The first systematic genetic study of Parkinson's disease (PD) was carried out by Mjönes in 1949. His results indicated autosomal dominant transmission with 60% penetrance. These conclusions, however, were long discounted because oligosymptomatic and atypical relatives were counted as secondary cases without clear justification. Subsequent surveys of patient and twin studies failed to confirm evidence of familial concentration and the hypothesis of a genetic etiology was over-shadowed by interest in possible environmental neurotoxins. A growing accumulation of pedigrees of histologically confirmed Lewy-body PD over the past several years has refocused attention on genetic factors. Fluorodopa positron emission tomography (PET) studies in oligosymptomatic co-twins of probands and recent clinicopathologic studies of atypical cases have provided retrospective support for Mjönes' methodology. A survey of a personal series of PD patients showed that the majority of those for whom pedigree information was available were familial. This along with another recently reported series confirm Mjönes' data showing similar segregation ratios for siblings and parents, a low ratio of maternal:paternal transmission and a marked asymmetry in the distribution of ancestral secondary cases. These findings favor monogenic autosomal dominant inheritance and show reason to argue against a multifactorial etiology or heteroplasmy. The clinical evidence justifies searching for gene loci linked to PD. Available methods are briefly outlined. Preliminary investigations have examined possible allelic associations, e.g., with alleles of MAO-A and debrisoquine hydroxylase and linkage to the tyrosine hydroxylase gene on chromosome 11p15.5 has been excluded in one study of juvenile familial parkinsonism. Linkage mapping studies are presently underway.