The effects of smoke exposure via mothers' milk and/or via passive smoking during the first year of life were investigated in a prospective longitudinal matched-pair study. The somatic and mental development of 69 infants whose mothers smoked more than five cigarettes per day throughout pregnancy and continued smoking after childbirth were compared with 69 children of non-smoking mothers. At birth, mean body weight of neonates from smoking mothers was significantly lower than the weight of neonates from non-smoking mothers. This weight difference between the two groups was no longer significant in infants at 12 months of age. With the methods employed by the authors, neither psychomotor nor mental development was affected by smoke exposure during pregnancy and early infancy. Infections of the lower respiratory tract were more frequent in the children of smoking mothers. These mothers weaned their babies earlier than non-smokers, but the different feeding behaviour did not influence any of the clinical parameters that were investigated in this study. In order to evaluate the extent of smoke exposure, cotinine was measured in children's urine and in breast milk once a month throughout the first year of life. Cotinine in the urine was significantly dependent on feeding behaviour: infants breast fed showed concentrations 10-fold higher than those who were bottle fed. Cotinine excretion in urine of infants from smoking mothers, who were not breast fed (nicotine exposure via passive smoking only) was even higher than that of adult passive smokers. If infants from smoking mothers were breast fed, their urinary cotinine excretion was in the range of adult smokers.(ABSTRACT TRUNCATED AT 250 WORDS)