Nevertheless, we cannot exclude the possibility that this increase in egg size under short day length could be a relic feature of a lost diapause capacity in tropical populations. Indeed, diapause can be quickly counter selected in laboratory (Pumpuni, 1989), or in field populations as described during the species’ colonization of Florida (Lounibos et al., 2003). On the contrary, even though Brazil was colonized by tropical strains of A. albopictus, in laboratory some populations still show a small but significant decrease in egg hatchability under short photoperiod ( Lounibos

et al., 2003). In consequence it is possible that tropical populations could be selected to express diapause again. The black cutworm Agrotis ipsilon is a good example of persistence

ZD1839 of diapausing metabolism: even though larvae and adults of this species are able of diapause elsewhere in the world, Japanese populations migrate from north to south of the island to overwinter. It allows them to avoid diapause initiation, but adults reared under short day length show a delay in the development of their ovarian maturation ( Tauber et al., 1986). We make the hypothesis that short day length is interpreted by females as stressful environmental conditions. If bigger eggs are better adapted to survival in harsh environment, then females will anticipate poor development conditions for its offspring by laying eggs of bigger size. For example, a higher amount of yolk in bigger eggs could increase the survival of embryos. On an interspecific Sirolimus manufacturer scale, larger eggs have an increased developmental duration (Gillooly and Dodson, 2000), and in Aedes (Stegomyia) species larger eggs are more resistant to desiccation (Sota and Mogi, 1992b). It is unclear how these interspecific observations could be verified inside a species. No modulation of desiccation resistance was observed in eggs of a tropical strain of Kuala Lumpur, Malaysia, reared under different photoperiodic conditions (Urbanski et al., 2010a). Photoperiodic

rearing conditions don’t modify maternal wing length; consequently we dismiss a possible indirect effect of maternal size on eggs volume. Previous works on mosquitoes showed that egg length is neither influenced 4��8C by mosquito wing size (Shannon and Hadjinicalao, 1941 and Pumpuni et al., 1992) nor by maternal larval nutritional regime (Pumpuni et al., 1992). However it must be noted that these studies did not measure egg width, which seems to be the main factor of egg volume variability. Finally, the ecological meaning of photoperiod is questioned in tropical populations. Indeed, annual day length variation is subtle under tropics and less well correlated with environmental events (Tauber et al., 1986). The comparisons of the embryogenesis chronology and egg size demonstrate the existence of photo-induced maternal effects in temperate and tropical populations of A. albopictus.