To assess the cortical contribution to breathing, low-intensity transcranial magnetic stimulation (TMS) was delivered over the motor cortex in 10 subjects during: (i) voluntary static inspiratory efforts, (ii) hypocapnic voluntary ventilation (end-tidal CO(2), 2.7±0.4% mean±SD), and (iii) hypercapnic involuntary ventilation (end-tidal CO(2), 6.0±0.7%). Electromyographic activity (EMG) was recorded from the scalene muscles (obligatory inspiratory muscles) and was significantly suppressed by TMS at short latency (17.2±1.7ms). The scalene EMG was reduced to 76±8% and 76±7% in voluntary breathing and the static inspiratory effort, respectively, but only to 91±10% during the involuntary ventilation, significantly less than during the two voluntary tasks (p