To screen isolates and to identify mecA alleles, published mecA sequences were analyzed, and a microarray for the rapid discrimination of mecA alleles was designed. A GenBank analysis yielded 135 full-length gene sequences annotated as mecA. These sequences clustered into 32 different alleles corresponding to 28 unique amino acid sequences and to 15 distinct hybridization patterns on this microarray. A collection of 78 clinical and veterinary isolates of Staphylococcus spp. was characterized using this assay. Nine of the 15 expected patterns, as well as one as-yet-unknown pattern, were identified. These patterns were detected in various epidemic methicillin-resistant Staphylococcus aureus strains, in S. pseudintermedius, and in coagulase-negative species such as S. epidermidis, S. fleurettii, or S. haemolyticus. There was no correlation between the different mecA hybridization patterns and the SCCmec type. Determination of MICs showed that mecA alleles corresponding to only four of these nine patterns were associated with β-lactam resistance. The mecA alleles that did not confer β-lactam resistance were largely restricted to coagulase-negative staphylococci of animal origin, such as S. sciuri and S. vitulinus. Because of the diversity of sequences and the different impact on β-lactam susceptibility, the existence of different mecA alleles needs to be taken into account when designing diagnostic assays for the detection of mecA.