Objective: To explore genetic and environmental influences on individual differences in autistic traits in early adulthood and to test if there is assortative mating (non-random partner choice) for autistic traits in the general population. Design: Twin family study using structural equation modeling. Setting: Population-based twin family sample from the Netherlands. Participants: Twins aged 18 years (n=370) and their siblings (n=94); parents of twins (128 couples). Main Outcome Measure: Self-reported Autism-Spectrum Quotient (AQ) scores, a quantitative measure of autistic traits. Results: Autistic traits were continuously distributed in the population. Twins and siblings did not significantly differ in AQ scores; men obtained significantly higher AQ scores than women (in twin-sibling sample, P<.001; twin-parent sample, P=.02). Individual differences in endorsement on autistic traits show substantial heritability (57%). No significant shared environmental influences were detected. The genes affecting autistic traits appear to be the same across the sexes. The correlation in AQ score between spouses was low and not significant (Pearson r=.05; P=.59). Conclusions: Previous general population twin studies reported high heritability for autistic traits in childhood and early adolescence. This study extends these findings to late adolescence and yields no evidence for sex-specific genetic influences on autistic traits in later stages of development. As autistic traits show substantial variation in the general population, future genetic studies may be facilitated by measuring autistic traits on a continuous scale like the AQ. No evidence for assortative mating for autistic traits was found, suggesting that, in the general population, there is no passive or active partner selection for autistic traits. © 2007 American Medical Association. All rights reserved.