This study quantified the prevalence of positive Darwinian selection in a number of bacterial genera. This was achieved using an automated pipeline that utilises newly developed maximum likelihood methods and also parsimony based methods. The percentage of genes found to be evolving under the influence of adaptive evolution was found to be similar to results observed in two eukaryote lineages. The evolutionary history of several membrane associated bacterial genes was also investigated. Presently a number of these genes are in phase I clinical trials in an effort to investigate their suitability as putative vaccine candidates. A number of these genes were shown to exhibit signatures of positive Darwinian section. This finding has serious implications as recent studies have shown that vaccines that target variable epitopes are less successful then those that target highly conserved epitopes. The implications of these finding are discussed. A phylogeny derived from four Neisseria genomes was proposed using character congruence and total congruence. These two methods are at the centre of heated debate as to which best describes the true phylogeny of large datasets. Systematic problems associated with the character congruence approach were investigated and possible solutions discussed. An analysis of 31 individual and concatenated protein data sets encoded in Reclinomonas americana and Marchantia polymorpha mitochondrial genomes revealed that based on the available sequence data, the Rickettsiaceae family are the mitochondrion ancestor. A previous study using the same data and similar methodologies concluded that Rhodospirillum rubrum came as close to mitochondria as any -proteobacterium investigated. A robust phylogeny for the -proteobacteria is also proposed.