We made a study of coordinated saccadic eye and head movements following random and predictable horizontal visual targets by applying transcranial magnetic stimulation (TMS) over the cerebellum before the start of the gaze movement. We have found three effects of TMS on eye/head movements under these conditions. Saccadic latency-effect. When stimulation took place shortly before movements commenced, significantly shorter latencies were found between target presentation and commencement of saccades: For predictable, to a lesser extent for random targets and TMS up to 75 ms before start of the saccade, latencies were significantly decreased when compared with no application of TMS. Without stimulation, latencies to random targets were within a range of 120–200 ms. Eye–head interaction-effect. Without TMS, for amplitudes greater than 25°, head movements usually preceded eye movements, as expected, especially for predictive responses. With the application of TMS shortly after the target display, the number of eye movements which preceded head movements, was significantly increased ( p<0.001), and the delay between eye and head movements was reduced or reversed ( p<0.001), compared with gaze movements without the use of TMS. Saccadic peak velocity-effect. Applying transcranial magnetic stimulation at 5–25 ms after the position change of the 60° target, and 50–5 s before the start of eye movement, mean peak velocity of synkinetic saccades increased up to 600°/s, compared with 350–400°/s without the use of TMS. We conclude that transient functional cerebellar deficits caused by the application of TMS can change the central synkinesis of eye–head coordination.