The Anopheles gambiae complex includes the major vectors of malaria in sub-Saharan Africa where > 80% of all world-wide cases occur. These mosquitoes are characterized by chromosomal inversions associated to the speciation process and to intraspecific ecological and behavioral flexibility. It has been postulated that introgressive hybridization has selectively transferred inversions on the second chromosome between A. gambiae and A. arabiensis, the two most important vectors of malaria. Here we directly test this hypothesis with laboratory experiments in which hybrid populations were established and the fate of chromosomal inversions were followed. Consistent with the hypothesis, "foreign" X chromosomes were eliminated within two generations, while some "foreign" second chromosomes persisted for the duration of the experiments and, judging from the excess of heterozygotes established stable heterotic polymorphisms. Only those second chromosome inversions found naturally in the species could be introgressed.