The study was designed to investigate the possible occurrence of "wearing-off" (WO) during dopamine agonist (DA) monotherapy. Sixty patients with "de novo" idiopathic PD were randomised into one of two DA monotherapy branches to receive oral ropinirole at 15 mg per day, or pramipexole at 2.1 mg per day. DA doses could be increased in the following two years but levodopa could not be added until the study ended. WO was assessed by self-evaluation charts confirmed by a blinded observation of a 30% or greater deterioration in the Unified Parkinson's Disease Rating Scale (UPDRS) motor score. Proc Mixed and Kaplan-Meier curves evaluated treatment variables as a function of time. T-tests were used to compare post-hoc variables reclassified according to WO occurrence. Thirty patients received ropinirole, and 30 pramipexole monotherapy. Eighteen patients (30%) experienced "wearing-off" 15-21 months after beginning monotherapy. No differences were observed between treatments. WO phenomena was observed 3.4+/-0.3 hours after intake of the morning or afternoon dose and consisted of UPDRS score worsening by 11.1+/-2.1 points (69-111% more than "on" score). Statistical evaluation gave evidence of differences between patients who experienced WO and those who did not: UPDRS motor scores obtained at admission to the study were higher (by 3.4+/-0.2 points, p=0.01 t-test) and DA doses at 6-12 months were higher in fluctuating patients. UPDRS motor scores deteriorated, however. similarly and there were no differences, in UPDRS scores recorded in ON conditions, between fluctuating and non-fluctuating patients at the end of the study. Our findings provide evidence of WO phenomena in patients with early PD receiving non-ergolinic DA monotherapy.