The "oviposition preference-offspring performance" hypothesis proposes that females lay their eggs in habitats that maximize the fitness of their offspring. The aim of this study was to assess the oviposition site selection by Aedes aegypti females and the success of their larvae in habitats with different detritus accumulation times, under conditions representative of the natural spatial variability of detritus quality and quantity in the Metropolitan Area of Buenos Aires, Argentina. Two experiments were performed, one assessing oviposition site selection and the other analyzing developmental success. In both experiments, two levels of detritus accumulation time were compared, one with short-time detritus accumulation (2 weeks), and the other with long-time detritus accumulation (8 weeks). Naturally fallen detritus was used in both experiments, collected in ten sites across the Metropolitan Area of Buenos Aires. In the oviposition experiment, two contiguous ovitraps corresponding to each accumulation time were placed at each of the ten sites and the number of eggs received for each accumulation time was compared. In the development experiment, always 19 larvae were raised in containers of both accumulation times and overall performance was compared using an integrated index that considers survival, development time and female size. A large variability in the amount of detritus collected at the different sites was observed. The number of eggs was significantly higher in the long-time than in the short-time detritus accumulation containers, and approximately proportional to the amount of detritus in each ovitrap. The performance was not affected by the detritus accumulation time, but a better performance was detected in containers that received a higher amount of organic detritus, regardless of the accumulation time. Leaves were on average the most abundant type of detritus, with an average of 53 % of the total detritus collected. The amount of leaves added 2 weeks before hatching showed a positive effect on larval performance. Our results do not support the "oviposition preference-offspring performance" hypothesis, since Ae. aegypti females laid eggs in containers where larvae did not show a better performance. Moreover, at larval densities related to the number of eggs actually laid in each of the accumulation times, it is expected that the performance would be even worse in the most selected containers, due to the density-dependent effects of crowding. Since the results obtained reflect the natural heterogeneity of the environmental conditions in the region studied, they might be a fairly good indicator of what occurs in natural larval habitats.