Associations between psychosomatic symptoms and two measures of social support, the quantity of social ties and the quality of primary intimate relationships, were investigated. The sample consisted of 214 predominantly low-income mothers of young children. The quality of relationships with family members was inversely associated with psychosomatic symptoms, whereas the quality of husband and boyfriend relationships was not. In contrast, the quantity of social ties was inversely related to psychosomatic symptoms among all the women. The quantity of social ties, the quality of relationships as modified by type of intimate, and the baseline level of symptoms measured five years earlier were significant predictors of psychosomatic symptoms among this sample of women.