A project was undertaken to develop and evaluate the effectiveness of a unit of instruction which would teach economically deprived first grade children some interesting scientific concepts and increase their curiosity behavior and willingness to offer hypotheses, carry out experiments, and evaluate outcomes. The instructional unit was based on the concept that all matter living and non-living, is composed of molecules which can undergo change. The unit was presented in two parts. In Part I, six 50-60 minute lessons were presented to an intact class over a two-week period. Comparisons between this experimental group and an uninstructed control class revealed that over the relatively short instructional period the experimental group did significantly better (p.01) on the subject matter test than the control group. On the Banta Curiosity Box Test, both the pretested and non-pretested experimental children and the pretested control children scored significantly higher than the non-pretested control children. The results to the other individual measures developed to assess curiosity behavior were ambiguous, primarily due to problems with the measures themselves. Modification and extension of the instructional program and the evaluation instruments were undertaken in Part II. One month after the completion of Part I, the same experimental class was presented five 30-minute lessons in a seven-week period. Experimental children continued to perform better (p.005) than their control counterparts under the revised instructional program. Results of Parts I and II point up the importance of the further development of instructional units designed to improve young children's scientific inquiry processes. (Author/CK)