Differentiating autism from other handicapping conditions, especially mental retardation, has been a constant problem for public schools. This study investigated the effectiveness of three instruments to discriminate autistic from trainable mentally retarded children. The Autism Screening Instrument for Educational Planning, the Childhood Autism Rating Scale, and the Diagnostic Checklist for Behavior Disturbed Children, Form E-2 were administered to 20 autistic and 20 TMR students. Discriminant analysis was used to determine the best linear combination of scores that would separate the two groups of children. All three instruments were found to separate the two samples of children. However, the CARS and the ASIEP provided for a greater separation of groups.