Whether Parkinson's disease patients have been shown to have abnormal visual evoked cortical potentials (VEPs) to flash and pattern stimulation or not, is a matter widely discussed. Actually it is said that this variability may be due to the stimulus method used and how the patients were selected. The retina is rich in dopamine and together with previous animal and human studies, this suggests that the abnormal VEPs in Parkinson's disease patients may be due to a biochemical and electrophysiological disorder in the retina. This hypothesis has been examined by studying the flash VEPs, pattern VEPs and with light electroretinogram in 29 Parkinson's disease patients and matched control subjects. We have found: a) Slower FVEPs and PEVPs latencies in Parkinson disease patients. Flash VEPS were more frequently abnormal than pattern VEPs. We have only found a significant relationship between slow flash VEPs and the patient age. b) Electroretinograms waves (a) and (b) amplitudes were smaller in Parkinson disease patients than in control subjects. b/a coefficient abnormal values must be due basically to wave (a) alterations. These values are statistically significant with respect to the number of years of L-Dopa in the evaluative stage and the evolutive stage of the illness. For all these reasons, we can deduce that an anomaly exists in the dopaminergic pathway in the retina, localized in the plexiform layer and neurophysiologically detectable by a study of the wave variations of the ERG and of the b/a coefficient.