Feeding bile acids (BAs) to rodents has been used to study BA signaling and toxicity in vivo. However, little is known about the effect of feeding BAs on the concentrations of BAs in serum and liver as well as the dose of the fed BAs that causes liver toxicity. The present study was designed to investigate the relative hepatotoxicity of individual BAs by feeding mice cholic acid (CA), chenodeoxycholic acid (CDCA), deoxycholic acid (DCA), lithocholic acid (LCA), or ursodeoxycholic acid (UDCA) at concentrations of 0.01, 0.03, 0.1, 0.3, 1.0, or 3% in their diet for 7 days. The data demonstrate that (1) the ability of the fed BAs to produce hepatotoxicity is UDCA<CA<CDCA<DCA<LCA; (2) the lowest concentration of each BA in the feed that causes hepatotoxicity in mice is CA and CDCA at 0.3%, DCA at 0.1%, and LCA at 0.03%; (3) BA feeding results in a dose-dependent increase in the total serum BA concentrations but had little effect on liver total BA concentrations; (4) hepatotoxicity of the fed BAs does not simply depend on the concentration or hydrophobicity of total BAs in the liver; and (5) liver BA-conjugation enzymes are saturated by feeding UDCA at concentrations higher than 0.3%. In conclusion, the findings of the present study provide guidance for choosing the feeding concentrations of BAs in mice and will aid in interpreting BA hepatotoxicity as well as BA-mediated gene regulation.