The study found that the overrepresentation of learning disorders among criminals appeared to be a significant phenomenon. There were no differences among sex offenders and controls in overall IQ or in the percentage of mentally retarded (MR) or learning-disordered (LD) cases, suggesting that the learning difficulties are not peculiar to sex offenders. The results indicated that there was a bias in referral source, with more mentally retarded, borderline-retarded, and/or learning-disordered cases being referred by the Children's Aid Society, prisons, and the Crown. This suggests that referral sources may play a significant role in evaluating intelligence and mental retardation among sex offenders, but the overrepresentation of learning disorders among criminals appears to be a significant phenomenon, regardless of referral source. The sex offenders were generally of average intelligence, and the mentally retarded were not overrepresented among them, however, the learning disordered were overrepresented. A sample from a forensic database of 2,286 male sex offenders and paraphilics and 241 nonsex offenders was evaluated for the prevalence of mental retardation and learning disorders, using the full Wechsler IQ scales and diagnosis of MR and/or LD. These persons were seen for assessment between 1966 and 2005 at a university psychiatric hospital and private clinic, and grouped on basis of known criminal history. Tables, references