A sensitive method was developed for the estimation of [3H] cocaine in biological materials. After an injection of 8 mg/kg i.v. in male Wistar rats, peak levels in brain, tissues and plasma occurred within 15 minutes and cocaine disappeared completely from brain and plasma 6 hours postinjection. The T1/2 of cocaine in brain and plasma was 0.4 and 0.3 hour, respectively. No significant differences were observed in the rates of disappearance of cocaine from the subcutaneous site in acute and chronically treated rats after an injection of 20 mg/kg. After a 20 mg/kg s. c. dose, the peak levels of cocaine were attained gradually in 4 hours in brain, tissues and plasma with the exception of heart (0.5 hour) and fat (2 hours). These peak levels shifted from 4 to 2 hours in the chronically treated group. Consistently higher levels of cocaine were found to be sequestered in fat in the chronically treated animals. The T1/2 of cocaine in brain and plasma of chronically treated rats was approximately 1.8 to 2 hours and that in the acutely treated animals, 0.8 to 1 hour. The brain/plasma ratios were also somewhat higher in chronically treated as compared to the acutely treated animals and were indicative of a high affinity of tissue for cocaine. Although cocaine did not persist in brains of acutely treated animals, measurable amounts were shown to persist in brain and other tissues of chronically treated animals long after the disappearance in plasma. Significantly high concentrations of metabolites of cocaine persisted in brain and plasma of acutely and chronically treated animals. No significant differences were observed in the plasma protein binding of cocaine in control, acutely and chronically treated rats. Unchanged cocaine was excreted in very small amounts in rat bile and approximately 36% of the dose (5 mg/kg i. v.) was excreted as metabolites 3.5 hours after injection. Excretion of free cocaine in urine and feces after a 20 mg/kg s. c. dose in acutely and chronically treated rats was 1.2 and 1.5%, respectively. Significantly higher excretion of total radioactivity occurred in feces in the chronic group (35.9%) as compared to the acute group (22.1%). Benzoylecgonine, benzoyl norecgonine, ecgonine methyl ester and ecgonine were identified as urinary metabolites in both acute and chronic animals. In addition, evidence was obtained for the presence of a phenolic metabolite and two other hydroxylated metabolites (with hydroxylation presumably in positions 6 and 7 of the pyrrolidine ring). Implications of these observations with respect to systemic toxicity, the absence of tolerance and physical dependence liability of cocaine are discussed.