Seventy male Sprague-Dawley rats were trained to discriminate which of two levers to press for milk reinforcement on a VI-20 schedule of reinforcement on the basis of whether they were injected subcutaneously with 0.75 mg/kg diazepam or 10.0 mg/kg pentylenetetrazol. Following discrimination acquisition, a dose-response function was generated for each drug during 5-min extinction periods. Subjects were then assigned to one of seven groups on the basis of their per cent responding during saline testing. Each group was injected with 5 mg/kg diazepam and then given a 5-min extinction test at intervals of 2, 4, 6, 8, 12, 16, 20, or 24 h subsequent to injection. The results indicated that at the shorter time intervals the animals responded on the diazepam lever. However, as the time interval between injection and testing lengthened, responding on the PTZ bar gradually increased until by 12 h following injection with diazepam the animals were responding as though they had received an injection of 5 mg/kg PTZ. Following this period of rebound, choice behavior returned to baseline by 24 h post-injection.