In the present work the M42-A disk diffusion protocols of the Clinical Laboratory Standards Institute were used by two laboratories to provide data on the epidemiological cut-off values for the Streptococcus phocae isolates obtained from diseased Atlantic salmon (Salmo solar) (n = 27) and seal (Phoca vitulina) (n = 4) to florfenicol (FLO(30)), erythromycin (ERY(15)) and oxytetracycline (OTC(30)). The data generated by the two laboratories for FLO(30) zone sizes were normally distributed with a mean of 27.1 mm and a standard deviation of 2.5 mm, while for ERY(15) they had a mean of 25.5 mm and a standard deviation of 1.7 mm. For FLO(30) and ERY(15) the wild type cut-off values (CO(wt)) calculated from the data for all 31 strains were <= 20 mm for both agents. Neither of the OTC(30) data sets from Lab 1 or Lab 2 was normally distributed. Using the data from both laboratories the CO(wt) for OTC(30) was <= 21 mm and applying this CO(wt), suggested that 11 strains should be treated as WT and 20 as non wild-type. Of the strains isolated from salmon, the two strains isolated in 1999 and 4 of the six isolated during 2001-2002 were classified, using the CO(wt) of <= 21 mm, as WT Of the 19 strains isolated from 2004 onward, only one, isolated in 2004, was classified as WT. The CO(wt) values, generated by normalized resistance interpretation, can be used as epidemiological cut-off values and have the potential to reduce errors in interpreting susceptibility test data for S. phocae. Moreover, the data generated in this work demonstrated the importance of reducing OTC use for the streptococcal treatment. (C) 2011 Elsevier B.V. All rights reserved.