The four decades until the present day have seen major changes in the theory and practice of bacterial identification. Heightened expectations from the public, government authorities and clinicians – including accurate identifications of zoonotic bacteria, of potential bioterrorism agents, and of emerging pathogens – have raised the bar for both routine clinical microbiology laboratories and for the specialist identification laboratory. Accreditation and compliance requirements (particularly with Security Sensitive Biological Agents) have added complexity to laboratory test standards and documentation. Perhaps most significant are the changes in technology. Labour-intensive methods (dependent on experience and an “apprentice – style” learning) were streamlined into miniaturised and automated substrate based systems; then molecular biology and mass spectrometry transformed both bacterial taxonomy and laboratory practice. I will describe my experience of identification of unusual and “difficult” bacterial isolates in a public health laboratory over these years; I will give examples of particular diagnostic puzzles and problems; and I will illustrate my opinion that for me - while genomics is the current state-of-the-art technology – Hans Christian Gram’s legacy remains a pillar of bacterial identification.