The purpose of this study was to investigate the effects of green tea ingestion on hepatocarcinogenesis before and after its initiation. Male Sprague-Dawley rats were fed an AIN76A diet with or without green tea. Initiation was induced by a single dose (200 mg/kg) of diethylnitrosamine at week 4 and 0.02% (w/w) 2-acetylaminofluorene was supplied in the diets. The control group had free access to water for 13 weeks (CTR13). Tea infusion was provided from the beginning of the experiment for 13 weeks (PRE13) or from the post-initiation stage until week 13 (POST13). Three other groups (CTR24, PRE24 and POST24) were added to examine the longer-term effects (24 weeks) with the same experimental design. The percentage area of liver sections that were positive for hepatic placental glutathione S-transferase (GST-P), which was used as a marker of preneoplastic lesions, was smaller in PRE13 ($20.2{pm}5.0%$, $mean{pm}SD$) and POST13 ($26.0{pm}4.8%$) than in CTR13 ($33.2{pm}5.8%$, p<0.05). Over the longer period, the GST-P lesions were significantly smaller for both PRE24 and POST24 ($21.6{pm}8.5%$ and $22.2{pm}4.0%$, respectively) than for CTR24 ($28.6{pm}5.1%$, p<0.05), but there was no significant difference between PRE24 and POST24. The liver content of thiobarbituric acid reactive substances was significantly lower in the tea groups than in the controls (p<0.05). However, no significant differences were observed among groups of GST activity. The results show that tea consumption exhibits a stronger short-term initiation-inhibiting ability in liver carcinogenesis, but over a longer period, the preventive effects of green tea ingestion do not differ in post- and pre-initiation.