A question of both theoretical and practical importance for the study of phonological development is whether there is a difference in the status of productions rendered spontaneously by the child and those repeated by the child after either an adult model or his own production. The relevant theoretical questions are: (1) Are all the child's productions mediated by his current phonological system? (2) What is the role of the adult model in phonological acquisition? The spontaneous and repeated utterances of four children were examined, and it was found that there were differences between the spontaneous and echoic forms produced. The child's productions immediately following those of the adult did not necessarily approach the model more closely than did spontaneous productions. In certain cases a "trade-off" occurred, i.e., a sound segment in an echoic form more closely approached the model while another sound segment fell short of the target. This suggests that echoic and other repeated forms are not "phonologically progressive," as had been proposed. (Author/CFM)