OBJECTIVE To evaluate the hepatic effects of colonic carcinogenesis induced by azoxymethane at different doses and times of exposure in rats. METHODS Forty-four Wistar rats were divided into four groups. The animals were eight weeks at the beginning of the experiment. group 1 received 1.0 ml of saline intraperitoneally once a week for two weeks. Group 2 received 15 mg/kg of azoxymethane intraperitoneally once a week for two weeks. These animals were killed at the 15th week of the experiment. The animals of group 3 received saline intraperitoneally once a week for two weeks. Group 4 animals received 20mg/kg of azoxymethane intraperitoneally once a week for two weeks. These animals were killed at the 26th week of the experiment. The fragments of liver tissue were stained with hematoxylin and eosin and evaluated microscopically. RESULTS Groups 1 and 2 differed significantly in relation to steatosis, no difference having been found between group 3 and group 4. However, in group 4 we observed pre-neoplastic lesions (foci of altered, clear, vacuolated, basophilic, amphophilic tigroid, oncocytic, small or acidophilus cells, spongiosis and peliosis) and neoplastic lesions (adenomas and colangiomas) containing atypical hepatocytes in between, not identified in group 3. CONCLUSIONS In the model of colorectal carcinogenesis, preneoplastic and neoplastic hepatic lesions appear and evolve in proportion to the time of exposure and dose of azoxymethane.