During the past 10 years, electroencephalographic recordings have been taken on children with cerebral palsy, who were attending a special clinic at the Royal Hospital for Sick Children, Bristol. In this paper an attempt has been made to evaluate these records and, in particular, to assess their importance to the clinician.The published findings on the electroencephalogram (EEG) in cerebral palsy are not extensive possibly because of the difficulty in obtaining records from these severely handicapped children. An early survey was made by Perlstein, Gibbs, and Gibbs (1946). Aird and Cohen (1950) found that 85% of spastics and 60 % of athetoids gave abnormal records, but that there was focal pathology in only 32% of athetoids. They suggest that if the EEG shows no evidence of epileptiform disturbance, the chance of a fit occurring is only 1 in 8. They found no correlation between an abnormal EEG and the intelligence quotient.Pampiglione (1958) reviewing the literature says that there is considerable agreement on four points: (a) asymmetry between the records from the two hemispheres is commoner in hemiplegia, (b) EEG abnormalities are recognizable in most cerebral palsy patients who have seizures, (c) one-third of the cases show normal records, and this is commoner with athetoids than with spastics, (d) there is a smaller proportion of abnormal EEGs among patients whose condition is ascribed to prematurity or anoxia than in those who have a known cerebral injury or infection.Denhoff and Robinault (1960) found that 24 % of a group of 95 cerebral-palsied children had normal records. Lundervold (1960) compared air-encephalographic and electroencephalographic findings and showed that the combined results had some prognostic significance.