Previous studies have suggested that among affected sib pairs with autism there is an increase in the frequency of twins over what would be expected in comparison to the prevalence of twins in the general population. In this study we sought to determine whether "sub-threshold" autistic traits were more pronounced in twins than in non-twins. The Social Responsiveness Scale (SRS) was administered in an epidemiologic twin sample (n=802) and in a separate population-based sample of non-twins ascertained from a local school district (n=255). For males (but not females), the mean SRS score was significantly higher among twins than among non-twins. As has been suggested for autism, twin status may incur increased liability to sub-threshold autistic symptomatology, particularly in males.