The taxonomic value of the different ‘groups’ into which the Enterobacteriaceae are divided is assessed, and consideration is given to the classification of different groups within the family. No scheme fits the requirements of all workers and two different approaches are made, one for workers in applied bacteriology, and a more logical one for taxonomic purists. For the former scheme it is suggested that common (vulgar) names should be used to describe the groups and their subdivisions. Linnaean binomials are essential for the more systematic scheme, in which the Arizona group is merged with Salmonella; Ballerup-Bethesda is combined with Citrobacter freundii; Sonne’s bacillus and Bacillus alcalescens are included in Escherichia; Cloaca and Hafnia (32011) are included as species of Klebsiella, and Providence (29911) as a species of Proteus. Few of these suggestions are new, but the scheme is a classification of a family and not a collection of ‘groups’.