Purpose: Our aim was to study whether transcranial direct current stimulation (tDCS) over premotor cortex (PM) can modify the excitability of the ipsilateral primary motor (M1) and somatosensory (S1) cortices via cortico-cortical connectivity. Methods: Ten subjects received, anodal, cathodal and sham tDCS (1mA) over left PM for 15min. PM was determined to be 2cm anterior and 3cm medial to the hotspot of right first dorsal interosseus muscle. Motor evoked potentials (MEPs) were recorded from right first dorsal interosseus (FDI) muscle with transcranial magnetic stimulation over left M1. Somatosensory evoked potentials (SEPs) to right median nerve stimulation were also recorded from left C3'. Both MEPs and SEPs were recorded before, immediately after and 15min after tDCS. Results: The amplitudes of MEPs after anodal tDCS were shown to decrease while those of SEPs tended to increase. In contrast, the effects of cathodal tDCS were opposite to those of anodal tDCS. Statistical analysis (ANOVA) revealed that a significant interaction among INTERVENTION (anodal, cathodal) x TIME (before, after, after15min) on both MEPs (p < 0.01) and SEPs (p < 0.05). Discussion: We infer that decreased MEP amplitudes resulted from inhibitory input to M1 from PM with anodal tDCS over PM, whereas the opposite effect was mediated from PM to M1 with cathodal tDCS. It is likely that changes in S1 excitability reflect the alternation of input-output modulation between M1 and S1. Conclusion: tDCS is useful for modulating the excitability of PM with which plastic functions of M1 and S1 can be assessed.