A whole population survey of an inner-city health district (population 130,000) was undertaken in order to study the association between social deprivation and non-organic growth delay in preschool children. Potential cases were identified from health clinic records. Cases comprised white, full-term singletons, whose weight and height lay below the tenth centile at 4 years of age. Allowance was made for parental stature. A comparison group was closely matched for socio-economic conditions, and other salient variables. Case children were comparatively significantly delayed in all areas of their cognitive development. One-third were seriously retarded and likely to require special education.