As the initial step in a systematic genome survey, 16 simple sequence tandem repeat polymorphisms that span the X chromosome at an average spacing of 10 cM were examined for allelic associations with typicalonset Alzheimer's disease (AD). The efficiency of this survey was substantially enhanced by genotyping pools of genomic DNA from 50 autopsy-confirmed AD cases and 50 autopsied controls who were similar in sex ratio, race, and age at death. The frequency of the DXS1047 202-bp allele was twice as common among AD cases (0.45 ยฑ S.E. 0.06) than controls (0.22 ยฑ S.E. 0.05), a finding that was reproduced in an independent and geographically disparate sample. Consistent with Hardy-Weinberg equilibrium, the proportion of women with AD who carried the 202-bp allele, 73% was nearly double that observed for men with AD, 38%. However, the frequency of the 202-bp allele was similar for men and women and the presence of this allele did not affect the age at onset of dementia in either sex. Furthermore, the frequency of the DXS1047 202-bp allele in AD cases and controls was unaffected by the APOE genotype, indicating that these two loci modulate AD risk independently. Finally, the frequency of the 202-bp allele among 50 autopsy-confirmed cases of Parkinson's disease (0.29 ยฑ S.E. 0.06) was indistinguishable from the control value, reflecting relative specificity for this allelic asso-