This article describes the technique of human embryo selection in two laboratories doing in vitro fertilisation in India and France. It is based on participant observations and a comparative ethnography. It shows how reproductive biology utilises naturalism as a global scientific approach to cells, and how biologists interact with patients who seek to realise a parental project. In the French laboratory two definitions of embryos, either as bunches of cells or as potential children, are critically related in a context where the scientific vocabulary is said to be limited or detrimental when communicating with patients. By contrast, the Indian laboratory does not debate the status of embryos, but instead communicates “facts” and scientific “truth” to patients in a context where fraudulent exchanges of gametes can occur. With reference to the notion of the “operational sequence,” this comparative ethnography characterises selection as an operational relation, through which local contextualisations determine the meaning and evaluation of scientific naturalism.‪