The Department of Radiation Therapy, Memorial Hospital for Cancer and Allied Diseases, and the Division of Medical Research, Sloan-Kettering InstituteThe few controlled studies of the use of antiemetic drugs in the treatment or prevention of radiation nausea and vomiting reported in the literaturel-9 have pointed out some of the problems inherent in evaluating new drugs in this clinical setting. A large number of factors other than drugs can influence the results of treatment, and with modem methods of radiation therapy only 10 to 20 per cent of unselected patients will develop nausea or vomiting and require medication for radiation sickness1• 9 ; and, furthermore, as many as 50 to 70 per cent of these patients will be relieved by placebos. 1 • 4 • 6 • 9 The difficulties involved in discriminating between an active drug and a placebo under these circumstances are obvious and call for a wellcontrolled clinical assay designed in a way such that these factors might be taken into account.