To investigate the brain stem control of saccadic eye movements, the paramedian pontine reticular formation (PPRF) in rhesus monkeys was temporarily and partially inactivated with the local anesthetic lidocaine. The influence on ipsilesional, contralesional, and upward saccades was examined. While the effects of the inactivation on contralesional and upward saccades were inconsistent and small, consistent and marked modifications were observed for ipsilesional movements. For ipsilesional, horizontal saccades, all lidocaine injections caused a decrease in peak velocity and a proportional increase in duration, which substantially altered the shape of the velocity profile. The rise in duration usually fell short of preventing hypometric saccades at the peak of the effect. However, as the lidocaine effect dissipated, the amplitude often returned to control, even though the velocity and duration remained compromised. For ipsilesional, oblique saccades, the effect of lidocaine on the horizontal component was similar to that for horizontal saccades. The vertical component of oblique saccades was also influenced, albeit to a much lesser extent: the duration of the vertical component typically increased, while the vertical peak velocity either decreased or exhibited no significant change. These results were compared with simulations of three prominent models for cross-coupling oblique saccades. In general, these results of the temporary inactivation of PPRF are consistent with the predictions of local feedback models for saccadic control.