Future studies of population structure and the evolution of bacterial species will require DNA sequences from multiple loci and large databases, such as are provided by MLST and SNP-typing. As these techniques become cheaper and faster, they will also become essential for typing and long-term epidemiology. This scenario demands independent curation in order to ensure that combined data from multiple laboratories are reliable. A further requirement is that sampling strategies reflect the biology of the particular pathogen and include representative material from the developing world. Cholera and tuberculosis provide prime examples where global collaboration on international problems is essential. For some bacterial species, the primary current problem is a lack of knowledge about population structure and the availability of adequate statistical methods. Understanding population structure demands knowledge about epidemic processes and pathogen evolution that are still lacking. Many important parameters for bacterial populations need to be calculated, including size, growth rate, and generation time. The relative importance of diverse evolutionary processes also needs quantification, including mutation, homologous recombination, and horizontal genetic exchange. In addition to these classical parameters,we are largely ignorant about the details of microevolution of bacterial populations in nature. Knowing population parameters would aid typing. For example, determining clock rates for different types of genes could facilitate the choice of appropriate genes for different questions. Similarly, genomic approaches such as microarray will help distinguish between neutral (core) and variable (accessory) genes. Neutral genes are more appropriate for analyses of the demographic and evolutionary history of bacterial populations whereas variable genes could shed light on the processes of adaptation and speciation. Increased knowledge about these processes will in turn facilitate