This paper is based on my M.Sc. dissertation: ‘The Origins of Numerical Taxonomy’ 1985, submitted to the University of Leicester during the tenure of a D.E.S. State Studentship. For this work I drew extensively on interviews with Professors A. J. Cain, G. A. Harrison, R. R. Sokal and P. H. A. Sneath. I am very grateful to them for their time, interest and encouragement. Without the indefatigable assistance of Jon Harwood, this paper would never have been finished.

The word ``deme'' was coined by the botanists J.S.L. Gilmour and J.W.Gregor in 1939, following the pattern of J.S. Huxley's ``cline''. Its purposewas not only to rationalize the plethora of terms describing chromosomaland genetic variation, but also to reduce hostility between traditionaltaxonomists and researchers on evolution, who sometimes scorned eachother's understanding of species. A multi-layered system of compoundterms based on deme was published by Gilmour and J. Heslop-Harrison in1954 but not widely used. Deme was adopted with a modified meaning byzoologists (...) leading the evolutionary synthesis – Huxley, Simpson, Wright,and Mayr. Connections are shown between Gilmour's ideas around definingthe deme, his role in founding the Systematics Association, and his chapter``Taxonomy and Philosophy'' in the book The New Systematics. Thishistorical episode raises questions about the role of carefully-definedwords in scientific practice. (shrink)