The term “radiation sickness” as applied in the management of patients with malignant disease refers to a well known but poorly defined symptom-complex. The symptoms include nausea, vomiting, anorexia, nervousness, insomnia, malaise, fatigue, headache, dizziness, tachycardia, etc. Many theories have been advanced as to the cause of clinical radiation sickness including choline poisoning, chloride depletion, glycogen dwindling, cholesterol alteration, and vitamin deficiencies (1). The etiology of the condition, however, remains unestablished. Numerous agents have been utilized in the treatment of this disorder, but there have been few properly controlled clinical trials for its adequate evaluation and response to management attempts. As a consequence, there has been little appreciation of the high incidence of placebo-responders to anti-emetic medications until quite recently (2, 3). This situation has resulted in the continued empirical use of a wide range of drugs for radiation nausea with little understanding of the clinical and radiotherapeutic variables involved in its induction and amelioration. Presented in the following sections are the results of careful statistical analyses in a “double-blind” study of the factors involved in the induction of radiation nausea and the effects of placebos and drugs upon it. Materials and Method The subjects of this study were 113 patients receiving placebos and 122 receiving drugs for nausea and/or vomiting during the course of radiation therapy for malignant disease. Approximately one-third of the subjects were treated at the U. S. Naval Hospital, San Diego, and two-thirds at the Strong Memorial Hospital, Radiation Therapy Centers. During 1961 to 1962, GS-95 (Torecan) was kindly supplied by the Sandoz Pharmaceuticals; during 1962 to 1963, SKF-478 (Diphenadol) by the Smith, Kline and French Laboratories at the Strong Memorial Hospital; and during 1962 to 1964, GS-95 by the Sandoz Pharmaceuticals at the U. S. Naval Hospital, Radiation Therapy Centers. The drugs and placebos were randomized alternately among patients or by the random-number method. The dose of medication was usually one tablet or capsule, four times daily, half an hour before meals and at bedtime. A few patients received two tablets or capsules four times daily when immediate response (within one day) was not obtained with the usual dose. Two days of medication were chosen as the time period of evaluation of response in order to minimize the influences of extraneous variables which might produce subsidence of the symptoms, e.g., changes in psychological, environmental, or radiotherapeutic factors. The improvement in nausea and/or vomiting was graded subjectively: 0 for none, 1 for equivocal or fair, 2 for good, and 3 for excellent or complete improvement. For the purpose of analysis, the responses were evaluated as binary rather than multinomial quantities, taking the dividing point between response-grades 1 and 2.