retical studies of population genetics and evolution of bacteria—many of which are cited herein—the formal (mathematical) and informal (verbal) theory of the mechanisms of adaptive evolution in bacteria is modest in its volume, breadth, and level of integration.
Bacteria are haploid, reproduce clonally, and rarely subject their genomes to the confusion of recombination. Thus, one might believe (and both of us once did believe) that the genetic basis of adaptive evolution in these prokaryotes would be simpler than that of so-called higher organisms and that the theory to account for it would be straightforward extensions of that already developed for sexually reproducing eukaryotes. As we shall try to convince the reader, in this personal (read, opinionated) discussion, this situation is by no means simple. Bacteria are different. The mechanisms of adaptive evolution in the prokaryotic world raise a number of delicious theoretical and empirical questions that have only begun to be addressed.
The most striking feature of retrospective studies of genetic variation and molecular evolution in bacteria is the extent to which these organisms are chimeras. Much of the DNA of bacteria classified as Escherichia coli has been acquired relatively recently: more than 17% of the open reading frames of the E. coli K-12 genome was acquired in the last 100 or so million years from organisms with G + C ratios and codon usage patterns distinguishable from those of other strains of E. coli and closely related Enterobacteriacae (Lawrence and Ochman, 1998). Moreover, a substantial amount of the variation in bacteria is not in their chromosomal genes. Bacteria commonly carry arrays of active and retired accessory genetic elements (plasmids, prophages, transposons, and integrons), the composition of which also varies widely among members of the same bacterial species. Although, at any given time, some of these elements, such as insertion sequences and cryptic plasmids, may not carry genes that code for specific host-expressed phenotypes, others are responsible for the more interesting adaptations of bacteria to their environment. For example, many of the genes coding for the adhesins, toxins, and other characters responsible for the pathogenicity of bacteria, “virulence factors,” either are present in clusters known as “pathogenicity islands” (Finlay and Falkow, 1997; Groisman and Ochman, 1996; Hacker et al., 1997; Lee, 1996), which almost certainly had former lives as accessory elements or as parts thereof, or are borne on functional acces-