The value of intracerebral injections as a means of relating brain structure and function is dependent on the degree of site specificity of the injection. The purpose of this study was to compare the distribution over time of naloxone and its quaternary derivative, methylnaloxonium, after intracerebral microinjection. One microliter of tritiated naloxone (NAL) or methylnaloxonium (MN 10.0 ng, 12.8 nCi for both drugs) was infused directly into the n. raphe pontis. Each animal was then decapitated at a specific time (2.5, 5.0, 15, 30, or 60 min), the brain was removed and dissected into hindbrain, cerebellum, midbrain and cortex. Tritium beta emissions of brain homogenates were measured 1 day later, MN remained better localized to the injection site than did the same volume of the more lipophilic NAL. Within 15 minutes, less than 5% of the NAL remained in the hindbrain compared with nearly 40% of the MN. These results that MN may be a better probe than NAL for investigating the relationship of opioid receptor anatomy and function, particularly for dependent variables requiring sustained time courses.