Linkage disequilibrium (LD) mapping was used to follow up reports of linkage between reading disability (RD) and an 18 cM region of chromosome 6p21.3-22. Using a two-stage approach, we tested for association between RD and 22 microsatellite markers in two independent samples of 101 (Stage 1) and 77 (Stage 2) parent/proband trios in which RD was rigorously defined. The most significant replicated associations were observed between combinations of markers D6S109/422/1665 (Stage 1, P=0.002 (adjusted for multiple testing); Stage 2, P=0.0001) and D6S506/1029/1660 (Stage 1, P=0.02 (adjusted), Stage 2, P=0.0001). The only two-marker association observed in both samples was with D6S422/1665 (P=0.01, 0.04). No single marker showed replicated association but D6S506 produced values of P=0.01 and 0.08 which were significant when combined (P=0.02). We observed weaker and less consistent evidence of association in a region of confirmed linkage to RD in previous studies. The most consistently significant haplotypic association D6S109/422/1665, showed association with single-word reading, spelling, phonological awareness, phonological decoding, orthographic accuracy and random automised naming, but not with vocabulary or Attention Deficit Hyperactivity Disorder. Our findings strongly support the presence of a gene contributing to RD in a region of chromosome 6 between markers D6S109 and D6S1260, but do not rule out the presence of a gene between D6S1556 and MOG.