This study was performed to determine the changes in pulmonary function in patients randomised to receive treatment with four cycles of bleomycin, etoposide and cisplatin (BEP) (27 patients) or with four cycles of etoposide and cisplatin (EP) (27 patients) for disseminated non-seminomatous testicular cancer. This enabled us to establish whether effects other than those due to bleomycin determined the detrimental effects of BEP on lung function assessments. Slow inspiratory vital capacity (VC), the transfer factor of the lungs for carbon monoxide (TLCO), the diffusing capacity of the alveolo-capillary membrane (Dm), the pulmonary capillary blood volume (Vc) and the transfer factor of the lungs for carbon monoxide per unit alveolar volume (KCO) were determined before and at 3 week intervals during chemotherapy. Both groups, similar in terms of factors that may influence pulmonary function, showed during therapy a significant decrease in TLCO compared with the pretreatment value. Only at the end of the therapy was a significant difference in TLCO between both groups observed. Dm diminished also significantly in both groups during treatment, but differences between both groups were not seen. VC and Vc decreased in patients receiving BEP but remained constant during treatment with EP. It can be concluded that the Dm, KCO, and the widely used TLCO are not suitable parameters to monitor specifically pulmonary toxicity induced by bleomycin as part of a multidrug regimen. However, VC and Vc appear to be proper lung function assessments which reflect specifically alterations induced by bleomycin.