We tested the hypothesis that fatigue in MS is related to a dysfunction in cortical areas involved in movement preparation. Thirty-three patients with clinically definite MS (16 with fatigue MS-F, 17 without fatigue MS-NF) and a relapsing-remitting course, matched for disease severity and duration, disability scores and level of depression were enrolled. They underwent a combined assessment with magnetic resonance imaging (MRI) and transcranial magnetic stimulation (TMS) and, for the electrophysiological study, were compared with 12 healthy controls. MRI was used to assess regional and total lesion-load volume (LL) on T1- and T2-weighted sequences and total brain volume on T1-weighted sequences. With TMS we tested central motor conduction time, short intracortical inhibition (SICI) and facilitation (ICF), pre-movement facilitation related to a simple reaction time paradigm and the effect of short trains of 5-Hz repetitive TMS (rTMS). No significant differences were found in total and regional LL between MS-F and MS-NF, except for a significant increase in frontal lobe LL in MS-F. Neurophysiological assessment did not disclose any difference of SICI and ICF among the three groups. The significant increase of MEP size produced by 5 Hz rTMS in controls was absent in both MS-NF and MS-F. MS-F lacked pre-movement facilitation compared with MS-NF and controls. The lack of pre-movement facilitation and the increased frontal lobe lesion load were significantly correlated to the FSS score, suggesting that central fatigue in MS is probably due to a dysfunction of cortical motor areas involved in movement preparation.