During the past two decades, the number of antibacterials that has reached the marketplace each year has declined, whilst resistance to existing antibacterials has increased. New antibacterials are needed to replace those that have become less effective as a result of the emergence of a high level of resistance amongst target bacteria. Antibacterials are developed by targeting live multiplying whole bacterial cells, or essential bacterial molecules such as enzymes. Using these targets, libraries of natural, recombinant or chemically synthesised compounds are screened. Most existing antibacterials have been developed by creating novel analogues of established antibacterials, which are themselves derivatives of natural compounds. Recently, live non-multiplying bacteria have been used as targets. Bacteria in such a phase are much more tolerant to antibacterials than logarithmic phase organisms. Targeting of non-multiplying bacteria has the potential to yield new antibacterials that would shorten the duration of therapy. This would be more convenient for the patient, could reduce the incidence of adverse effects of treatment, and might reduce the emergence of antibacterial resistance. However, there is much to learn about non-multiplying bacteria, particularly the mechanisms that lie behind their profound antibacterial tolerance. New terminology has been proposed for susceptibility tests for antibacterial agents against non-multiplying bacteria, namely: the minimum stationary-cidal concentration and the minimum dormicidal concentration, which are defined as the minimum concentrations of drug that will kill stationary and dormant bacteria, respectively. The relationship between the antibiotic susceptibility of stationary and logarithmic phase bacteria is the stationary/logarithmic ratio. This terminology is suitable for both planktonic and biofilm cultures. In the future, it is likely that most antibacterial drug design will be based on existing antibacterial structures, but an increasing number of new molecular antibacterial structures may emerge from screening against multiplying and perhaps non-multiplying bacteria. The genomic approach has been disappointing so far, but it is still hoped that this will produce novel antibacterial agents.