The purposes of this study are: (1) to investigate the correlations between the teacher's view on child development and preferences for early childhood programs; (2) to compare and analyze the differences on the rank order of the teacher's functions of language uses according to the teacher's view of child development; (3) to examine the differences of the teacher's verbal interaction types among teachers with different views on child development and the differences of the teacher's verbal interaction types depending on activity types; (4) to compare and analyze the differences in the move-and-response types between the teacher and the child depending on the teacher's views on child development. For the correlation between the teacher's view on the child development and the program preferences, 62 teachers from 18 kindergarten or Saemaul day care centers are consulted. Five teachers are selected for each of the three views of child development : behavioral-environmental view, normative- maturational view, and cognitive-interactional view. EPIE(1972) checklists are used. The teacher-child verbal interaction styles are analyzed with respect to five linguistic activities, 27 linguistic functions, topics, and domains of discourses, as suggested by Wood(1980). The move-and-response types between the teacher and child are also analyzed according to Wood(1980). For the measurements of the analysis and results, Pearson r, Kendall's Test of Concordance, three-way analysis of variance with repeated measures on the last two factors, and one way analysis of variance were used. The results are as follows: 1. The Seoul area preschool teachers' views on child development are predominantly cognitive-interactional. 2. There are significantly high correlations between the preschool teacher's view on child development and preferred program types. 3. There are no significant differences in the rank ordering of the functions of language uses of preschool teachers according to the teacher's view on child development. Such language functions as Asks for Description, Directs, and Instructs, that is, Insturction or Managements are used most frequently. Most of the teachers are using verbal interaction types of the behavioral approach. Therefore, we can conclude that there is no significant relationship between the teacher's view on child development and the verbal interaction types used. 4. There are significant differences in the teacher's uses of verbal functions according to daily activity types. Verbal interactions for Instruction are significantly high during the snack, work, and group discussion time. During the snack and work time, verbal interactions for management are also significant. And verbal interactions for rapport as well as management are significant during the group discussion time. 5. There are no significant differences in topics of discourse due to the teacher's view on child development. Talking about the child is significant in the teacher-child verbal interactions. During group discussions and work time, talking about objects and possessions is also significant. 6. The most significantly frequent domains of discourses are here-and-now. During the group discussion time, timeless talks are also significantly frequent. The domains of discourses are significantly related to the topics of the discourses, as Bartlett (1981) claimed. 7. The teachers who have the behavior-environmental view on the child development use closed questions significantly more than teachers with other views on child development. And children in a classroom with a teacher using closed questions frequently also use closed questions more frequently. The teachers with cognitive-interactional view on child development make significantly more contributional responses in discourse. Other response types showed no significant differences among teachers with different views on child development.