This study was designed to replicate and extend the work of Steffe on the effect of using different set names as variables in addition problem statements. The behavior of both first- and second-grade students was studied and the usefulness of visual aids in solving addition problems was also investigated. One-hundred twenty first- and second-grade students were selected from a school; a random half of these were assigned to the visual aids treatment while the remainder used no visual aids. Steffe's instrument, containing two item types, was used; items of one type used one set name (e.g., 5 cars + 2 cars = ? cars) while others used three set names (e.g., 4 cats + 2 goldfish = ? pets). Data on these story problems were submitted to analysis of variance; differences between means were subjected to t-tests. Results indicated that first-grade students had a difficult time solving addition story problems involving three set names. Second-grade students had more difficulty with problems involving three set names than one set name when visual aids were used, but not in the absence of these pictures. In general, second-grade students were more able than first-grade students to solve the problems. (Author/SD)