Résumé
Agriculture faces great challenges to overcome global warming and improve system sustainability, requiring access to novel genetic diversity. So far, wild populations and local landraces remain poorly explored. This is notably the case for the two diploid species, Brassica oleracea L. (CC, 2n=2x=18) and B. rapa L. (AA, 2n=2x=20). In order to explore the genetic diversity in both species, we have collected populations in their centre of origin, the Mediterranean basin, on a large contrasting climatic and soil gradient from northern Europe to southern sub-Saharan regions. In these areas, we also collected 14 populations belonging to five B. oleracea closely related species. Our objective was to ensure the absence of species misidentification at the seedling stage among the populations collected and to describe thereafter their origins. We combined flow cytometry, sequencing of a species-specific chloroplast genomic region, as well as cytogenetic analyses in case of unexpected results for taxonomic verification. Out of the 112 B. oleracea and 154 B. rapa populations collected, 103 and 146, respectively, presented a good germination rate and eighteen populations were misidentified. The most frequent mistake was the confusion of these diploid species with B. napus. Additionally for B. rapa, two autotetraploid populations were observed. Habitats of the collected and confirmed wild populations and landraces are described in this study. The unique plant material described here will serve to investigate the genomic regions involved in adaptation to climate and microbiota within the framework of the H2020 Prima project ‘BrasExplor’.