A randomized, controlled study was begun in 1982 on the effect of keyhole-limpet hemocyanin and mitomycin C in the prevention of recurrent superficial bladder cancer (stages pTa to pT1, grades 0 to 3) in 44 patients. All tumors were resected and all patients were presumed to be free of tumor at initiation of the prophylactic instillations. Before the bladder instillation program was begun all patients in the keyhole-limpet hemocyanin group Ia were immunized with 1 mg. keyhole-limpet hemocyanin intracutaneously and then monthly bladder instillations of 10 mg. were given. The control group Ib received 20 mg. mitomycin C monthly. Of the 21 patients in the keyhole-limpet hemocyanin group Ia (mean followup 20.7 months) 3 (14.2 per cent) had recurrences, compared to 9 of 23 (39.1 per cent) in the mitomycin C group Ib (mean followup 18.3 months). The over-all preventive effect was significantly better (p less than 0.05, chi-square) in keyhole-limpet hemocyanin-treated patients than in those given mitomycin C. In 1984 a new single drug study (group II) was started with keyhole-limpet hemocyanin alone, administered as in group Ia. Of 81 patients in group II (nonrandomized, mean followup 22.8 months) 17 (20.9 per cent) had recurrences. Of the patients given keyhole-limpet hemocyanin 20 of 21 (95.2 per cent) in group Ia and 70 of 81 (86.4 per cent) in group II had complete and partial prevention (downgrading), compared to 16 of 23 (69.5 per cent) in group Ib. Our study was established to analyze the effect of a new method of immunotherapy; no adverse local or systemic side effects were noted.