Parkinson's disease (PD) is frequently reported to be associated with pesticide exposure but the issue has not yet been solved because the data are inconsistent and the studies suffer from several biases and limitations. The aim of this article is to summarise available biochemical and toxicological data on some pesticides, particularly on paraquat, that might help in the evaluation of epidemiological data. The nigrostriatal system appears to be particularly sensitive to oxidative damage caused by different mechanisms and agents, thus supporting the epidemiological evidence that Parkinson's disease is in fact an environmental disease. In available experimental studies, animals have been treated with a high single or a few doses of pesticide, and have been followed up for a few days or weeks after treatment. Moreover, experimental data indicate additive/synergistic effects of different pesticides that act on different targets within the dopaminergic system. In these conditions and to a different extent, pesticides such as paraquat, maneb and other dithiocarbamates, pyrethroids, rotenone, and dieldrin cause neurotoxic effects that may suggest a possible role in the development of a PD-like syndrome in animals. Although, all the characteristics of PD cannot be reproduced by any single chemical, these data can be of help for understanding the role of pesticide exposure in human PD development. On the other hand farmers are exposed for days or weeks during several years to much lower doses than those used in experimental studies. Therefore, a firm conclusion on the role of pesticide exposure on the increased risk of developing PD cannot be drawn. However, it is suggested that close follow up of survivors of acute poisonings by these pesticides, or identification in epidemiological studies of such subjects or of those reporting episodes of accidentally high exposure will certainly provide information useful for the understanding of the relevance of actual human exposure to these pesticides in the development of PD. Also exposure to multiple pesticides, not necessarily at the same time, should be evaluated in epidemiological studies, as suggested by the additive/synergistic effects observed in experimental studies.