The aim of this study was to investigate the difference in rates of cerebral palsy (CP) between singletons and twins by considering factors that may be predictive of CP. Data were taken from the Scottish Register of Children with a Motor Deficit of Central Origin and the Scottish Morbidity Record series. All children born in Scotland between 1984 and 1990 inclusive comprised the cohort. There were 646 children with CP (370 males, 276 females) of whom 57 were from twin pregnancies. Prevalence of CP was higher in twins than in singletons. Also, for singleton and twin births, the prevalence of CP was higher for infants who had low birth weight for gestational age (GA), were preterm, and who were male. Prevalence of CP by GA followed a different pattern for twins than for singletons, being lower for twins in the middle range of GAs than for singletons. After allowing for GA and birth weight, twins appeared to be at increased risk for CP compared with singletons. The type of CP in singletons and twins also differed with 64.9% of twins having spastic bilateral CP compared with 48.5% for singletons. The aetiology of CP in twins and singletons may differ.