Robert, Thierry, Luxereau, Anne, Mariac, Cédric, Kairou, Ali, Allinne, Clémentine, Amoukou, Adamou I., Bani, Jibril, Banoin, Maxime, Beidari, Yakouba, Bezançon, Gilles, Cayeux, Sonia, Couturon, Emmanuel, Dedieu, Valérie, Gamatché, Ibrahim, Hamidou, Abdoulaye, Moussa, Djibo, Sadou, Mahamane, Seydou, Moumouni, Seyni, Ousmane, Tidjani, Moussa, and Sarr, Aboubakry
Le projet a été réalisé au Niger (pays Zarma-Songhai) à l’aide d’analyses anthropologiques et génétiques, autour de deux axes de recherche : i) l’étude des relations entre la diversité des pratiques agricoles, les flux de gènes entre variétés et l’organisation de la diversité aux niveaux agromorphologique et moléculaire ; ii) l’analyse des relations génétiques entre formes cultivées, sauvages et intermédiaires (soun). Les résultats montrent que les pratiques agricoles favorisent les brassages génétiques entre populations variétales, et la dissémination des soun à l’échelle du village et de la région, via les flux de semences. Ceci expliquerait l’abondance des soun dans les champs de la région. L’absence de différenciation génétique (au niveau des marqueurs moléculaires) entre variétés démontre aussi l’importance quantitative des flux de gènes par le biais des hybridations. Ainsi, la gestion des semences, sans doute moins autarcique qu’auparavant, les pratiques culturales et la forte allogamie du mil conduisent à un processus d’homogénéisation, au niveau régional, des variétés cultivées par les agriculteurs. Il est toutefois difficile de savoir s’il s’agit d’un processus récent. Ce processus aurait pour conséquences une déstructuration rapide des qualités d’adaptation des variétés aux conditions locales, et une augmentation de l’introgression par des gènes des formes sauvages dont la présence des soun est une manifestation. Il pourrait à l’extrême conduire à une « dé-domestication » du mil dans ces régions soumises à des pénuries de production chroniques et donc participer à l’aggravation du phénomène. The project was settled mainly in two localities in West Niger (Zarma-Songhai region). It includes both anthropologic and genetic analyses and aims at: i) studying the relationships between farmers’ practices, gene flow and the organization of diversity at morphological and molecular levels: ii) analysing the genetic relationships between domesticated, wild and intermediate forms (soun) of pearl millet based on their genetic and morphological evaluation at the field level. The purpose was to understand the origin of so-called soun (plants with intermediate domesticated/ wild phenotypes) and to evaluate the level of introgression of cultivated populations by genes from the wild. The results showed that, although diverse, farmer practices favour very large genetic admixture at the village and even regional scale, through important seed flows (seed exchanges and supply networks). A consequence of this is the near-absence of any genetic differentiation (at the molecular level) between populations of the same variety cultivated by different farmers, even in remote villages. Both the abundant seed flows and the existence of a large seed bank in the soil could explain why soun are so abundant in large areas in the region, even where no wild forms are found. Moreover, the very low level of genetic differentiation between different varieties (including between early and semi–late flowering ones) can be used as a proof of the sizeable gene flow through hybridization at the field level (different varieties are often found in the vicinity of each other or even in the same field). Farmers are able to identify the different varieties they grow and to manage them separetely. Nevertheless, different factors act together to lead both to the homogeneization of the cultivated gene pools, and then varieties, at large geographical scales and to the existence of gene flow between domesticated and non domesticated forms of pearl millet. In particular, farmers pressures during the seed selection process cannot be neither strong nor evenly applied due to the scarcity in production. We think also they are probably less strong that they used to be; cultural practices and the largely preponderant outcrossing mating system promote hybridization between different varieties; weeding is not efficient and/or discriminant enough to remote all the soun growing in the field, so that the proportion of these plants can sometimes reach several tens per cent. However, it is difficult to ascertain whether we are facing a recent evolutionary process due to dramatic social and ecological (repeated drought) changes. If this were the case, this could lead to a decrease in the level of adaptation of cultivated populations to their local environment and to an increase in their introgression by genes from the wild. At the end, this could lead to a “de-domestication” of pearl millet in this region that already suffers of scarcity in the production of this crop, reinforcing therefore this situation.