In this paper we combine the findings from two recent studies relating to participation and attainment in school science --a re-analysis of existing official data for England and a review of wider international research evidence in the literature relevant to the UK. Although the secondary data are drawn mainly from England, the comprehensiveness of these datasets, together with our inclusion of a review of international studies on maths and science participation provides a useful reference point for an international audience. The research was prompted by concerns over a reduction in the uptake of the physical sciences post-16 and especially in higher education and interest in ways of encouraging the study of science by students from less prestigious socio-economic status backgrounds. Such concerns are not unique to the UK. Using large-scale official datasets we show that participation and attainment in science are stratified by socio-economic status. Students from poorer families are less likely to take sciences at post-16 than many other subjects and those who do are then less likely to obtain grades high enough to encourage further study of the subject. No conclusive evidence has been found to explain this satisfactorily. Plausible reasons suggested in the literature include the relative scarcity of local opportunities putting off those who do not wish to study away from home or the perceived time demands of studying science, and so the difficulties of combining part-time study and part-time work for those needing to continue earning while studying. Direct support from professional parents may also lead to greater participation in post-16 science for students from higher SES. Perhaps the simplest explanation is that participation in science at any level is often predicated upon success at the previous educational stage. There are clear differences in science attainment at age 16 between students of differing backgrounds, which could explain the subsequent differential participation. However, these differences are not dissimilar to those for all subjects. The largest gap presented in the paper is between students eligible and not eligible for free school meals. We also show that these patterns appear early in the life of children. At ages 7 and 11, attainment in the three core subjects (English, maths and science) is negatively related to living in an area of deprivation. The paper ends with a discussion of suggestions for research, policy and practice emerging from this review of the evidence. (Contains 4 notes and 12 tables.)