A detailed explicit definition of learning disability (LD) is presented in terms of the diagnostic procedures and decision points used in identifying that category among a group of juvenile delinquent and control group adolescent males. The described procedures include a review of existing records, interviews, and diagnostic assessments. Rationales for the choice of pertinent variables and measurement instruments are given. The assessment process and administrative arrangements are explained. The factor analytic studies of Witkin and his colleagues were deemed especially relevant and the Wechsler Intelligence Scale for Children (Revised) was used. Decision points eliminated boys who were mentally retarded (intelligence quotient below 75), and boys who were, or had been, diagnosed as emotionally disturbed (those diagnosed as untestable or presenting bizarre symptoms). Those chosen for further diagnosis included boys whose achievement score differed by two or more years from their ability score, or boys whose academic achievement was two or more years retarded. Interviews and diagnostic testing were carefully standardized and required about three hours for each boy. Training, quality control, and other administrative procedures are described. (CTM)