Recent research on the pharmacologic treatment of tinnitus is reviewed, emphasizing studies in which controls have been used. Several double-blind cross-over studies have found that lidocaine can reduce tinnitus in about 50 to 75 percent of subjects. Unfortunately, it cannot be used clinically because it must be administered intravenously and its effects are very brief. Other drugs have been much less successful. A few controlled studies have found success rates between 33 and 56 percent using oxazepam, clonazepam, sodium amylobarbitone, flunarizine, and eperisone hydrochloride. None of these studies have been replicated, however. Closely controlled studies using specified etiologic subgroups with subjective and objective psychophysical measurements are needed.