Eight hundred and twenty-one strains of the genera Streptomyces and Streptoverticillium were physiologically characterized using a total of 329 miniaturized tests. Overall similarities of all strains were determined by numerical taxonomic techniques using the UPGMA algorithm and the SSM and the SJ coefficients as measures of similarity. Test error was within acceptable limits. Comparison of photometric and visual test reading revealed overall differences of 7.45%. A total of 15 major clusters (six or more strains), 34 minor clusters (less than six strains) and 40 single-member clusters were defined at the 81.5% similarity level (SSM). Two clusters containing physiologically, and in some cases morphologically and genetically, different groups could be further subdivided at the 84.0% similarity level (SSM). Generally, similar groupings were obtained with the Jaccard coefficient at similarity levels ranging from 59.6% to 64.6% similarity (SJ), with changes in the definition of clusters and subclusters. The cophenetic correlation coefficients rCS for the UPGMA/SJ and the UPGMA/Ssm analysis were 0.6929 and 0.8239, respectively. Several phena showed significant overlap with others, indicating the physiological variability within the species. The phenetic data in most cases confirm the major phena of the study of Williams et al. (1983), Journal of General Microbiology 129, 1743-1813 (although the cluster-groups defined in that study could only be detected in part) and the results indicate that the genus Streptomyces is still overspeciated. Some of the major groupings obtained were very much in line with chemotaxonomic and genetical data. However, several clusters containing only a few strains should be regarded as preliminary ‘species’ until further information is available. The majority of Streptoverticillium strains presently assigned to different species formed a homogeneous subcluster defined at the 84.0% similarity level (SSM). Thus, on the basis of numerical phenetic and (published) molecular genetic and chemotaxonomic data, our study supports the suggestion that members of the genus Streptoverticillium be reclassified into the genus Streptomyces.