Theoretical and correlational considerations suggest that prelinguistic communication lays the foundation for several aspects of later linguistic communication and language development. Mentally retarded children frequently show language levels substantially below that expected for their mental ages. This unexpected language delay may be due, in part, to deficits in prelinguistic communication development. the primary purpose of the research is to investigate the empirical basis for inferring a causal relationship between prelinguistic and linguistic communication in mentally retarded children. To do so, we will determine whether teaching prelinguistic communication skills to mentally retarded children influences their parents' linguistic input to their children and the children's later developing linguistic development. At the pretreatment period, children's means-end relations, babbling and vocal communication complexity, and the parents' responsivity and HOME scores will be taken as potential predictors of which dyads benefit most from the intervention. The subjects will be mildly to moderately retarded children who demonstrate emerging intentional communication (approximate developmental age 9 months) but do not yet use spoken or signed words to communicate (approximate developmental age 12 months). Given the narrow selection criteria, three waves of subject recruitment with 18, 21, and 21 children (N = 60) are proposed. Children will be randomly assigned to a treatment or control group. Children in the treatment group will receive an intervention designed to facilitate prelinguistic development six times/week for 6 months or until they reach mastery criteria. Children in the control group will participate in a play session of equal length to the intervention sessions to control for nonintervention explanations to between-group differences. In addition to the pretreatment measures already mentioned, prelinguistic communication, parental linguistic input, and/or linguistic communication and language development will be assessed at five periods; pretreatment, posttreatment, and 6, 12, and 18 months after the posttreatment assessment. Statistical analyses involved in testing the research questions will be MANOVA, repeated measures ANOVA, MANCOVA, multivariate multiple regression, multiple regression, and the Johnson-Neyman technique.