In 13 villages in the savannah zone and 21 villages in the forest zone of Côte d'Ivoire, the biting density of the principal malaria vector, Anopheles gambiae, was studied as a function of rice cultivation in the inland valleys in a 2-km radius around each village. In the savannah villages, during the main season cropping period, surface water on rice-cultivated and to a lesser extent on uncultivated inland valleys seems to contribute strongly to the A. gambiae population density. For the off-season cropping period (which starts after the first light rains in the savannah zone), correlations were weaker. Breeding sites other than in inland valleys may play an important role in the savannah zone. In the forest zone, however, the A. gambiae population density was strongly correlated with the surface water availability (SWA) in the rice-cultivated inland valleys, whereas the correlation with the SWA in other (uncultivated) inland valleys was weak. The requirement of sunlit breeding sites for A. gambiae might explain this difference between zones. In the forest zone, only inland valleys cleared for rice cultivation meet this requirement, whereas all other inland valleys are covered with dense vegetation. In the savannah zone, however, most undergrowth is burnt during the dry season, which permits sunlight to reach puddles resulting from the first rains.@en