The experimental hepatic cirrhosis was induced either by bile duct ligation (BDL) or by pretreatment with dimethyinitrosamine (DMNA). The pharmacokinetics of theophylline were studied after a single intravenous or a single oral administration. Using the ultrafiltration method, protein-drug binding experiments were also carried out. The bilirubin level was several-fold increased by BDL, but not by DMNA treatment. The albumin content was decreased in both cirrhotic groups. The total clearance (Clt, ml/kg/hr) of theophylline in both hepatic cirrhosis groups significantly decreased and the terminal half-life $(t_{1/2})$ in the cirrhotic rats was increased about two-fold after intravenous and oral administration. The volume of distribution at steady state (Vdss, ml/kg) was increased slightly in the cirrhotic groups. Protein binding in BDL $(8.67{pm}4.85%)$ decreased about four-folds, but in DMNA $(73.00{pm}9.85%)$ similar result war observed as compared with the control. Increased free fraction of theophylline did not increase the volume of distribution in BDL. Therefore decreased total body clearance of theophylline was mainly due to decreased intrinsic clearance of theophylline in the liver. The absolute bioavailability of theophylline in these experiments was between 63.8 and 72.8%(66.1% in BDL, 63.8% in Sham operated and Control, 72.8% in DMNA). These results suggest that in the experimental hepatic cirrhosis model, administration route does not affect the disposition of theophylline.