The search for indicators of vulnerability has been important in schizophrenia research, but, as in many areas, progress has been impeded due to the heterogeneity of schizophrenic disorders. How one conceptualizes the observed heterogeneity is dependent upon the particular genetic model to which one subscribes. In this article, we delineate several models that may account for the distributions of vulnerability indicators in groups of schizophrenic patients and their ill and well relatives. We present these models for heuristic purposes so that they may serve to guide the interpretation of data with respect to the issue of teasing apart familial and nonfamilial environmental components of putative vulnerability indicators. It is suggested that investigators will profit by: a) efforts to combine psychiatric genetic paradigms in order to maximize the yield of family study data, and b) thinking in terms of comparing the ability of different models to account for research findings.