Previous research has shown that low birthweight is a predictor of several adverse educational outcomes, including special educational placement, by middle school age. Most low-birthweight follow-up studies that have extended to school age have focused on very small infants-- < 1500 or < 1000 g; less is known of the school age outcomes for infants with only moderately low birthweight (1500-2000 g). This study examines the prevalence of special educational placement and the relationship of such placement to grade retention, verbal and performance scores on tests of general intelligence, reading and maths achievement scores and classroom hyperactivity among low-birthweight children. In a regional birth cohort of 1105 infants born between 1984 and 1987 and weighing 500-2000 g, 868 children were available for follow-up at age nine. Information on special education placement as well as grade retention, intelligence, academic achievement and classroom behaviour was available on 645 (74% completion rate). Nearly a third of the cohort was classified as needing special education. Special education placement followed a birthweight gradient, occurring among 29.3% of children with birthweights 1500-2000 g, among 32.5% in children 1000-1500 g and 49.4% in children < 1000 g. Among children in special education, a similar birthweight gradient was found for maths achievement and hyperactivity, but not for reading achievement or IQ scores. Among children not in special education, only maths achievement showed such a decline with birthweight. A substantial proportion of low-birthweight children, including those of moderate low birthweight, receive special education services, although the need is greatest among those with the lowest birthweights. Maths achievement declined with birthweight regardless of educational placement. The medical and social risk factors that accompany low birthweight and may account for these findings, require further study.