A series of studies in schizophrenia, using several varieties of the twin study method, are reviewed and five major conclusions or hypotheses drawn. A modified twin study design—the co-twin comparison method—is described, with which these conclusions and hypotheses can be tested; and two additional major issues—specificity to schizophrenia, and the antecedent-consequent question—dealt with. A stress model of schizophrenia pathogenesis is presented, which attempts to integrate genetic, biochemical, and experiential variables.