Halothane (HAL) is an inhaled anesthetic that induces a severe, idiosyncratic liver injury also referred to as HAL hepatitis, in 1 out of 6,000-30,000 human patients. We used known human risk factors (female sex, adult age, genetics) as well as probable risk factors (fasting and inflammatory stress) to develop a murine model having the known characteristics of human HAL hepatitis. Female and male BALB/cJ mice treated with halothane developed dose-dependent liver injury within 24hrs; however, only females developed severe liver injury. Livers had extensive centrilobular necrosis, inflammatory cell infiltrate, and steatosis. Fasting rendered mice more sensitive to HAL hepatotoxicity, and 8 week-old female mice were the most sensitive compared to males of the same age and younger (4 week-old) females. HAL-treated female BALB/cJ mice had higher plasma concentrations of tumor necrosis factor-alpha than male HAL treated mice. Also, neutrophils were recruited to the liver more rapidly and to a greater extent in HALtreated female BALB/cJ mice. AntiCD18 serum attenuated HAL-induced liver injury in the female mice, suggesting that neutrophil migration and/or activation are required for injury. Ovariectomized (OVX) BALB/cJ mice developed only mild liver injury at a HAL dose (15mmol/kg, ip) that produced severe liver injury in control mice. Plasma interferon-gamma (IFN-γ) was elevated 10-fold in HAL-treated females compared to similarly-treated OVX and male mice. IFN-γ knockout mice were resistant to severe HAL-induced liver injury. The deactivation of NK cells with anti-asialo GM1 treatment attenuated liver injury and plasma IFN-γ compared to IgG-treated control mice. Mice with mutated perforin, a protein involved in granule-mediated cytotoxicity, were protected from HAL, whereas wild-type mice were not. Furthermore, HAL increased the activity of NK cells in vivo, as indicated by increased surface expression of CD69, an early activation marker. Hepatocyte surface proteins are altered in response to HAL treatment in vivo, as indicated by the decreased expression of self, MHC class I molecules, H-2D^d, and increased expression of stress ligand, Rae-1. In conclusion, this is an animal model of an idiosyncratic adverse drug reaction that is based on human risk factors and produces reproducible, severe hepatitis from HAL exposure with lesions characteristic of human halothane hepatitis. Furthermore, this dissertation provides evidence that IFN-γ, perforin, and NK cells have essential roles in the development of HAL hepatotoxicity and contribute to the sexual dimorphic response seen in HAL-treated mice.