The phenetic structure of Aedes albopictus.

The large and recent migrations of the main vector of the Chikungunya virus, Aedes albopictus, raise questions about the epidemiological impact of possible microevolutionary changes in new territories. Quantitative traits are suitable to detect such changes as induced by environmental adaptations, local competition and founder effects. Using landmark-based geometric morphometrics, we compared the size and shape of 22 populations (1572 females) of Ae. albopictus. The collection sites covered five countries around the world, with special emphasis on Asia, which is assumed to be the origin of the vector. Some collections came from places where an important epidemic outbreak of Chikungunya has recently occurred. Quantitative comparisons were based on 18 anatomical landmarks on the wing. To contrast geographic and possible interspecific shape variation, close species were introduced in the sample, namely five Aedes riversi and nine Ae. flavopictus from Japan. The three species had similar size, but they were clearly separated by shape. Within Ae. albopictus, there was general agreement on size variation with the available climatic data. Shape variation was less related to climatic data than to either geography or a known itinerary of past territorial expansion. Thus, two main clusters were distinguished by wing shape variation: the first one composed of the Southeast Asian sample, the second one grouping Japan, Florida, Hawaii and the Reunion Island samples. The Southeast Asian countries, assumed to be at the origin of the geographic expansion of the mosquito, had similar wings and constituted a distinct group where localities clustered into northern and southern localities. Contrasting with this homogeneous group, very distant localities such as United States (US) and Japan shared a common shape pattern. The US Ae. albopictus samples (Hawaii and Florida) were indeed very similar to the Japanese samples, with Florida behaving exactly like a northern Japanese locality, close to Tokyo and Yokohama. Shape proximity among these distant areas could be explained by a common and relatively recent ancestor, as generally suggested in the literature. The three Reunion Island samples conformed a group external to the Southeast Asian countries, loosely connected to the remaining localities. Thus, areas where recent epidemics took place, the Reunion Island and the southernmost localities of Thailand, did not cluster together, but represented separate groups. In sum, the size of the wing was apparently under the influence of climatic factors, while its shape could contribute information on species and geographic differences. Apparent departures from expectations based on geography were explained by known past and present migratory routes. These patterns of shape variation were compatible with genetic drift, suggesting microevolutionary changes probably induced by the expansion of the Ae. albopictus mosquito.