This study lies within the constructivist tradition in science education which holds that the learner's prior knowledge is a key factor in concept development. The topic was photosynthesis, which was known to be difficult for students although little was known of their prior understandings about plant activities. A three phase research programme was designed to arrive at a description of these pre-teach views (Part 1) and to use this knowledge in an action research programme which modified children's existing knowledge more effectively than current classroom practice (Part 2), Phase I comprised interviews with 28 pupils (aged 8 to 17 years). This lead to Phase II in which nearly 6000 pencil-and-paper survey responses were obtained from students from standard four (10- year-olds) to first year university (18+ years). The interviews and surveys showed that pre-teach children had a number of separate views about plant drinking, plant breathing, plant growth, how plants acquire energy, and plant feeding (these were collectively identified as views about 'plant nutrition'), but the children did not possess knowledge directly comparable with photosynthesis. As with earlier scientists' explanations, analogies and metaphors relating plants to animals were important in children's understandings. In contrast with what is known about children's prior knowledge generally, their views about plant nutrition were held with varying degrees of conviction and on some issues no views were evident. Phase III approached classroom action research by evaluating three existing strategies. Each of these (the guided discovery, element analysis, and trophic conflict strategies) was found to be deficient because children's prior knowledge was not considered and/or the scientists' view was not adequately addressed (i.e. food-making was considered at the expense of carbohydrate production or energy storage). A new strategy which explored the material aspects of photosynthesis (carbohydrate production)and which was based on the generative learning model was developed. The new strategy resulted in a teaching package entitled "Where Does The Wood Come From?" which was trialled by an experienced and sympathetic teacher with a middle ability class of 26 fourth formers (14-year-olds). The class was observed throughout the four weeks of teaching, and individual students were also interviewed out of class at key times. After this unit, 71% of the students had acquired a view of photosynthesis as a carbohydrate-producing process. This contrasted with the usual guided discovery strategy, where a foodmaking view was the major outcome. Important observations were that the students perceived the unit as non-threatening, and that they underestimated the importance of their own ideas. The study also documented some practical applications of the generative learning model (and constructivist theory generally) in the classroom. Investigations 1 surveys, a self-teach booklet, and a series of checkpoints were developed, and these features of the teaching package may have wider application. Also, it was suggested that the apparent similarities between children's views and those of earlier scientists may be used to facilitate classroom discussion and expose ideas. The findings of the action research resulted in suggested modifications to the generative learning model itself, especially its apparently sequential nature.