Royal Society Publishing

Peter Henry Andrews Sneath. 17 November 1923 — 9 September 2011

Dorothy Jones, William D. Grant

Abstract

Peter Sneath revolutionized the practice of bacterial taxonomy over a 30-year period. He was the first to apply Adansonian principles to bacterial taxonomy, arguing that a robust system required that bacteria should be subjected to many phenotypic (phenetic) tests, all given equal weight, with analysis of the subsequent binary test scores being used to derive groups that could be given taxonomic rank (taxa). The procedures came to be known collectively as ‘numerical taxonomy’. A further innovation was the realization that the then embryonic discipline of computing could be harnessed to derive taxa from these very large sets of data. Computer-aided numerical taxonomy became the method of choice for classifying bacteria by the early 1960s. Much of Peter’s effort as Director of a Medical Research Council (MRC) Research Unit in Leicester in the 1960s and early 1970s was to reassess the taxonomy of most of the medically important bacteria. This information was then interrogated to determine the minimum number of tests required to identify new isolates reliably. The tests available in commercial identification kits in use today directly reflect these original numerical analyses. Later, after appointment to the Foundation Chair of Medical Microbiology at the University of Leicester, he carried out, together with colleagues, the most important revision of bacterial nomenclature for more than a century, the 1980 ‘Approved lists of bacterial names’. He was also a member, Vice Chairman and then Chairman of Bergey’s Trust, the organization responsible for Bergey’s manual of determinative bacteriology (later Bergey’s manual of systematic bacteriology), the definitive account of bacterial taxonomy and properties. He continued to edit volumes and contribute sections right up to his death.