Previous research has shown that the children's use of the fingers in additon changes with age. In this study, a part of data on the strategy of finger use by Asakawa and Sugimura (2009) was reanalyzed to clarify the relationship between, short-term memory, finger dexterity and addition skills. A two-way ANOVA showed a significant interaction between memory span and finger use. Examination of simple main effect indicated that significant effect of memory span at the group of the children who did not use fingers and significant effect of finger use at the group of the children whose memory span were small. However, there was no significant interaction between finger dexterity and finger use. Results suggest that finger use can compensate for small memory span.