Autism is the most concrete developmental disorder. Upon recent epidemiological studies, the prevalence of autism is about one in every 165 children. Only 25% of them have intellectual disabilities. So, it was thought that the disorder is not necessarily with severe cognitive damages. While the poor performance of children's communication is the main features of them, language profile of symptoms varies from person to person. On the one hand, children have ability to perform their language as the norm, and on the other hand, children who do not have the ability to speak. So, Language impairment is one of the main characteristics of children with autism; however, findings of the language deficits in these children have been inconclusive, and even less is known about the language profiles of Persian children with autism. Due to the heterogeneity in autism, we also need extensive investigations on the properties of their language in different languages. The aim of this study was to evaluate grammatical characteristics in children with autism by revealing the differences between age matched and language-matched normal children.Rizzi believes each structure will consist of three layers from top to bottom as follows: The complementizer phrase (CP), inflectional phrase (IP) and verbal phrase (VP). So, the Complement layer is syntactic materials to build question phrases, complement phrases and relative clauses. Inflectional layer is related to inflection, negative and auxiliary verbs while verb phrases are related to verb and its arguments.Research data were collected at 10 autistic (ages 6 to 9) and 20 normal children (10 age-matched and 10 language-matched). First, we used TheGARS-2, and the ASSQ for diagnosing high-functional Autistic children. TheGARS-2 is a norm referenced instrument that assists teachers and clinicians in identifying and diagnosing autism individuals aged 3 years to 22 years and in estimating the severity of the child's disorder. The ASSQ is a questionnaire which was filled out by the parents or teachers of children or adolescents (6 to 17 years of age). It stands for screening autism spectrum disorders. Children’s expressive GRAMMAR was measured by using the Persian Photographic GRAMMAR Expression Test. It is the first reliable and valid test that exclusively and accurately evaluates grammatical characteristics of Persian-speaking children. It concludes 32 grammatical structures it has good content validity (CVI>80), convergent validity of two test (r P=0.5) and a correlation coefficient equal to 0.91 and greater than 0.9 respectively. Also, the test has a good internal consistency (kr21=0.82). And finally data were analyzed with Mann-Whitney test.The result of this research showed that there is significant difference between autistic group and age-matched group in all structure except in WH question, yes/no question, different type of relative clauses, tense, aspect, mood, different type of pronouns and preposition. But comparison between autistic group and language-matched group showed that there is no significant difference in most structures except WH question, yes/no question, complementary relative clauses, different type of pronouns and preposition. Overall, autistic children demonstrated consistent expressive impairment in questions, relative clause and pronouns. It should be noted that these factors should be seriously considered in identification and treatment of this phenomenon.