Most individuals interested in reading disability favor the view that disordered language processing is the main cause of children's reading problems and that visual problems are seldom, if ever, responsible. Nevertheless, in a preliminary study (Eden, Stein, & Wood, 1993) we showed that visuospatial and oculomotor tests can be used to differentiate children with reading disabilities from nondisabled children. In the present study we investigated a larger sample of children to see if these findings held true. Using 93 children from the Bowman Gray Learning Disability Project (mean age = 11.3 years; 54 boys, 39 girls), we compared the phonological and visuospatial abilities of nondisabled children (children whose reading at fifth grade rated a Woodcock-Johnson reading standardized score between 85 and 115), and children with reading disability (whose reading standardized score was below 85 on the Woodcock-Johnson). In addition to performing poorly on verbal tests, the children with reading disability were significantly worse than nondisabled children at many visual and eye-movement tasks. A high proportion of the variance (68%) in reading ability of both the nondisabled children and those with reading disability could be predicted by combining visual and phonological scores in a multiple regression. These results provide further support for the hypothesis that reading disability may, to some extent, result from dysfunction of the visual and oculomotor systems.