Most models in life history assume that maximization of λ holds. Demographic genetic theory supports this assumption provided (i) the fecundity of each mating is determined solely by the age and genotype of the female, (ii) the two sexes experience the exact same selection intensities, and (iii) the two sexes are equal in their age specific schedules of survival and fecundity. But these assumptions will often not hold in nature: thus a more general means of analysis closely integrated with a broader genetic treatment is attempted here. First, operators are derived to map sex specific non-overlapping generation models into the more general case in which generations can overlap. Application of such operators then suggests a broader representation of age structured selection in which evolution acts according to a single composite Wrightian topography whose gradient is a weighted mean taken over selective gradients in all of the age classes. Application of this view is straightforward, showing, for example, that (i) the evolutionary maximization of λ, which is the basis of present optimization theories for life history, is extremely fragile with biological deviation from mathematical assumptions used in its derivation, (ii) the cost of meiosis argument is shown not to conflict with traditional life history theory, although its correct application will depend critically and subtly upon how genic effects on fertility are measured, and (iii) correct weightings for age specific selection converge to those of Hamilton (1966) and Schaffer (1974) only when sex specific differences in fertility and survival are nonexistent or when selection occurs only in the demographically dominant sex. For life history problems beyond assumptions underlying tradition approaches the composite adaptive topography seems to provide a suitable alternate means for analysis.