Research on the use of the Rorschach in the diagnosis of schizophrenia is reviewed. The analysis of this research (from that done by Rorschach to the most recent) indicates that schizophrenics do not respond in any unique fashion to any particular determinant or any particular card. The research also indicates that the computation of percentages and ratios, no matter how complex, does not facilitate the identification of a schizophrenic process. What the research does demonstrate is that schizophrenia, in its overt as well as latent and borderline forms, is detectable on the Rorschach through an analysis of the patient's thinking. The thinking of the schizophrenic reveals highly personal, illogical, and bizarre associations to the blots. This analysis of thinking is not accomplished through assessment of the formal properties of the blots, e.g., the analysis of response to color, movement, shading, the color-to-movement ratio, or form level (low form level proves to be a function of psychoticism, not schizophrenia, per se) but more in terms of the phenomenology of the responses.