A clinical interview method was used to investigate the addition procedures used by first- and second-grade children on both written and orally presented problems of different magnitudes and to determine whether these children vary their strategies according to the nature and magnitude of the problem. The results showed that first graders used a variety of counting procedures to solve addition problems, while second graders used both counting and noncounting procedures. In general, second-grade children efficiently adjusted their strategies according to the magnitude of the problem's addends. Implications for education are discussed.