Recently several chronic toxicity/carcinogenicity studies of di-isononyl phthalate (DINP) have been reported. These studies defined effect levels for liver tumors in male and female F344 rats at dietary levels exceeding 700 mg/kg/day; the no effect levels were 359 mg/kg/day in males and 442 mg/kg/day in females. Similar results were found in male B6C3F1 mice, but in female mice a significant increase in liver tumors was found at 336 mg/kg/day, making 112 mg/kg/day the NOAEL for liver tumors in that sex and species. DINP induces peroxisomal proliferation, and that, along with evidence that DINP is not mutagenic, is presumptive evidence for peroxisomal proliferation as the underlying mode of action for liver tumor development. To further explore the relationship between peroxisome proliferation and tumor induction in male and female mice, indicators of peroxisomal proliferation including liver weight, peroxisomal volume density, induction of peroxisomal enzyme activity, enhanced replicative DNA synthesis, and rates of apoptosis were measured at all of the dietary levels used in the chronic study in mice (500, 1500, 4000, and 8000 ppm, or approximately 100, 300, 800, and 1600 mg/kg/day). Liver weights, peroxisomal volume, and peroxisomal enzyme activity were significantly elevated in both male and female mice at the tumorigenic levels. Cell proliferation was also elevated in male and female mice, although the increases at levels below 4000 ppm in female mice were not significantly different from control values. Apoptosis was elevated at the 4000 and 8000 ppm levels, paralleling the increases in liver weight. These data along with previous results satisfy the criteria of the International Agency for Research on Cancer (IARC) and demonstrate that peroxisomal proliferation was indeed the mode of action for DINP-induced liver tumor induction in mice.