The extent of smoke exposure via mother's milk and passive smoking was investigated in a prospective, longitudinal matched-pair study by comparison between children, whose mothers smoked substantially throughout pregnancy and nursing period and children whose mothers did not smoke. Our preliminary results show that not only infants of smoking mothers but also those of smoking fathers show reduction of birth weight. Smoking mothers weaned their babies earlier than non-smokers. Cotinine concentrations in breast milk depended on the number of cigarettes smoked. The highest urinary excretion of cotinine (as expressed by ng cotinine/mg creatinine ratios) were observed in infants fully breast-fed by smoking mothers. After weaning the values were in the same range as those of formula-fed infants of smoking mothers (exposed to passive smoking only). In the group of non-smokers only small or undetectable amounts of cotinine were found. Thus it is demonstrated that both nursing and--to a lower degree--passive smoking contribute to the exposure of infants to nicotine and its metabolite cotinine.