Signs of riboflavin deficiency require many months to develop in patients placed on riboflavinrestricted diets (1-6). The demonstrations that riboflavin deficiency induced by dietary restriction or riboflavin antagonists resulted in regression of rodent tumors (7-13) prompted several efforts in this laboratory to produce riboflavin deficiency in patients with neoplastic disease (14, 15). Riboflavin antagonists were used in these studies with the objective of producing the deficiency state rapidly, since a regimen requiring many months of dietary restriction would not lend itself to therapeutic evaluation in patients with metastatic cancer. These earlier investigations did not result in the production of clinical evidences of riboflavin deficiency or in objective demonstrations of tumor regression. The administered riboflavin antagonists, selected on the basis of their effectiveness in rats, were shown to be metabolized differently in man. The most active of these agents against rat tumors, 6,7-dimethyl-9-(2'-acetoxyethyl) -isoalloxazine, was rapidly and almost completely hydrolyzed in man, although a significant amount of the absorbed drug was excreted unchanged by rats. With the objective of producing human riboflavin deficiency, an analog was sought that would be an effective riboflavin antagonist in rodents and would be metabolized similarly in rodents and in man.The present paper describes studies with 1O-Ddulcityl -7,8-dimethyl-isoalloxazine, galactoflavin,