This paper considers some of the varied practices that went into making populations into legible biological entities in the 1950s, work that was not simply self-evidently ‘biological’. Similar to other recent historical studies, I describe how the construction of blood groups as devices for producing human biological difference relied on notions of kinship, racial and national identity, multiple historical traditions, and questions and practices imported from geography and anthropology. But it is my argument that, this recourse to ‘cultural’ knowledge was not disingenuous or naïve, as some have implied; rather, this made blood-group genetics distinctively ‘human’. The paper concerns the work of Arthur Mourant, avid collected of blood groups and prolific author of what he called ‘blood group anthropology’. By calibrating his blood-group frequency data researchers like Mourant sought to give it the potential to reveal new knowledge about human difference and identity, even if the production of that knowledge was constantly deferred. By focusing on blood groups I outline an account of the way in which human difference was made genetic, and about how genetics was made human.