Low participation rates in the study of science, technology, engineering, and mathematics (STEM) post-16 are a matter of international concern. Existing evidence suggests children's science aspirations are largely formed within the critical 10 to 14 age period. This article reports on survey data from over 9,000 elementary school children in England (age 10/11) and qualitative data from 160 semi-structured interviews (92 children aged 10/11 and 78 parents), collected as part of an ongoing 5-year longitudinal study in the United Kingdom tracking children from 10 to 14. Drawing on the conceptual framework of Bourdieu, the article explores how the interplay of family habitus and capital can make science aspirations more "thinkable" for some (notably middle-class) children than others. It is argued that while family habitus is not deterministic (there is no straightforward alignment between family habitus, capital, and a child's science aspirations), social inequalities in the distribution of capital and differentially classed family habitus combine to produce uneven (classed, racialized) patterns in children's science aspirations and potential future participation. (Contains 4 notes and 1 table.)