This paper will report on the development of a research program by a group of science educators at Kristianstad University, which has its roots in a longitudinal study I conducted concerning students' developing understandings of ecological processes. Following the insights generated in this first study concerning the nature of student understandings, and the potential of the longitudinal design, a research group has developed at Kristianstad which has extended this work into related areas. This paper will describe my own work and its development, and link it with three projects that use a longitudinal design, which we have subsequently undertaken and in some cases completed. The emphasis within the paper will be on the findings generated by these studies on student learning and response to science, and the particular features of the longitudinal design that allow such insights to emerge. The paper will explore patterns of change, and continuity, as a way of appreciating the particular contributions of longitudinal studies.Most of us who taught scientific subjects in teacher education at Kristianstad University during the late 1980s and early 1990s came to teacher education after several years of teaching science at the lower or upper secondary school level, to students of age 13-19 years. Consistent with research findings in the science education literature, we frequently experienced students' difficulties in understanding and describing scientific phenomena, and their use of alternative ideas when explaining scientific concepts and processes. Studies of such conceptions form a substantial portion of the science education research literature.I found that I could identify common features in students' conceptions at different levels, and also signs of patterns of development in their descriptions of the phenomena. I asked myself, as a teacher and researcher: From where do these ideas and attitudes come, and how do they develop? If you identify a student's conception at one point in time, it might be helpful to know more about the roots of such conceptions in order to improve teaching at an early age. It might also be interesting to know how the student could develop when she/he gets older. A longitudinal research design makes it possible to gain insights into students' individual learning pathways by following their developing understandings year by year. Such insights could then be used to develop teaching strategies in order to more effectively support development of students' understanding. The insights can also be used in teacher education. We ask student teachers to identify alternative explanations of scientific phenomena as