This is a longitudinal study on the development of basic emotions of school children and adolescents. The subjects were 212 children of the third grade, 216 children of the sixth grade and 231 adolescents of the ninth grade. All the subjects were given the emotional questionnaire twice, the first time two years ago and the second time this year. They were sampled as to their self-expressed happiness, anger, sadness, fear and worry and their reactions to these emotions. Comparing the responses of the subjects to both measurements, three groups of longitudinal data were collected. The results indicated that the developmental trends were more evident on the emotion of happiness, anger and fear than sadness and worry. They also varried according to the sex of the subject. 　　The first group of longitudinal data were based on the response changes of the subjects of the third grade between this year and two years ago. The results indicated that the joy derived from recreational activities tended to increase with ago. The joy derived from holiday celebrations tended to decrease with ago. Personal restrictions and intrusions as the resources of anger tended to increase with age. The anger due to other people's behavior tended to decrease with age. The fears of being left in the dark and the contents of TV program tended to increase with age for girls. No change was found as far as the sadness and worry were concerned. As regarding to their reactions to these emotions, the major pattern was affective response, but more positive responses were found while they were getting older. 　　The second group of longitudinal data were derived from the response changs of the subjects of the sixth grade between this year and two years ago. The results showed a different pattern of development. Girls seemed to think school and personal achievement as the happiest experience in life as well as receiving gifts. The joy of sport, and recreational activities tended to decrease with age. The fears of natural phenomena, animals and unhappy experiences tened to decrease with age but the oppsite was true for the fear of school life. No developmental changes were found as regarding to the emotion of sadness and worry. More positive pattern of emotional responses were found. 　　The third group of longitudinal data were derived from the response changes of the subjects of the ninth grade between this year and two years ago. The results showed that boys mentioned less frequently about sport, school and personal achievement and interpersonal relationship as their happy experiences in life. The opposite tendencies were found for girls. The anger of losing a personal belonging, frustration derived from school work, the behavior of other people and personal restrictions tended to increase with age. The fears of school life, being left alone in the dark, mass media and unhappy experiences tended to decrease with age for boys. Girl's feeling of sadness on losing a friend, personal belonging, death of the family and personal defecit tended to increase with age. The major pattern of emotional response of adolescents became more self-controlled and more indirect as they were getting older. 　　The third group of longitudinal data were derived from the response changes of the subjects of the ninth grade between this year and two years ago. The results showed that boys mentioned less frequently about sport, school and personal achievement and interpersonal relationship as their happy experiences in life. The opposite tendencies were found for girls. The anger of losing a personal belonging, frustration derived from school work, the behavior of other people and personal restrictions tended to increase with age. The fears of school life, being left alone in the dark, mass media and unhappy experiences tended to decrease with age for boys. Girl's feeling of sadness on losing a friend, personal belonging, death of the family and personal defecit tended to increase with age. The major pattern of emotional response of adolescents became more self-controlled and more indirect as they were getting older. 　　Generally speaking, the developmental trends derived from longitudinal data were mostly congruent with the results collected by cross-sectional method.