A follow-up study was performed to investigate the stability of IQ measures in a group of dyslexic teenagers and young adults. Earlier research had shown contradictory results. The 65 subjects, 12 years old on the average at first test, were retested after a mean interval of six and a half years. There was a significant relative decrease in verbal IQ (VIQ), which was interpreted as either an effect of low reliability of tests used, or an effect of the dyslexic individuals' less experience with reading and writing, and as a consequence, a lag in verbal ability, the second interpretation being in line with earlier findings in groups of children with learning disabilities. Performance IQ improved significantly and the tentative interpretation was that of a compensatory process, in the sense that the dyslexic children might develop a more visual, intuitive and creative way to process information and solve problems. The conclusion was that caution should be taken, before making important decisions on the basis of a single intelligence test, and that dyslexic children might be at risk to lag behind their peers in terms of VIQ, especially if they are not provided with suitable special education. (Contains 7 tables.)