To assess head circumference in children with autism, 148 charts were retrospectively reviewed. All of the children met the Diagnostic and Statistical Manual of Mental Disorders (DSM-III or DSM-III-R) criteria for autism and had no known underlying condition that might affect head circumference. In addition, data were collected regarding height, weight, brain imaging, cognitive development, adaptive behavior, and language. The children were divided into two groups: those with head circumference at or above the 98th percentile (Group 1) and those with head circumference below the 98th percentile (Group 2). Group 1 consisted of 27 (18.2%) of the children. Height measurements were significantly higher in Group 1 as compared with Group 2 (P = .0006) as were weight measurements (P = .0003). Group 1 had a significantly lower percentage of females (P = .04) and lower adaptive behavior scores (P = .0067) than Group 2. Routine brain imaging studies could not explain the macrocephaly in Group 1. The etiology of large head circumference and increased growth indices in children with autism is unclear.