1. The initial ocular pursuit of small target spots (0.25 degrees diam) that suddenly start to move at constant speed (ramps) was recorded in four rhesus monkeys with the electromagnetic search coil technique. All target motions were horizontal, and both eyes were monitored. 2. In agreement with the observations of Keller and Khan, stationary textured backgrounds substantially reduced the initial eye acceleration achieved during pursuit but did not affect its latency. Correlation techniques were used to assess the changes in the eye speed profiles and indicated that the reduction in eye acceleration due to the background was a linear function of the logarithm of target speed over the range investigated (5-40 degrees/s), averaging 60% with the fastest targets. 3. Selectively excluding the background texture from the path of the target with a horizontal strip of card (vertical width, 4 degrees) reduced the impact of the background only slightly, and, even when the vertical width of the card was increased to 60 degrees, the effect of the background was not entirely eliminated. Thus the effect involves regions of the visual field well beyond the target and is not due simply to the reduced physical salience (contrast) of the target spot. Such spatially remote interactions suggest that the neurons decoding the target's motion have very extensive visual receptive fields. 4. Textured backgrounds also caused similar reductions in the eye acceleration during initial pursuit when, before the ramps, the fixated target spots stepped forward, i.e., stepped in the direction of the subsequent ramps (step ramps). In this situation, as with no steps, initial target ramps were foveofugal. When the fixated target spots were stepped back before moving forward so that initial target ramps were foveopetal, textured backgrounds now also delayed the onset of pursuit, and the reductions in eye acceleration were not seen until some time later when tracking resulted from foveofugal target-ramp motion. Selectively excluding the texture from the path of the target with a narrow strip of card eliminated any delays in the onset of pursuit to step ramps, but the later reductions in eye acceleration were still evident. These step-ramp data indicate that the mechanisms decoding foveofugal and foveopetal target ramps differ markedly in their sensitivity to textured backgrounds. That backgrounds can influence the latency and the initial eye acceleration independently is consistent with the idea that there are independent trigger and drive mechanisms for the decoding of target motions.(ABSTRACT TRUNCATED AT 400 WORDS)