The ocular following responses elicited by brief unexpected movements of the visual scene were studied in eight rhesus monkeys. Test patterns were random dots except in one experiment when sine-wave gratings were used. Test stimuli were velocity steps of 100-ms duration applied after spontaneous saccades. Two response measures were used: the initial peak in the eye velocity profile (ei), and the average final eye velocity over the period of 110-140 ms measured from stimulus onset (ef). Responses were best when the test ramps began soon after saccades and attenuated progressively as the postsaccadic delay interval was increased: postsaccadic enhancement of ocular following. The decline in ei was roughly exponential: average time constant, 60 ms; average asymptote, 22%. Later measures (ef) were generally less affected. We suggest that this transient enhancement aids the visual suppression of postsaccadic ocular drifts (glissades) and the tracking of moving images newly acquired with a saccade. The magnitude of the postsaccadic enhancement was dependent on the amount of retinal stimulation during the antecedent saccade; when this stimulation was compromised, as when a vertical saccade was made while viewing a grating pattern with vertically oriented stripes, subsequent enhancement of ocular following was much reduced. Further, saccade-like conditioning movements of the visual scene resulted in an enhancement of the ocular following, elicited by subsequent test ramps, that was similar in magnitude and time course to that in the wake of real saccades. We conclude that the postsaccadic enhancement of ocular following is largely due to the visual stimulation produced by the saccade sweeping the scene across the retina. Data obtained with the visual field partitioned into central and peripheral regions (center 20-60 degrees diam) and with gaze centered suggested that the short-latency ocular following system and the enhancement mechanism that modulates it both receive their major inputs from the central 40 degrees of the retina. Further, when this central region was partitioned, enhancement was obtained only when the conditioning and test stimuli were presented to the same region of retina. Visual enhancement showed only weak interocular transfer: the conditioning and test stimuli had to be seen by the same eye to produce appreciable enhancement. These data suggest that the enhancement involves local spatial interactions at an "early" point in the visual pathway before the inputs from the two eyes have converged. When the conditioning and test stimuli impinged on different regions of the retina, brief powerful suppression of ocular following was obtained.(ABSTRACT TRUNCATED AT 400 WORDS)