A variety of studies have reported possible genetic associations between bipolar affective disorder and different loci using relative risk (case-control) comparisons. An alternative approach is to construct a contrast group using parental alleles which were not transmitted to an affected individual [Falk and Rubinstein, 1987: Ann Hum Genet 51:227-233]. We have used both approaches to test for possible associations between alleles of the dopamine D3 receptor gene and bipolar affective disorder. For relative risk studies, the probands of multiple incidence bipolar affective disorder families have been compared to alcoholic and psychiatrically normal contrast groups. Nontransmitted allele approaches have used bipolar affective disorder and alcoholic probands in which both parents were available for genotyping. Using the BalI restriction enzyme site polymorphism of Lannfelt et al. [1992: Psychiatr Genet 2:249-256], we have found no differences in the allele or genotype frequencies for bipolar vs. alcoholic or psychiatrically normal controls. In contrast, we have found evidence for an increased frequency of allele 1 and allele 1 containing genotypes in transmitted alleles from bipolar families.