The Christchurch Child Development Study is a longitudinal study of a birth cohort of 1265 New Zealand children who have been studied over an 11-year period using data from multiple sources including parental interview, medical records, teacher questionnaires and direct testing of children. The article provides a review of the major lines of epidemiological research examined in the Study. These include: breast feeding and child health; parental smoking and child health; the effects of low level lead exposure; childhood asthma; nocturnal bladder control; the effects of early hospital admission; the distribution of child health services; and the consequences of private medical insurance. In addition a number of general topics (sample attrition, measurement error, individual differences and causal inference) relating to longitudinal designs are discussed briefly. It is concluded that the longitudinal design is a powerful and cost-effective method of gathering data for general paediatric epidemiological purposes but that research in this area would benefit from an increased use of emerging methods of statistical modelling.