1. In the cats trained to follow a target spot with their eyes, activity was recorded from omnipause neurons (OPNs). OPNs discharge at a relatively high steady tonic rate (50-130 spikes/s) during visual fixation and smooth-pursuit eye movements but exhibit a complete cessation of discharge that begins before saccades in any direction. They are located in a compact region of the dorsal pontine tegmentum near the midline, just rostral to the abducens nucleus. 2. The average duration of the horizontal or vertical component of a saccade increases monotonically with pause duration, but a given pause duration is associated with a large range of individual saccade parameters and the timing of the pause, such as the latency from the pause onset to saccade onset or the interval from the maximum saccade velocity to the end of the pause, is no better. However, OPNs can be divided into two distinct groups on the basis of the timing of the pause relative to the parameters of the saccade. One group ceases discharging 32.4 +/- 4.6 ms, on average, before the saccade, while the second pauses 18.2 +/- 3.4 ms before the saccade. 3. Microstimulation at the site of OPNs affects the occurrence and trajectory of saccades but not smooth pursuit or fixation. Sustained electrical stimulation (20 micro A) lasting several seconds prevents the occurrence of saccades while brief trains (10-60 ms), timed to occur early in the saccade, interrupt it in midflight for the duration of the train. The latency to the interruption is about 26 ms. These data support the view that OPNs tonically inhibit the saccadic machinery between saccades and must be turned off to allow a saccade to occur. 4. Almost every (65 of 69) feline OPN exhibited a brief transient increase in activity for visual stimuli moving in any direction with a wide range of velocities. A moving 1 degree spot was generally more effective than a moving full-field, striped background. All units also showed a transient increase in firing when the spot was turned either on or off. Receptive fields plotted with the spot were greater than 250 deg2 and always included the area centralis. Two-thirds of the cells tested also responded to auditory stimuli. 5. Interaction between the excitatory visual input and the saccade-related pause was tested by comparing OPN activity and the saccadic trajectory during eye movements in the dark versus the light and by presenting brief flashes of light during a saccade. During saccades in the dark, the steady firing of OPNs was less than during saccades in the light. Only by stabilizing a flashed spot of light to occur on the area centralis at the beginning of the saccade was it possible to activate an OPN artificially to interrupt the saccade in midflight. Therefore, rather than being instrumental in specifically controlling the saccade trajectory, the visual input, along with the auditory and other sensory inputs, probably serves, under normal visual conditions, to help establish the tonic rate of OPNs. 6...