Abstract : The studied model is the vascular plant pathogen Ralstonia solanacearum, with a particular focus on phylotype II strains. This telluric bacterium has a wide diversity, both on genotypic and phenotypic levels. Its evolving classification reflects the need to clarify its unusual biodiversity and seek to identify ecotype structure in this species complex, i.e., groups of strains with both genotypic and specific biological traits. Within the framework of this model pathosystem, we initially focused on deeply revisiting pathotypes among ecotypes, although well described in the literature, or describing new ecotypes (African phylotype III). We observed high phenotypic convergence between strains from phylotype III from the African highlands and Brown rot strains from phylotype IIB-1, both able to trigger wilt symptoms on potato and other Solanaceae at cold temperatures. Adaptation of diverse strains for cold tolerance led us to investigate the R. solanacearum situation in Europe and more specifically in the Mediterranean regions. This strategy allowed us to appreciate the significant divergence towards pathogenicity (virulence and aggressiveness) on Solanaceae within clonal-like structure of strains in the Brown rot ecotype, which also established latent interactions in the banana vascular system. In the mean time, phenotypes of banana pathogenic strains unifying the Moko ecotype, was also revisited on Solanaceae, and was able to trigger symptoms on both susceptible and resistant genetic resources to bacterial wilt. All these experimental data yielded selection criteria for choosing three new candidate strains in the R. solanacearum species complex for complete genome sequencing. Our genomic comparative approach allowed us to describe the first pangenome of this pathogen: all targeted identified genes of this species complex. These data were analyzed by various bioinformatic approaches and allowed us to design a complete reshaping of R. solanacearum species complex into three distinct genomic species, firstly clustering strains from phylotype I (Asia) with strains from phylotype III (Africa); strains from phylotype II (America); and lastly, strains from phylotype IV (Indonesia). This pangenome was then used for designing a DNA microarray, a high resolution tool that allowed us to explore a wide set of genomes. The density of accumulated data allowed for a molecular ecological approach to retrieve a certain amount of the evolutionary past of R. solanacearum phylotype II strains. Furthermore, a deeper analysis of these genomic data, combining phylogeography with population structure analysis of the Brown rot ecotype, revealed a dual epidemic situation in Europe, both across Andean and African influences. Similarly, the Moko ecotype presents three distinct genetic structures. These data were analyzed within the purpose of tracking the main gene flows in the ancestral states of phylotypes and to unravel the strong contribution of the mobile elements, genes related to environmental adaptation, and pathogenicity as a major driving force into the evolution of this successful plant pathogen.