In Exp. I rats received 100% or 50% reinforcement for blocks of 10 lever-presses separated by 15-sec. intertrial intervals. During 360 acquisition trials 50% reinforcement led to longer latencies and response times than 100% reinforcement. In Exp. II 600 trials of 100% reinforcement of blocks of 10 lever-presses preceded the switch to 50% reinforcement for the partial reinforcement group. 50% reinforcement led to significantly slower response times than 100% reinforcement during Trials 601 to 900 and, while the crossover typical of the partial reinforcement acquisition effect occurred, 50% reinforcement did not lead to significantly faster response times during Trials 901 to 1200.