Kindergarten curricula (Bahovec et al. 1999, 37) cover different areas of education that are sensibly interconnected and integrated. Science is one of the areas that represent child’s first learning about the surrounding world and the first introduction to nature. Science education is a very suitable approach of introducing children into basic research work, since in nature objects and phenomena are more concrete, and hence the children are instinctively attracted. Therefore, science education could act as a starting point for all other areas of education. Teaching science comprises several difficulties, which are mainly connected with the way how contents are introduced to children. This is often a great challenge and great responsibility for pre-school teachers, because the scientific content needs to be explained in an appropriate way taking into consideration the child's age, the use of correct terminology, and at the same time avoiding inadequate generalization and over-simplification. Buoyancy is a natural phenomenon that is experienced by every child, but which, on the other hand, is quite difficult to explain. With the present study we wished to assess the knowledge considering buoyancy of the part-time students within the Pre-school Teaching educational programme at the Faculty of Education of the University of Primorska, which is performed on different locations throughout Slovenia. These students already teach in the kindergartens and should be well acquainted with buoyancy from previously passed physics courses at the Faculty of Education. We examined how they explained the buoyancy to children in kindergartens, and whether the knowledge about buoyancy is affected by their working experience or the location of their study. The results show that the students’ knowledge about buoyancy is insufficient and incomplete. In addition, many misconceptions about buoyancy are transmitted to the children in the process of teaching. Furthermore, it can be stated that neither the working experience nor the location of the study affects the interpretation of buoyancy (p<0.05).