Population genetic studies of vectors are essential for (i) the determination of their taxonomic status and consequently the definition of their vectorial role in the transmission of pathogenic agents; (ii) the evaluation of the species genetic variability and the estimation of their capacities of adaptation to selection pressure; (iii) an estimation of gene flow among populations in order to evaluate their degree of isolation and gene circulation, especially resistance genes. Among the malaria vectors taken as examples on three continents, Africa, South-East Asia and Latin America, the large majority of the species showed an important polymorphism. The Gambiae Complex, which is by far the most studied one, includes at present 7 species with the recent description of An. quadriannulatus A and B from Ethiopia. An. gambiae s.s. includes itself 5 chromosomal forms. One of them, the Mopti form, should be considered as a species unto itself. For An. arabiensis, a strong differentiation has been observed among the populations from Senegal and the Indian Ocean Islands. The kdr mutation, which confers resistance to pyrethroid knockdown effect, has never been found either in the Mopti form, or An. arabiensis, indicating a restricted gene flow between these latter two and An. gambiae s.s. The speciation process of the Gambiae Complex seems to be a recent phenomenon due to environmental selection pressure. Species of the Funestus Group are distinguishable by morphological characters. The genetic study of An. funestus s.s. did not show the presence of a complex, in spite of high polymorphism and population structure. Anophelines from eastern areas present an important biodiversity. The Minimus Complex includes two species, A and C, which are widely distributed in South-East Asia. Species A is strongly endophilic, on the contrary species C is at once more exophilic and zoophilic. The latter species might have been selected by DDT indoor house spraying. After numerous taxonomic investigations, the Dirus Complex includes now 7 species. In Latin America, An. pseudopunctipennis clustered into three geographic populations which are under a speciation process. One covers North America and Guatemala, the other South America and Belize, whilst the last one is restricted to Grenada Island. On the contrary, An. darlingi showed little morphologic and genetic variability throughout the species geographic range suggesting the existence of a single species. The main objective of these studies is to implement a more selective approach of vector control programs in relation to the incriminated species, their bioecology and their role in malaria transmission. The improvement of efficiency and selectivity of vector control is becoming a major goal in order to make the best out of the available tools and control the impact of interventions on the environment.