The control exerted by various portions of fixed-time and fixed-interval schedules was assessed with a trace-conditioning procedure. The intervals were segmented into 10 bins. In all but 1 of those bins, the stimuli were presented in different random orders on each trial. In 1 bin, the stimulus was the same on each trial. The position of this trace stimulus was varied across phases. The results indicated that a trace stimulus can come to control behavior and that differential control can extend to even the second tenth of an interfood interval. The results were interpreted as indicating that traditional explanations of the rate loss in earlier portions of an interfood interval are inadequate and that models such as Palya's (1993) bipolar model or Miller and Schachtman's (1985) comparator model may provide a principled framework with which to understand within-trial effects.