In the view of animal welfare considerations, we investigated the suitability of modifying the rat liver micronucleus test with partial hepatectomy to include administration of an analgesic drug to minimize pain and distress as much as possible. The effects of the analgesic, buprenorphine, on the genotoxicity evaluation of structural chromosome aberration inducers (cyclophosphamide, diethylnitrosamine and 1,2-dimethylhydrazine) and numerical chromosome aberration inducers (colchicine and carbendazim) were examined. The genotoxicants were given orally to 8-week-old male F344 rats a day before or after partial hepatectomy and hepatocytes were isolated 4 days after the partial hepatectomy. Buprenorphine was injected subcutaneously twice a day with at least a 6-hr interval for 2 days from just after partial hepatectomy. As results, buprenorphine caused neither change in clinical signs (except for one animal death) nor increase in the incidence of micronucleated hepatocytes of vehicle treated animals. In the case of concomitant treatment of buprenorphine and a genotoxicant, one out of 8 animals died in each group given buprenorphine with cyclophosphamide, carbendazim or colchicine (lower dose level only). Slight changes in clinical signs were noted in the group given buprenorphine with cyclophosphamide or carbendazim. A statistically significant increase in the incidence of micronucleated hepatocytes was obtained in concomitant treatment of buprenorphine and genotoxicant compared with genotoxicant alone for 1,2-dimethylhydrazine, colchicine and carbendazim. It is concluded that use of buprenorphine as an analgesic drug to minimize pain and distress for rats that are given partial hepatectomy is not appropriate under the present experimental conditions, because it could enhance the general toxicity and genotoxicity of the test chemical.