Twenty-four Ss were required to detect infrequently and randomly occurring visual and auditory signals under six different schedules. In four of the sessions there were 40 visual signals and, respectively, 0, 20, 40, or 80 auditory signals. In two sessions there were 20 visual signals and 20 or 60 auditory signals. Auditory signals (increments in a steady noise background) were relatively easy to detect; visual signals (double jumps of a moving clock hand) were relatively difficult to detect. Missed detections of visual signals increased with time on task for each condition. Latency of response to visual signals increased progressively for all conditions except the one with no auditory signals (control condition). Latency was least for this control condition but decreased monotonically with number of auditory signals in other conditions. Visual missed responses were intermediate in frequency for the control conditions and apparently decreased monotonically as a function of the ratio of auditory to visual signals in the other conditions. Latency was higher for conditions with more visual signals, but no such simple relationship is apparent for missed detections.