An experiential condition for creative aspects of morphemic use was explored using a miniature linguistic system. Both numbers of exemplars presented during the study trial and exemplars tested during the test trial were controlled. The major findings were that 1) a small number of experienced exemplars was insufficient for novel situations to be described, and 2) the number of exemplars (20 out of 32) was a critical variable which made a criterion of adequate creativity possible. The results were interpreted to suggest that a sufficient number of different exemplars should be involved in experiencing linguistic data, and were discussed in terms of a theory emphasizing that language acquisition is based on the interaction between linguistic and nonlinguistic semantic aspects of experienced context.