The recognition that some low birthweight babies (<2500 g) were the victims of intrauterine growth retardation rather than premature birth was a milestone in perinatal medicine.13 Up to 10% of all liveborn babies and at least 30% of those of low birth weight suffer from intrauterine growth retardation; their perinatal mortality is four to 10 times higher than that of normally grown babies?both stillbirth and neonatal deaths contributing. Poor growth also exposes the fetus and the newborn to perinatal complications, which leave their scars in the form of later neurodevelopmental disability.4 No widely accepted, reliable definition of intrauterine growth retardation is applicable before birth. Instead, those babies who are small for dates?who weigh, for example, less than the 10th centile for their gestational age at birth?have, by inference, suffered intrauterine growth retardation. This approach may cause problems: the precise gestational age may be uncertain, especially in those pregnancies vulnerable to intrauterine growth retardation; the lower limit of birth weight for gestational age is variously defined as the 3rd, 5th, 10th, or 25th centile, or less than two standard deviations; intrauterine growth and standards of birth weight for gestational age are influenced by ethnic and geographical factors56; small for dates babies may be the result of genetic constraint rather than pathological growth retardation; and, finally, birth weight is only one index of growth failure, and babies of normal weight may none the less have failed to achieve their genetic growth potential.