Dysbindin (DTNBP1) is a positional candidate gene for 6p22.3-linked schizophrenia (SZ). However, so far, no disease-causing alleles have been identified. DTNBP1 is immediately adjacent to JARID2, a member of the ARID (AT-rich interaction domain) family of transcription modulators. We have previously suggested that proteins which bind to AT-rich domains could play a role in SZ pathogenesis. Consequently, we explored the possibility that JARID2 itself could be a candidate gene for 6p22.3-linked SZ. We used a case control design to analyze single nucleotide polymorphisms (SNPs) and insertion/deletion variants affecting AT-rich domains in both the DTNBP1 and JARID2 genes. Three of the DTNBP1 SNPs analyzed had previously been shown to be associated with SZ. We did not detect any significant difference in allele, genotype or haplotype distribution for any of these DTNBP1 markers. However, we did detect a significant difference in allele distribution for a tetranucleotide repeat polymorphism in the JARID2 gene that affects an AT-rich domain. A significant increase in short alleles (less than 11 repeats) was found in patients with SZ (chi(2) = 7.02; P = 0.008). No other JARID2 marker displayed statistically significant allele and genotype distributions. Our findings suggest that JARID2 should be viewed as a candidate gene for 6p22.3-linked SZ.