P300 was recorded, using an 'odd ball' paradigm, in 18 parkinsonian patients before and during dopaminergic monotherapy. The data were compared with those of a homogeneous standard group of 20 subjects. The main finding was an increase in the P300 latency of parkinsonian patients before therapy, which recovered during dopaminergic monotherapy. In 11 voluntary healthy subjects the same therapy did not produce a reduction of the P300 latency. The data are discussed in relation to a possible dopaminergic component in P300 origin.