BUNNEY and Fawcett have posed a very interesting question: "Could a chemical test aid in the prediction of suicidal intent in depressed patients?" They noted that in view of the high suicide rate (19,000 reported deaths in this country per year) and an estimated 4,000 patients each year who see physicians within a week of the time they commit suicide, such a test would be of great assistance in the prediction of potential suicide patients.1 Their paper presented urinary 17-hydroxycorticosteroid (OHCS) determinations in a group of 36 depressed inpatients. Of these 36 patients, three patients who committed suicide showed high mean urinary 17OHCS. In addition these patients had relatively low clinical ratings of suicidal behavior making such a test of great practical value. The authors have collected serial 24-hour urine collections in a group of hospitalized patients two of whom suicided while under observation. Stimulated by the report