This thesis explores preschool class children's meaning making processes when they encounter evolution. By adopting social semiotic and sociocultural perspectives on meaning making, three group-based tasks were designed. Video data from the activities were analysed using a multimodal approach. The analysis focuses on how the communicated science content affects the science focus of the tasks, how different materials function as semiotic resources and influence meaning making, and interactive aspects of doing science in the meaning-making processes. The findings reveal that, by using the provided materials and their previous experiences, the children argue for different reasons for animal diversity and evolution. Throughout the tasks, a child-centric view of life emerged in a salient manner. This means that, apart from the science focus, the children also emphasise other aspects that they find important. The child-centric perspective is suggested to be a strength that enables children to engage in science activities. The results show that the provided materials had three functions. Children use materials as resources providing meaning. This means that the children draw on the meaning potential of the materials, a process that is influenced by their previous experiences. Moreover, in interaction with peers, the materials also serve as communicative and argumentative tools. Thus, access to materials influences the children's meaning making and enables them to discuss evolution and "do science". The findings also reveal an intimate relationship between task context and interaction. More scripted tasks convey more child-adult interaction (scaffolding) while less scripted tasks, during which children build on previous experiences instead of communicated science content, stimulates child-child interaction (mutual collaboration). In scaffolding interactions, a greater emphasis is placed on the science topic of the task due to guidance from the adult. Consequently, meanings made by children in more scripted tasks are more likely to be "scientifically correct". However, if the teacher or the adult steps back and allows the children to engage in mutual collaboration, they engage in multiple ways of doing science through evaluating, observing, describing and comparing. v Overall, the research reported in this thesis suggests that task contexts and materials have a great impact on children's meaning making and how science is done.