This study examined the association between women's retrospective reports of smoking during pregnancy and subsequent language, cognitive, behavioural and physical development in their children up to age 9 years. While there was a strong association between maternal smoking and an index of disadvantageous child rearing, maternal smoking was not associated with more general family disadvantage. After controlling for levels of background disadvantage, no relationship was found between reports of smoking and language, cognitive or physical development. However, smoking was related to maternal reports of behaviour problems at age of school entry. Possible explanations for this relationship are discussed.