The Exeter Family Study of Childhood Health is a prospective study, set up to test the fetal insulin hypothesis, and to identify genetic polymorphisms that play a role in determining birthweight and early postnatal growth. We recruited 1017 families from a postcode-defined area in central Exeter. Specific inclusion criteria were established to obtain a homogeneous, non-diabetic, UK Caucasian cohort. Detailed anthropometric measurements were taken from both parents at 28 weeks' gestation, and from their children at birth, 12 weeks, 1 year and 2 years of age. Insulin and other biochemical analysis were measured in fasting parental samples and an umbilical cord blood sample taken at delivery. Parental and offspring DNA were extracted to allow molecular genetic analysis of candidate genes implicated in fetal growth. This comprehensive data set is available on a disease-free cohort, and detailed preparatory work has ensured high levels of quality control. The study would welcome collaboration within strict confidentiality rules.