Automation is nothing new in microbiology. Any microbiological analysis using blood as a substrate and where the definitive diagnostic method is the identification of fragment of a molecule can readily be reduced to a test done using a commercial kit on an analyser in a clinical chemistry conveyor belt. Infectious diseases serology and, to some extent, clinical virology, automated a decade ago. Much of the work in these microbiological sub-specialities now consists of maintaining machines and the computerised assessment the quality of their diagnostic output, applying the standard methods familiar to any process of mass production. 

Bacteriology and mycology testing have proved recalcitrant to automation. The late 20^th century molecular revolution has failed to deliver on its promises for these fields. Partly this has been due to the variety of substrates but mainly because of the variety of species requiring detection, their genomic and antigenic complexity and the demand for antimicrobial susceptibility testing. However, overall specimen numbers have increased exponentially and high level managerial decisions from the 1990s on to reduce staff numbers and consolidate testing into fewer locations, necessitates more tests being done by fewer people but utilising the same space. In my particular location the response of senior management has been to experiment with automation using the COPAN-WASP and 24 hour services.