A probability matrix for computer-assisted identification of vibrios has been constructed, based on the API20E system. Data were gathered from 173 strains representing 31 taxa of vibrios and related organisms, from a variety of sources. The matrix was tested internally by four statistical programs. Program OVERMAT and program MOSTTYP tested the separation and homogeneity of the taxa. Most of the taxa were satisfactory but a few were less so; reasons for this are discussed. Program CHARSEP and program DIACHAR tested the separation and diagnostic values, respectively, of the 50 characters used and from the results a shorter diagnostic test set was drawn up. The overall test error rate was 4.5%. The matrix was assessed externally by its performance in the identification of vibrio-like strains isolated from freshwater. The overall success rate for identification was 84%, using a Willcox score of ?0.99. The data generated for 172 of the reference strains and 243 wild strains were subjected to numerical taxonomic analysis. The main purpose of these analyses was to verify the quality of the identification scheme. Nevertheless the results produced were in good agreement with those of more detailed taxonomic studies. The taxonomic position of some named and some unnamed groups of Vibrio and Aeromonas was clarified, and taxa showing poor distinction were highlighted. An ecological survey was carried out to determine the distribution and seasonal occurrence of vibrios in various freshwater sites in the U.K. The results showed that species of vibrio both pathogenic and non-pathogenic for man, are widely distributed in rivers and canals. Some of the species considered to be indigenous to the waters. The role of these organisms in the freshwater environment remains to be determined.