An important issue in the study of the evolution of aging in Drosophila melanogaster is whether decreased early fecundity is inextricably coupled with increased life span in selection experiments on age at reproduction. Here, this problem has been tackled using an experimental design in which selection is applied directly to longevity. Selection appeared successful for short and long life, in females as well as males. Progeny production of females selected for long life was lower than for short-lived females throughout their whole life. No increase of late-life reproduction in long-lived females occurred, as has been found in selection experiments on age at reproduction. This discrepancy is explained in terms of the inadequacy of the latter design to separate selection on life span from selection on late-life fecundity. Moreover, starvation resistance and fat content were lower for adults selected for short life. In general, the data support the negative-pleiotropy-disposable-soma theory of aging, and it is hypothesized that the pleiotropic allocation of resources to maintenance versus to reproduction as implicated in the theory might involve lipid metabolism. It is argued that further research on this suggestion is urgent and should certainly comprise observations on male reproduction because these are for the greater part still lacking. In conclusion, the longevity of D. melanogaster can be genetically altered in a direct-selection design, and such an increase is accompanied by a decreased general reproduction and thus early reproduction.