1. In a series of 2,100 consecutive admissions to the Veterans Administration Hospital, Lyons, N. J., 72% of the schizophrenics were photographed in the standardized somatotype position shortly after admission.2. These somatotype photographs were rated by Dr. William H. Sheldon, who had no knowledge as to the diagnosis, disposition, or any other pertinent information about the patients. Since these patients were included in a much larger series made up of psychoneurotics, neurological cases, and nonneuropsychiatric patients, in addition to the schizophrenics, the rater had no way of knowing even that they were schizophrenics.3. The patients were allowed to follow their normal course in the hospital without awareness on the part of the staff as to their somatotype rating. The staff had no particular training in possible relationships between somatotype and psychiatric factors.4. In all, 1,000 relatively consecutive admissions were somatotyped in the present series and an extremely detailed review made of all pertinent material, including records while in service, laboratory data, cardiac and ECG consultations, symptomatology, etc.5. Two years after the first patient in the series was admitted (and 6 months after the last series admission), the disposition of the patients was reviewed.6. There was found to be a significantly positive correlation between mesomorphy and good prognosis and a suggestively poor one with high endomorphy.7. This correlation between prognosis and somatotype proved to be independent of correlations between somatotype and diagnosis. In other words, relationship between prognosis and somatotype is more basic than relationship between prognosis and diagnostic subcategory.8. Significant correlations were found to exist between somatotype and diagnosis: Mesomorphs tended to be paranoid and not hebephrenic, whereas, ectomorphs tended to be hebephrenic and not paranoid.9. It was also found that those who were overendowed, or underendowed from an absolute point of view, had a poor prognosis.10. On the basis of other parts of the study, it is suggested that with the somatotype as a frame of reference a much-needed factor will be introduced into the organic approach to schizophrenia, making comprehensible much of the data that are now characterized only by their great variability.Abstract Teaser