To determine whether increased teacher use of student ideas would produce increased verbal initiation, an experiment was conducted with 16 second grade teachers and their classes. First, a series of observations were made under normal class conditions to determine the usual percentages of teacher use of student ideas and student initiation. Then, during a 20- to 25-minute time period, each teacher attempted to elicit student initiation on 2 topics selected by the experimenter. Finally, each teacher again attempted to elicit student initiation on 2 more topics. Each teacher and her class was observed using Flanders' Interaction Analysis categories to code teacher and student behavior, and a statistical analysis was made of the results. It was concluded that those teachers who increased their use of student ideas increased student initiation, whereas teachers who did not increase their use of student ideas did not obtain more student initiation. (A 6-item bibliography and a list of categories for interaction analysis are included.) (SG)