Hypotheses which have been proposed to account for the unusual seasonal birth pattern observed in schizophrenic populations are discussed. These competing hypotheses were tested by retrospectively studying season of birth in 975 schizophrenics divided according to family history of psychiatric illness. Information was obtained from case notes, item sheets, and questionnaires sent to general practitioners. The results were inconclusive, but there was a trend for high genetic risk cases to be born less often in the first quarter of the year. Although no clear support could be provided for one or other season of birth hypothesis, it is tentatively suggested that a seasonal constitutional damage factor may be responsible for the excess of births described in schizophrenic populations in the early months of the year.