This report describes three experiments designed to measure the changes in children's social speech produced by features in the physical environment that are designed to alter verbal communication contingencies between players. All experiments utilized A-B-A-B withdrawal designs. Both conditions consisted of free play at an ordinary open table, but, in the experimental condition, partitions between players allowed only partial visual contact. In the first experiment, the partitions produced sharp drops in the amount of speech referenced to playthings but no changes in the level of specificity. In the second experiment, no changes in verbal referents were produced, but significant increases in specificity of speech occurred when the partitions were in place. Similar results were obtained in Experiment 3. The results are considered as confirmation of the importance of physical environmental factors in the regulation of children's play, with differential results attributed primarily to differential reinforcement of specific and nonspecific speech when the partitions were in place.