Recent genome-wide association studies of Parkinson's disease have nominated 3 new susceptibility loci (PARK16-18) and confirmed 2 known risk genes (MAPT and SNCA) in populations of European ancestry. We sought to replicate these findings. We genotyped single-nucleotide polymorphisms in each of these genes/loci in 1445 Parkinson's disease patients and 1161 controls from northern Spain. Logistic regression was used to test for association between genotype and Parkinson's disease under an additive model, adjusting for sex, age, and site. We also performed analyses stratified by age at onset. Single-nucleotide polymorphisms in MAPT (rs1800547; P = 3.1 × 10(-4) ) and SNCA (rs356219; P = 5.5 × 10(-4) ) were significantly associated with Parkinson's disease. However, none of the markers in PARK16-18 associated with Parkinson's disease in the overall sample, or in any age stratum, with P values ranging from .09 to .88. Although our data further validate MAPT and SNCA as Parkinson's disease susceptibility genes, we failed to replicate PARK16, PARK17, and PARK18. Potential reasons for the discordance between our study and previous genome-wide association studies include effects of population structure, power, and population-specific environmental interactions. Our findings suggest that additional studies of PARK16-18 are necessary to establish the role of these loci in modifying risk for Parkinson's disease in European-derived populations. © 2011 Movement Disorder Society.