Eoliths' were crude but purportedly humanly worked stones that exercised a great deal of scientific interest between about 1870 and 1930. They became a problem in the context of the debate surrounding the existence of pre-humans in Europe before the beginning of the Pleistocene epoch, and are now mostly reckoned to be of non-human origin. This paper addresses the way in which a network of geologists and prehistorians associated with Benjamin Harrison, the celebrated collector of the first English eoliths, attempted to make sense of barely recognizable artefacts in the period immediately following the establishment of human antiquity in the face of orthodox creationist chronologies. Harrison and his associates did so by innovating a series of criteria, names, categories and crosscutting classifications drawn from their own cultural experience, and typologies available to them through the comparative ethnography of technology. Using concepts and insights developed in cognitive anthropology, we shall attempt to shed light on a controversy in the history of science that has implications for our understanding of the way in which scientists more generally employ provisional classifications', folk categories and vernacular terminology in order to make sense of domains of intractable data at the frontiers of knowledge.