The precise nature of the pathogen having caused early plague pandemics is uncertain. Although Yersinia pestis is a likely candidate for all three plague pandemics, the very rare direct evidence that can be deduced from ancient DNA (aDNA) analysis is controversial. Moreover, which of the three biovars, Antiqua, Medievalis or Orientalis, was associated with these pandemics is still debated. There is a need for phylogenetic analysis performed on Y. pestis strains isolated from countries from which plague probably arose and is still endemic. In addition there exist technical difficulties inherent to aDNA investigations and a lack of appropriate genetic targets. The recently described CRISPRs (clustered regularly interspaced short palindromic repeats) may represent such a target. CRISPR loci consist of a succession of highly conserved regions separated by specific "spacers" usually of viral origin. To be of use, data describing the mechanisms of evolution and diversity of CRISPRs in Y. pestis, its closest neighbors, and other species which might contaminate ancient DNA, are necessary. The investigation of closely related Y. pestis isolates has revealed recent mutation events in which elements constituting CRISPRs were acquired or lost, providing essential insight on their evolution. Rules deduced represent the basis for subsequent interpretation. In the present study, the CRISPR loci from representative Y. pestis and Yersinia pseudotuberculosis strains were investigated by PCR amplification and sequence analysis. The investigation of this wider panel of strains, including other subspecies or ecotypes within Y. pestis and also Y. pseudotuberculosis strains provides a database of the existing CRISPR spacers and helps predict the expected CRISPR structure of the Y. pestis ancestor. This knowledge will open the way to the development of a spoligotyping assay, in which spacers can be amplified even from highly degraded DNA samples. The data obtained show that CRISPR analysis can provide a very powerful typing tool, adapted to the systematic, large-scale genotyping of Y. pestis isolates, and the creation of international typing databases. In addition, CRISPRs do constitute a very promising new tool and genetic target to investigate ancient DNA. The corresponding genetic targets are small (<70bp), present in multiple copies (usually more than 10), highly conserved and specific. In addition, the assay can be run in any laboratory. Interpretation of the data is not dependent on accurate sequence data.