Understanding of the pharmacology of the narcotic antagonist naloxone has been limited by the lack of a convenient and sensitive method of assay. A radioimmunoassay for naloxone has been developed and is described. It is applicable for drug analysis in either serum or brain. The limit of sensitivity of the assay was 0.1 ng. Naloxone glucuronide, noroxymorphone (nor-naloxone) and morphine were not recognized by the antibody whereas naltrexone and 6-hydroxynaloxone were able to displace naloxone-3H from the antibody. The assay was of sufficient sensitivity to follow the serum levels of naloxone in man for up to 2 hours after an i.v. injection of 0.4 mg. In animal studies, the biologic half-lives of naloxone or morphine (5 mg/kg) were compared after s.c. injection in rats. The peak serum levels A (1 mu/mo), time to peak serum levels (less than 1/2 hour), and serum half-life (40 minutes) were comparable. However, the brain entry and egress of the two compounds differed markedly. Peak brain levels of naloxone occurred within 15 minutes and had declined by 50% within 1 hour, whereas the peak brain levels of morphine were sustained for up to 2 hours. At peak serum levels, the brain/serum ratio for morphine was 0.1 whereas for naloxone it was 15 times greater. We suggest the high brain/serum ratio of naloxone contributes to its potency whereas the rapid egress from the brain is important in the short duration of action of naloxone.