This monograph presents 11 papers on a study of the problem of specific diagnosis with children who do not communicate effectively. About 500 children with inadequate communication skills, selected as having a developmental language disorder, autism, or low IQ without autistic features, were studied by a multidisciplinary team. Data collected included family, developmental, and medical histories; parent/teacher questionnaires regarding the children's behavior and sociability; results of a neuropsychological and language battery; and analysis of spontaneous language and symbolic play skills. Behavioral and play measures proved most efficient for separating autistic from non-autistic children, whereas few historic or neurological features differentiated the groups. Titles and authors of individual chapters are: (1) "Introduction to the Study" (R. Morris and I. Rapin); (2) "Classification of Developmental Language Disorders" (N. Hall and D. M. Aram); (3) "Classification of Autistic Disorders" (L. Waterhouse); (4) "Methodology (Subjects and Measures)" (R. Morris and others); (5) "Historical Data" (I. Rapin); (6) "Neurological Examination" (I. Rapin); (7) "Neuropsychological and Language Data" (D. Fein and others); (8) "Behavioral Data" (L. Waterhouse and D. A. Allen); (9) "Play" (L. Wainwright and D. Fein); (10) "Classification Issues" (I. Rapin and others); and (11) "Overview and Conclusions" (D. Fein and others). The measures used are appended, including the first publication of two new assessment instruments: the Wing Schedule of Handicaps, Behaviour and Skills (HBS) and the Wing Autistic Disorder Interview Checklist (WADIC), which was the single best discriminator in the study. (Contains approximately 500 references.) (DB)