The methodology of this research tried to span the small scale and the large, the qualitative and the quantitative, present knowledge and proximal development. It took into account the results of previous small scale questionnaire studies, interviews ofgroups of students and action research in the classroom. These informed the results ofa large-scale questionnaire study of the nature of science.Few questions were asked which referred to knowledge acquired from two different domains: a. out-of-school images of science and scientist; b. ideas drawn from their own school experiences of science in the classroom and the laboratory. Knowing the strong effect of context we were anxious to keep the questions as general as possible. They were to probe the connection between theory and experiment in terms which, we had found, students ofage 15 years, could understand. The results showed a strikingly significant relation between the class teacher and the responses to most questions. The exceptions were in out-ofschool knowledge and confirmed our hypothesis about the two origins of students' knowledge about the nature of science.Looking for significant correlations between the students' answers to different questions revealed the presence of two interesting groups of students: the 'Explainers' and the 'Imaginers'. The first of these seemed to be reflective, and as having a more explanatory perception of science. However this was limited to their own laboratory experiences and did not extend to other cultures, or to the use of imaginative mental models. The second group -the 'imaginers' -were far fewer in the large sample of 15 year olds, but comparatively more numerous in a small sub-group ofolder students at age 17. Separating out this disappointingly small group of pupils we found that they were more interested in what goes on in the minds of scientists.

This paper is derived from data gathered during a year-long classroom study on how pupils' views of the nature of science changed when some of their learning materials were historically situated. Five classes were involved and the evidence comprised pre-and post-tests, interviews and class tests.It is shown that life-world images of scientists as people, and the equipment with which each is associated, can be taken to represent pupils' epistemologies. This is related to contemporary thinking about the nature of knowledge and of technology.The questionnaire responses showed substantial progress during the year. The older ideas were not eliminated but stories from history seemed to have provided alternative images which generated more reflection. However, the results were complicated by the conflicting meanings for 'explain' held by many of the pupils. Some supplementary work carried out during the course explored the classroom methods by which teachers emphasized the causative rather than descriptive nature of explanation, and the pupils' ease of learning this.

No abstract