Nondepolarizing muscle relaxants (MRs) diminish the indirectly evoked single twitch due to their binding to the postsynaptic receptors. Additionally, the MRs produce progressive diminution of successive twitches upon repetitive stimulation (fade). Our study addresses the generation of fade as observed under clinical situation. The study was conducted in two phases. In the clinical part, we have evaluated the time course of twitch depression and fade following the administration of several doses of three MRs (rocuronium, pancuronium, and cisatracurium). In the second part, we have modified our model of neuromuscular transmission to simulate the time course of twitch depression and fade. The MR was assumed to bind to a single site on the presynaptic receptor to produce fade. The rates of interaction with the presynaptic receptors were characterized in terms of the arbitrarily assigned equilibrium dissociation constant and the half-life for dissociation of the presynaptic complex. A method was developed to relate the release of acetylcholine to the occupancy of the presynaptic receptors. The strength of the first and the fourth twitch was calculated from the peak concentration of the activated postsynaptic receptors, i.e., of those receptors with both sites occupied by acetylcholine. Our results indicate that, while the affinity of the MR for the presynaptic receptor plays little role in the time course of fade, the rate of dissociation of the complex between the presynaptic receptors and the muscle relaxant may be critical in determining the time course of fade. Tentative estimates of this parameter are offered.