Careful comparison of symptomatic individuals with normal controls has revealed the primary biochemical abnormality in many human genetic diseases, particularly recessive disorders1. This strategy has proved less successful for most human disorders which are not recessive, and where a single copy of the aberrant gene has clinically significant effects even though the normal gene product is present. An alternative approach that eliminates the impediment of a normal protein in affected individuals is to study homozygotes for the mutant allele2. For virtually all dominant human disorders in which homozygotes have been described, symptoms have been significantly more severe in the homozygote than in the heterozygote3. Thus, these disorders do not conform to the classical definition of dominance which states that homozygotes and heterozygotes for a defect are phenotypically indistinguishable3-5. Instead, they display incomplete dominance, indicating that the normal allele may play a role in ameliorating the disease process. The D4S10 locus, defined by the probe G8 and linked to the gene for Huntington's disease (HD), has permitted us to identify individuals with a high probability of being homozygous for this autosomal dominant neurodegenerative disorder6-9. These homozygotes do not differ in clinical expression or course from typical HD heterozygotes. HD appears to be the first human disease of genetically documented homozygosity that displays complete phenotypic dominance.