[Forest ecosystems and Ebola virus].

Despite data collected since the emergence of the Ebola virus in 1976, its natural transmission cycle and especially the nature of its reservoirs and means of transmission are still an enigma. This means that effective epidemiological surveillance and prevention are difficult to implement. The location of outbreak areas has suggested that the reservoir and the transmission cycle of the Ebola virus are closely linked to the rainforest ecosystem. The fact that outbreaks seldom occur suggests the presence of a rare animal reservoir having few contacts with man. Paradoxically, various serological investigations have shown a high prevalence in human beings, especially in forest areas of the Central African Republic (CAR), with no pathology associated. This would appear to suggest a circulation of both pathogenic and non-pathogenic strains as well as frequent contacts with man. The ecological changes resulting from human activity (agriculture and logging) account for the modification of the fauna (movement of rainforest fauna, introduction of savannah species) and could explain a multiplication of contacts. Likewise, it is interesting to note that the centre of outbreaks has always been in areas bordering on forests (ecotone foreset-savannah in the Democratic Republic of Congo, savannah in Sudan). All these considerations have led us to establish a permanent "watch" in areas bordering on forests in the CAR, involving a multidisciplinary approach to the virological study (strain isolation, molecular biology) of the biodiversity of small terrestrial mammals. The results of a study conducted on 947 small mammals has shown for the first time the presence of the Ebola virus genome in two species of rodents and one species of shrew living in forest border areas. These animals must be considered as intermediary hosts and research should now focus on reservoirs in the ecosystem of forest border areas where contacts with man are likely to be more frequent.