Heretofore most studies dealing with the association between perinatal complications and autism have used a normal comparison group. In this study obstetrical records of 59 autistic children were compared to those of 28 nonautistic children whose intelligence has a similar range and distribution as the autistic sample. Using an optimality score to reflect number of obstetrical complications, we found that the nonautistic controls experienced less optimal conditions than the autistic sample. Abnormal presentation at birth is the only factor that occurred more frequently for the autistic sample than control sample.