The effects of simulator delays on performance, control behavior, and transfer of training were investigated with a group of subjects who had no experience with flight control tasks. Two types of aircraft were simulated: one with highly responsive dynamics and one with sluggish dynamics. Subjects were assigned to one of four time-delay conditions and to one of the two aircraft types. In the first phase of the experiment, subjects participated in fifty trials (ten trials per day) with a particular time delay (50, 100, 200, or 400 milliseconds). After this “training” phase, all subjects “transferred” to the minimum time-delay condition (50 milliseconds) for another fifty trials. The experimental task required that the subjects maintain constant heading and altitude in the presence of pseudo-random roll-rate and pitch-rate disturbances. There were statistically significant effects of time delay on root-mean-square heading and altitude errors in both the training and transfer phases of the experiment. The effect of delay on transfer of training was greater for the aircraft with sluggish dynamics.