A physiological dosimetric model of the disposition of TCDD in the rat (Kohn et al., Toxicol. Appl. Pharmacol. 120, 138-154, 1993) was extended to include effects of dioxin on serum concentrations of thyroid hormones in the rat. The extended model included distribution of blood among major vessels and tissue capillary beds and resorption of TCDD released into the gut lumen from the liver by cell lysis consequent to cytotoxicity. TCDD metabolism was represented by Hill kinetics. Parameter values were estimated by fitting time-course data for a single oral subcutaneous injection of TCDD and dose-response data for biweekly oral dosing. The extended model included new compartments for the thyroid and thyroxine-sensitive tissues (e.g., pituitary, kidney, and brown fat), secretion and tissue uptake of thyroid hormones, binding of 3,5,3'-triiodothyronine (T3) and 3,5,3',5'-tetraiodothyronine (thyroxine, T4) to proteins in blood and tissues, deiodination of iodothyronines, and glucuronidation of T4 by the hepatic UDP-glucuronosyltransferase (UGT) activity induced by TCDD. Secretion of thyroid hormones was modeled as regulated by thyrotropin (TSH), whose secretion was modeled as regulated by the hypothalamic factors thyrotropin releasing hormone and somatostatin. Release of the hypothalamic factors was modeled as under feedback control by the blood T4 level. Induction of UGT was modeled as stimulated by the Ah receptor-TCDD complex. The extended model fit the observed dose-response of P450 isozymes and Ah and estrogen receptors following repeated oral doses with comparable accuracy as the earlier model. The fit to liver and fat TCDD levels following single and repeated oral and subcutaneous doses was improved over the earlier model. The revised model's predicted liver TCDD concentrations at very low doses were verified experimentally. The model reproduced the responses observed for blood T3, T4, and TSH after 31 weeks of biweekly oral dosing of rats with TCDD. The model also predicted responses of UGT mRNA and UGT enzymatic activity comparable to those observed in TCDD-treated rats in experiments whose data were not used in constructing the model. Calculated increases in blood TSH levels are consistent with prolonged stimulation of the thyroid and may represent an early stage in the induction of thyroid tumors identified in previous two-year bioassays. Thus, increases in UGT activity may be useful as a biomarker for tumorigenic changes in hormone levels subsequent to TCDD exposure.