Transcranial magnetic stimulation (TMS) has opened important perspectives on the pathophysiological bases and potential targets of treatment strategies for idiopathic Parkinson's disease (IPD). Studies have been mainly focusing on motor cortical inhibitory phenomena. However, differences in patients and methods caused several discrepancies, particularly on the so-called long-latency cortical inhibition (LICI). We wanted to challenge such controversies by studying early, drug-naïve patients, and by reproducing the original method that detected a pathologic LICI in IPD. We studied the motor potentials evoked in the first dorsal interosseous muscle on the more and the less parkinsonian side of the body in 18 asymmetrical untreated IPD patients in the early stages of their disease. We had 12 healthy controls. The TMS variables were the active motor threshold, the size of the motor-evoked potential, the cortical silent period and LICI. Average active motor threshold was similar in patients and controls, but its variability was significantly higher among patients (P<0.05). There was a trend for the cortical silent period to be shorter on the more affected side of the patients (P=0.1). Patients, especially on their more affected side, exhibited a late LICI peak, which was absent among controls (P<0.05). This effect was independent of the silent period duration. However, patients and controls having a short silent period also had a shorter LICI (P<0.05). Changes in LICI had a strong trend to correlate to the severity of the parkinsonian signs (P=0.1). Thus, the present study definitely reinforced several previous TMS findings in IPD as a feature of the "pure" disease pathophysiology. The pathologic enhancement of late LICI can act as a candidate physiological hallmark of IPD, to be tested in various diagnostic or therapeutic circumstances.