Two experiments were designed to investigate the effect of tracing and copying practice methods on handwriting skills of Japanese letters in young children. In Exp. I, 106 first grade children were randomly divided into three groups ; tracing, copying, and control groups. Only two experimental groups were required to trace or copy 4 katakana letters ( o , ta , ya , mo ) 5 times in a session for five days. In Exp. II, almost the same procedure was conducted for 77 preschool children (4 and 5 years old) in order to retest the result of Exp. I. The result of both experiments were the same, and for poor handwriters, copying practice was significantly more effective than tracing practice, but for good handwriter, there was no difference among the groups. It was suggested that copying practice should be better than tracing practice for handwriting beginners.