmental triggers has proved difficult. Problems during the birth process have been shown to be associated with ASD but whether these problems are an independent cause of ASD, or a consequence of an existing liability for ASD, is unknown. Previous research has shown that many apparently environmental variables show genetic influence. Objectives: The current project aimed to explore whether perinatal problems are associated with dimensional measures of autistic behaviours, whether birth problems are themselves influenced by genetic factors, and whether there is evidence that the association between birth problems and autistic traits could be due to environmental influences. Methods: Parents of over 20,000 twin pairs from the Twins Early Development Study, a UK-based longitudinal twin cohort, reported on birth-related problems one year after the twins were born. When the twins reached middle childhood, parents and teachers of over 6000 pairs reported on the twins' autistic behaviours. Results: Weak but significant and positive correlations were observed between birth problems and dimensional measures of autistic behaviours at ages 7-9. Twin model-fitting revealed that child-specific birth problems were modestly heritable (~30%), as well as showing significant shared and nonshared environmental influences. Difference scores between identical (monozygotic, MZ) twins, on a child-specific birth problems scale correlated significantly with MZ twin difference scores on autistic traits. Conclusions: These data reveal that at least some types of birth problems may be driven by genetic influences, raising the possibility that part of the association between birth problems and autistic behaviours could be genetically influenced. The correlations between MZ differences reveal, however, that nonshared environment also plays a significant role in this association.