Aims: Evaluation of composition and evolution of the coagulase-negative staphylococci (CNS) communities in two traditionally fermented sausages (salsiccia and soppressata lucana) produced in Basilicata, southern Italy. Methods and Results: A culture-dependent approach based on isolation on selective media and identification with phenotypic and molecular methods was used. Phenotypic data of 471 strains were analysed by multivariate statistical methods by using 28 strains from culture collections and 48 strains identified by molecular methods (such as 16S rDNA sequencing, species-specific PCR assays, intergenic spacer region-PCR and PCR-denaturing gradient gel electrophoresis) as a reference. The CNS microflora of the sausages was found to be dominated by different biotypes of Staphylococcus xylosus (51.2%), followed by S. pulvereri/vitulus, S. equorum and S. saprophyticus (13.4, 10.2 and 10%, respectively). Other species (S. succinus, S. pasteuri, S. epidermidis, S. warneri and Macrococcus caseolyticus) were also present at lower levels. Identification of 25% of the isolates was impossible. Conclusions: The composition of CNS communities varied significantly with sausage type, plant and ripening time and clear differences were found among communities of salsiccia and soppressata at the end of ripening. Significance and Impact of the Study: Phenotypic characterization, supported by molecular and statistical analyses, can be considered a useful approach for typing a large number of isolates and for monitoring the evolution of staphylococcal communities during sausage fermentation but does not always provide a satisfactory identification of the isolates.