Over the past two decades efforts to control malaria have halved the number of cases globally, yet burdens remain high in much of Africa and the elimination of malaria has not been achieved even in areas where extreme reductions have been sustained, such as South Africa(1,2). Studies seeking to understand the paradoxical persistence of malaria in areas in which surface water is absent for 3-8 months of the year have suggested that some species of Anopheles mosquito use long-distance migration(3). Here we confirm this hypothesis through aerial sampling of mosquitoes at 40-290 m above ground level and provide-to our knowledge-the first evidence of windborne migration of African malaria vectors, and consequently of the pathogens that they transmit. Ten species, including the primary malaria vector Anopheles coluzzii, were identified among 235 anopheline mosquitoes that were captured during 617 nocturnal aerial collections in the Sahel of Mali. Notably, females accounted for more than 80% of all of the mosquitoes that we collected. Of these, 90% had taken a blood meal before their migration, which implies that pathogens are probably transported over long distances by migrating females. The likelihood of capturing Anopheles species increased with altitude (the height of the sampling panel above ground level) and during the wet seasons, but variation between years and localities was minimal. Simulated trajectories of mosquito flights indicated that there would be mean nightly displacements of up to 300 km for 9-h flight durations. Annually, the estimated numbers of mosquitoes at altitude that cross a 100-km line perpendicular to the prevailing wind direction included 81,000 Anopheles gambiae sensu stricto, 6 million A. coluzzii and 44 million Anopheles squamosus. These results provide compelling evidence that millions of malaria vectors that have previously fed on blood frequently migrate over hundreds of kilometres, and thus almost certainly spread malaria over these distances. The successful elimination of malaria may therefore depend on whether the sources of migrant vectors can be identified and controlled.