Ethanol-induced emesis were investigated using Suncus murinus and the emetogenic mechanisms of ethanol were compared with those of cisplatin. Intraperitoneal injection of ethanol caused dose-dependent emesis with ED50 value of 22.3% (v/v) when injection volume was adjusted to 4 ml/kg. Intraperitoneal and subcutaneous injection of acetaldehyde also caused dose-dependent emesis (ED50 = 3.5% (v/v) with an extremely shorter latency (6% i.p.: 1.0 +/- 0.3 min cf. 40% ethanol: 13.0 +/- 1.9 min). Neither ethanol nor acetaldehyde caused emetic responses when injected intracerebroventricularly. Pretreatment with disulfiram, an inhibitor of liver aldehyde dehydrogenase, potentiated the emetogenic effects of ethanol. Surgical abdominal vagotomy, which blocks cisplatin-induced emesis completely, did not prevent ethanol-induced emesis. 5-HT3 receptor antagonists, which also cause complete inhibition of cisplatin-induced emesis, did not affect the responses. However, ethanol-induced emesis was prevented by the pretreatment with 8-hydroxy-2-(di-n-propylamino)tetrarin hydrobromide (8-OH-DPAT) and N-(2-mercaptopropionyl)-glycine (MPG) dose-dependently. The tackykinin NK1 receptor antagonist (+)-(2S, 3S)-3-(2-methoxybenzylamino)-2-phenyl-piperidine (CP-99,994) also attenuated ethanol-induced emesis. Taken together, these results suggest that 1) acetaldehyde is probably responsible for ethanol-induced emesis, 2) active site for ethanol maybe peripheral, 3) ethanol-induced emesis is mediated by free radicals, and 4) mechanism of ethanol-induced emesis and that caused by cisplatin are different in many respects, although in some they are similar and that the precise pathways remain to be identified. Therefore, the tolerance to emetogenic effects of cisplatin in alcoholic patients cannot be explained as a simple cross desensitization of the pathway.