Three publications on the carcinogenicity of N-nitrosodiethylamine (NDEA) in the livers of F-344 or Wistar rats were examined for concordance of the data. Two reports recorded the appearance of tumors after treatment with NDEA, although one used a different dosing schedule that included phenobarbital promotion. Two studied glutathione S-transferase-placental positive (GST-p+) foci in liver at several doses. One also analyzed DNA for adducts from NDEA. This analysis revealed that when the dose was calculated in molecules/kg/day, the thresholds for the incidence of liver tumors were different by about 1.5 orders of magnitude. But when the dose was calculated as the total cumulative dose, the thresholds for tumor appearance (about 1020.3 molecules/kg) were in agreement within the error of calculation. Combining the data for GST-p+ foci revealed remarkable agreement between the two reports and a threshold for the appearance of these foci at about 1019.5 molecules/kg of total cumulative dose of NDEA. DNA adducts fit an exponential curve better than a linear. GST-p+ foci and adducts from NDEA were observed at doses below the threshold dose for the appearance of tumors. These results suggest that: cumulative dose is a better metric than daily dose and that adducts and GST-p+ foci appear at doses below those at which tumors appear. These results further support the observations of the authors that thresholds for carcinogenicity of this genotoxic carcinogen exist and that adducts and altered foci appear at lower doses than the threshold for carcinogenicity.