Small surface electrodes were placed bilaterally over the intermediate or lateral cerebellar cortex of cynomolgus monkeys to determine how electrical stimulation of different areas of the cerebellar cortex affected average evoked responses and a sequential forelimb movement. Biphase electrical stimulation was applied between various electrode combinations, and various intensities and frequencies were established for each combination. Transcortical stimulation between the right and left intermediate cerebellar cortex required the lowest intensity (1.5 microC/sq cm/ph) to elicit an average evoked response in the sensorimotor cortex; stimulation between the electrodes over the contralateral intermediate or lateral cortex required slightly higher levels (2.0 microC/sq cm/ph). No response could be elicited from stimulating the ipsilateral cortex. Likewise, 1 minute of transcortical stimulation was more effective than comparable stimulation of the contralateral intermediate or lateral cortex in altering the waveforms of a somatosensory evoked response. Transcortical stimulation also modified the forelimb movement, whereas contralateral stimulation of the intermediate or lateral cortex had little or no effect. Transcortical stimulation at 2.0 microC/sq cm/ph, with frequencies of 150 Hz or higher, increased the time required to execute the forelimb movement but did not affect the accuracy of the movement. High-speed motion pictures indicated that transcortical stimulation decreased the velocity of forelimb movement and in some cases also affected the limb trajectory. These results indicate that consideration should be given to the area of the cerebellum stimulated and to the mode of stimulation, in the hope of achieving optimum clinical benefit.