The concept of arteriosclerotic parkinsonism has created some confusion due to the absence of a precise definition of the parkinsonian syndrome and to a prolonged absence of knowledge concerning the topography of the lesions in "idiopathic" Parkinson's disease as well as the anatomic relations of the basal nuclei. Although certain authors at the beginning of the century suggested that Parkinson's disease could have a vascular origin, an opposition rapidly developed between Parkinson's disease and arteriosclerotic parkinsonism, largely based on the work by Cricthley (1929). The concept of arteriosclerotic parkinsonism is a quite ambiguous notion, but was generally accepted up to the sixties before severe criticisms were made when the anatomo-clinical characteristics of Parkinson's disease were identified. Recent work has again suggested that parkinsonism can be of vascular origin but the observations reported show a heterogeneous collection of lesions with different localizations, clinical expression and clinical courses. Clinically, two rare circumstances can be identified (vascular parkinsonism similar to Parkinson's disease and unilateral vascular parkinsonism due to contralateral lesions) and inversely much more frequent "atypical" parkinsonism syndromes. These differ from Parkinson's disease by their parkinsonism symptomatology (no resting tremor, predominance of gait disorders), the presence of associated neurological signs and lower sensitivity to levodopa. The attribution of the parkinsonism disorders to vascular lesions identified at imaging is often quite hazardous, but a few anatomo-clinical observations have confirmed that vascular parkinsonism does exist. The heterogeneous nature of the observations hinders the development of diagnostic criteria.