This review describes the symptomatic presentation of schizophrenia with onset in childhood. Phenomenologic data from an original sample of 35 children with onset before the age of 12, from the University of California, Los Angeles (UCLA) Childhood-Onset Schizophrenia Research Program, are presented and compared with similar data from two other major studies. Similarities and differences between childhood-onset and adult-onset forms of the disorder are discussed, with special emphasis on gender ratio and mode of onset. Among the 35 children in the UCLA study, 80 percent presented with auditory hallucinations, 74 percent with flat or inappropriate affect, 63 percent with delusions, 40 percent with formal thought disorder, and 37 percent with visual hallucinations. In the vast majority of cases onset was insidious. The mean age at onset of general psychiatric symptoms was estimated to be 4.6 years, the mean age at onset of psychotic symptoms was 6.9 years, and the mean age at diagnosis was 9.5 years. The phenomenology of the UCLA sample is compared with two other major studies of childhood-onset schizophrenia. The relative frequency of core symptoms, with the exception of thought disorder, was strikingly similar across the three studies, as was the mode of onset. The groups were also similar in age at diagnosis, gender ratio, and IQ. Limited comparisons with studies of first-onset schizophrenia in adults suggest basic similarities between schizophrenia with onset in childhood and adulthood. The qualitative nature of the symptoms reported is similar to that seen in adult cases with the expected developmental variations, for example, delusions are less complex in children and reflect childhood themes. Limited data from a cross-sectional assessment using DSM-III criteria indicate that the relative frequency of core symptoms may also be similar to that seen in adult cases. In contrast to previous reports, this review suggests that the high male to female ratio seen in childhood-onset cases represents a continuum with young adult cases, and gender ratio does not truly distinguish childhood-onset from (young) adult-onset forms of the disorder. Schizophrenia with onset before adolescence does seem to differ from later-onset cases in the very high rates of insidious as opposed to acute onset. The insidious onset may also help explain the clinical observation that in some children psychotic symptoms, particularly those of long duration, can be relatively ego-syntonic.