This study aimed to examine the relationship between rates of smoking cessation in adults and whether or not they have dependent children at home. Previous work has suggested that among women, particularly poor women, caring for young children might work against giving up smoking, while the relationship between having children and giving up smoking in men is largely unexplored. The analysis employed data from 3 years (1988, 1990 and 1992) of the General Household Survey, and assessed rates of self-reported smoking cessation among adult ever-smokers of cigarettes aged 16-49 years. After adjustment for a wide range of potential confounding variables, adults with dependent children were more likely to have given up smoking than those without. Among women there was a linear increase in the odds of cessation with each additional child. Using the combined data from the 3 years, and by comparison with those with no children, the adjusted odds of cessation were 1.42 (95% CI 1.21-1.67) in women with one dependent child, 1.77 (1.50-2.09) in those with two and 2.19 (1.74-2.74) in those with three or more. The effect was marginally present in men, with an increase in the odds of cessation of about 25% in those with two or more dependent children. (Estimated odds from the combined data 1.30 (1.09-1.50) for men with two dependent children and 1.23 (0.97-1.51) for those with three or more.) The association of increased rates of smoking cessation with number of dependent children did not vary with level of deprivation. It is concluded that having children is associated with smoking cessation in parents, whether poor or affluent. Health education campaigns targeted at families with children could aim to amplify this effect.