In 1930 Womersely left Imperial College to take up a position as a junior research officer at the Shirley Institute, Manchester. There he applied mathematical techniques to problems in textile manufacture, including research on cotton spinning, drafting fibrous materials, and, through L. H. C. Tippett, the use of mathematical statistics in industrial production and quality control. While at the Shirley Institute he also met Leslie Comrie and became interested in computational techniques. As a result, he spent a month at HM Nautical Almanac Office, London learning Comrie's numerical approaches. In 1936 he collaborated with Douglas Hartree who had built a Differential Analyser at the University of Manchester; together they devised a much cited method for the numerical integration of partial differential equations. In 1937, with war looming, he joined the armaments research department at Woolwich as a scientific officer, and worked on using statistical techniques applied to ballistics and ammunition proofing. In 1942, after the outbreak of World War II, he was appointed assistant director of scientific research at the Ministry of Supply and asked to set up and head the Advisory Service on Statistical Methods. This organization was responsible for advice and research into ammunition supply, engineering factories and the investigations of a range of Government Inspectorates. It was particularly important in ensuring quality control and promoting sample inspection methods to British industry during wartime. In 1944 he joined the British Association mathematical tables committee and in the same year he was appointed as the first superintendent of the Mathematics Division of the National Physical Laboratory. In addition to being responsible for statistical quality control, NPL was tasked with building an electronic computer, for which Womersley coined the name Automatic Computing Engine, echoing Babbage's Analytical Engine. Womersely was therefore responsible for the set up and operation of the first national computing centre in UK. One of his first actions was to visit the US for a fact-finding tour, where he learned about ENIAC, Howard Aitken's Harvard machine, George Stibitz's machines and von Neumann's plans for the binary computer, EDVAC. On his return, he recruited Alan Turing to work on the ACE section and backed him strongly. Womersley also recruited Donald Davies in 1947. However, progress on the ACE project was delayed and Turing developed a dislike of Womersley and disdain for his abilities and left the project in 1948. Davies then took over from Turing and a small experimental model, pilot ACE was produced in 1950. Whether Womersely led NPL successfully has been questioned, although the consensus seems to be that he did a good job in difficult circumstances. He himself left the project in 1950, before the prototype pilot ACE was completed, to join the British Tabulating Machine Company, a forerunner of International Computers Limited. There he recognised that the computers previously developed by academia or governments were too large and expensive to be commercially viable and he recruited Andrew Booth who had developed the All Purpose Electronic Computer at Birbeck University, as a consultant to develop a smaller inexpensive computer. The computer copied from Booth's original design by Ray Bird was named the Hollerith Electronic Computer, and was Britain's first mass-produced business computer. In 1954 Womersley left BTM and joined a research team, led by Donald McDonald at St Bartholomew's Hospital, who were studying blood flow in arteries. This change seems to have been a temporary arrangement to 'fill in time whilst awaiting completion of arrangements to come to WADC'. His collaboration with McDonald may have been prompted by his eldest daughter, Barbara, who was studying medicine at St Bartholomew's. Whatever the reason, this move led to a new and highly productive period in his research, as he applied mathematical and computational techniques to the analysis of blood flow and hemodynamics. Most notably in 1955 he published an article which described a dimensionless parameter which characterised the nature of unsteady flow; subsequently this has been called the Womersley number. In July 1955, as planned, he moved to WADC to take a post as acting chief of the Analysis Section, System Dynamics Branch Aeronautical Research Laboratory. In 1956, he was promoted to Supervisory Mathematician and then Supervisory Aeronautical Research Engineer, although he continued to publish on mathematical aspects of blood flow until his early death in 1958. His 1957 monograph on 'An elastic tube theory of pulse transmission and oscillatory flow in mammalian arteries' is widely regarded as a major influence in the field. In 1957 he returned to Britain for treatment of cancer. He underwent a number of operations in London and returned to Ohio in 1957, but never fully recovered and died at Ohio State University Hospital, Columbus, on 7 March 1958.
Personal life
Womersley married Jean Isobel Jordan in Hammersmith, London in 1931. The couple had three daughters, Barbara, Ruth and Marion. Womersley's wife, Jean lived in Dayton until 1996, and as of 2014, they were survived by two daughters and six grandchildren living in the US and Canada.
Selected publications
Womersley, JR. An Elastic Tube Theory of Pulse Transmission and Oscillatory Flow in Mammalian Arteries, Wright Air Development Center Technical Report 56-614,
A complete bibliography was compiled by Brian E. Carpenter and Robert W. Doran
