The assets and liabilities of current methodologies for studying the development of schizophrenia are reviewed under three major headings: retrospective methods which draw inferences concerning developmental factors in the lives of subjects who are already ill by using data collected after the onset of the illness, childhood records methods which draw developmental inferences about pathology in adults on the basis of developmental data contained in records collected previously during childhood, and prospective or “high risk” methods which examine developmental factors in subjects not yet ill but having a higher than normal probability of becoming ill. The primary focus of the review is upon the relative capacities of existing methodologies to generate developmental statements at a high degree of specificity and validity. Retrospective methods reviewed include the systematic use of the case history interview, direct observations of patients’ behavior, psychological testing, and family interaction experiments. Childhood records methods examined include both “follow-back” methods in which early life events of adult patients are collected from childhood records and also “follow-up” methods in which a sample of subjects for whom childhood records are available is followed up in adulthood. Prospective methods examined include experimental designs employing a variety ofstrategies(longitudinal, cross-sectional, and compound designs) andcriteriafor selecting a high risk population. The advantages of prospective methods in solving methodological roadblocks inherent in retrospective and childhood records methods are discussed along with their liabilities, and details of current high risk projects and current trends in high risk research are described.