Factors associated with a higher level of incest are lower socio-economic status, family discord and separation, and social isolation. This study also investigates the Westermark hypothesis of the aetiology of sibling incest, which maintains that proximity in childhood leads to incest avoidance in adulthood. This hypothesis has been supported by studies of non-related individuals raised together as young children. This is the first test of the proximity hypothesis with actual siblings within the family unit. More than 500 individuals responded to a questionnaire asking about family background, the nature of early interactions with siblings, and the presence and extent of later sexual contacts. Respondents with sibling sexual experience tended to come from a lower socio-economic class and from crowded conditions; they also changed residence more frequently, had fewer friends, and experienced greater family discord than did respondents without sibling sexual experiences. There were indications that heterosexual incestuous behavior was more likely when siblings had a large age difference between them, and when they were not raised together as young children, but measures of proximity and physical intimacy within a household were not predictive of later sexual contacts. 24 references, 7 tables. (Author abstract modified)