Among the radioisotopes which have therapeutic value is radioactive gold, AU198. It has a half-life of 2.8 days and decays with both beta and gamma emissions. One peculiar advantage of this element is that it can be prepared as a colloid which is chemically stable and relatively inert biologically. Dr. Paul Hahn, of Meharry Medical College, has introduced the use of this isotope for direct injection of tumors and intravenous administration (1, 2, 3). Its value as a means of irradiating body cavities for the control of fluid accumulation is the subject of the present discussion. This use for the isotope was first suggested by J. Muller (4), of Switzerland. During the last two years our group at the Oak Ridge Institute has been carrying on a program of experimental treatment with radioactive colloidal gold. The present discussion is based largely upon material which has already appeared in other publications (5–8). We have injected the isotope into pleural and peritoneal spaces of patients with effusions caused by neoplasms and have attempted to study its fate, distribution, and effects. Some of the patients treated have been relatively unsuitable, having advanced disease with such complications as mediastinal compression and impending intestinal obstruction. In some of these more unfavorable cases, the need to collect information, as well as the rather distant hope of improvement for the patient, was a reason for giving the treatment. In administering the isotope, we have usually aspirated a considerable portion of the fluid, inserted a polyethylene tube through the needle used for aspiration, removed the needle, and then allowed the radioactive colloidal gold to flow through the tube into the cavity. This procedure requires a simple infusion device similar to those used for giving intravenous fluids. Physiological saline, 200 to 400 ml., is used for rinsing. The operator receives radiation doses far below tolerance. The method of administration is illustrated in Figure 1. During the first few hours after the gold has been given, the patient is asked to move about frequently to aid free mixing in the cavity. There is a tendency for the gold in the pleural cavity to settle in the costophrenic angles and for that in the peritoneal space to concentrate in the pelvic region. I t is important, therefore, for the patient to be in a head-down position at least part of the time immediately after administration. The optimal dosage of gold has yet to be determined. We have employed various amounts, including some that were probably larger than necessary. Most of the patients treated recently have been given 150 mc. for ascites and 75 mc. for pleural effusion. In a few cases we have repeated the treatment of the same cavity several weeks after the initial dose.