Tn this study pupils in grades 1 through 9 (ages 7-16) of the Swedish comprehensive school rated their interests in mathematics and Swedish (separately for reading and writing). They also judged other aspects of their school work and situation, such as need for control and challenging tasks, their perception of the teacher, their concern with how well other pupils succeeded, and their well-being at school. Teachers rated their level of achievement in Swedish and mathematics. It was found that interests tended to decrease over time, especially in mathematics. High achievers were more interested but they as well as low achievers showed a drop in interest over the years at school. Need for control rose and general well-being dropped. The developmental trends of the interest variables could not be accounted for by a rising need for control. The decrements in challenge and well-being could explain a part of these trends in Swedish but not in mathematics.Cognitive psychology dominates present applications of psychology to educational problems, in spite of widespread concern about lack of student motivation, especially a lack of interest, topics usually ignored in the cognitive tradition. There are several reasons for this state of affairs, the most important probably being that interest has never been a fashionable topic in psychological research. Currently it surfaces in psychology most often in the form of «intrinsic motivation»! and it is closely tied to experimental studies of the effects of rewards on experimental tasks that carry a high level of interest in themselves. This is a tradition that is little helpful in the discussion of real-world educational problems, as we shall point out in some detail in the discussion.In our own work we conceive of interest as an emotion (Izard, 1977) and we see it as a driving force of critical importance in real-world educational settings. In previous studies at the secondary school level, interest has been related to subject matter attributes (Sjoberg, This study was partly suppoTled by a grant from the Office of the Chancellor of the Swedish Universities. We wish to thank several people in Boras who were helpful in making the study possible, viz. Chief Education Officer Hans Rosendahl and the School Managers. We are also grateful to the pupils who participated, their teachers and parents, for their support of the study. Finally, we are much grateful to Bengt Jansson for his help with data analysis, and likewise to Bjorn Areskoug of the Gothenburg Computer Center. Henry Montgomery gave valuable comments on the manuscript.