Results of previous studies suggest that N-desmethyladinazolam, the major metabolite of adinazolam in man, contributes substantially to psychomotor effects and sedation observed following adinazolam administration. Therefore, the pharmacokinetics and pharmacodynamics of N-desmethyladinazolam were explored following administration of single oral doses of placebo and solutions contianing 10, 30, and 50 mg N-desmethyladinazolam mesylate in a double-blind, randomized, four-way crossover design to 15 healthy male volunteers. Plasma concentrations of N-desmethyladinazolam were determined by HPLC. Psychomotor performance tests (digit symbol substitution and card sorting by fours and suits), memory tests and sedation scoring were also performed following drug administration. N-Desmethyladinazolam pharmacokinetics were dose independent over this range. Dose-related performance effects were observed at 1,2, and 6 h after dosing. Memory was likewise affected at 2 h. Psychomotor performance decrements correlated with log N-desmethyladinazolam plasma concentrations. Analysis of the relationship between percentage decrements in digit-symbol substitution and plasma N-desmethyladinazolam using the Hill equation revealed a EC50 of 325 ng/ml. These results establish the relationship between N-desmethyladinazolam plasma concentrations and performance effects; these data will be helpful in assessing the contribution of N-desmethyladinazolam to clinical effects observed after adinazolam administration.