This paper reports on an experiment undertaken to delineate more clearly the relationship between the naming process in children and certain aspects of the environment which may play a role in that process. The investigation concerned the effect of manipulation and of object novelty on naming. Sixteen children, ten girls and six boys, ranging in age from 16 to 23 months, were tested in their homes. The objects used in the experiment were selected so as to vary along the dimension of novelty. The children were shown objects familiar to them as category items, objects less familiar but still members of familiar conditions: the manipulation (M) condition in which the child was shown the object as it was named, and the non-manipulation condition (NM) in which the child did not handle the object. In each condition, the experimenter repeated the name of the object three times during the conversation with the child. Presentation conditions were alternated for each object continuing until all objects had been presented. It was found that manipulation does not have any significant effect on naming but that familiarity is a strong factor. A possible explanation is that familiar objects were more often named since the child did not also have to engage in concept formation and thus could direct his energies to naming. (AMH)