For many years it has been suspected that severely impaired somatic growth during early postnatal life can be associated with the subsequent impairment of mental abilities. This study aimed to test that hypothesis on the basis of data gathered from a prospective whole population survey of infant development in south London. A year's birth cohort of 1558 full-term singletons was monitored; 47 otherwise healthy cases with serious growth faltering in the first year were recruited. Mental and psychomotor abilities were assessed at 15 months. Potentially confounding psychosocial variables, including cognitive stimulation received at home, were measured contemporaneously. A statistical model was constructed that enabled the timing, duration and severity of growth faltering to be used as predictors of mental functioning. Up to 37% of the variance in cognitive and psychomotor outcome at 15 months can be explained by the model. The first few postnatal months appear to constitute a "sensitive period" for the relationship between growth and mental development.