Inspiratory loading in awake humans is associated with electroencephalographic signs of supplementary motor area (SMA) activation. To provide evidence for a functional connection between SMA and the diaphragm representation in the primary motor cortex (M1(DIA)), we tested the hypothesis that modulating SMA activity using repetitive transcranial magnetic stimulation (rTMS) would alter M1(DIA) excitability. Amplitude and latency of diaphragm motor evoked potentials (MEP(DIA)), evoked through single pulse M1(DIA) stimulation, before and up to 16 min after SMA stimulation, were taken as indicators of M1(DIA) excitability. MEPs from the first dorsal interosseous muscle (FDI, MEP(FDI)) served as a control. Four SMA conditioning sessions were performed in random order at 1-wk intervals. Two aimed at increasing SMA activity (5 and 10 Hz, both at 110% of FDI active motor threshold; referred to as 5Hz and 10Hz, respectively), and two aimed at decreasing it (1 Hz either at 110% of FDI active or resting motor threshold, referred to as aMT or rMT, respectively). The 5Hz protocol increased MEP(DIA) and MEP(FDI) amplitudes with a maximum 11-16 min poststimulation (P = 0.04 and P = 0.02, respectively). The 10Hz protocol increased MEP(FDI) amplitude with a similar time course (P = 0.03) but did not increase MEP(DIA) amplitude (P = 0.32). Both aMT and rMT failed to decrease MEP(DIA) or MEP(FDI) amplitudes (P = 0.23 and P = 0.90, respectively, for diaphragm and P = 0.48 and P = 0.14 for FDI). MEP(DIA) and MEP(FDI) latencies were unaffected by rTMS. These results demonstrate that 5-Hz rTMS over the SMA can increase the excitability of M1(DIA). These observations are consistent with the hypothesis of a functional connection between SMA and M1(DIA).