In some cases, sequential hermaphroditism can convey a selective advantage to an individual by increasing its reproductive potential relative to nontransforming members of the population. This is because age-specific fecundity in many populations is not distributed in the same way for males and females. By functioning as that sex which has the higher fecundity in a particular age span, an individual could increase its reproductive potential relative to lifetime males or females. The magnitude of the reward that comes from such sex changes depends on both the demography of the population and its spawning habits. Protandry may be selected for in populations where female fecundity increases with age and where individuals mate at random. Factors which favor the evolution of protogyny are those which tend to depress male fecundity values at early ages, such as inexperience, territoriality, or female mate selection. Selection for protogyny can also exist when female fecundity decreases with age, although this situation seems rare in the field. It appears that the nature of the fecundity schedule for females has a greater effect on the magnitude of the selection pressure for sequential hermaphroditism than does the age structure of the population. If sequential hermaphroditism becomes genetically fixed in some members of a population with differential age-specific fecundities, the frequency of the hermaphrodites should increase over time until they eliminate the gonochoristic individuals. The pattern of sexual transformation eventually reached in a population is predictable through the use of a simulation model developed here. The pattern depends on the age structure, female fecundity schedule, and mating characteristics of the population. For two examples, the predictions appear consistent with what is seen in nature. At equilibrium, the simulated populations consist of individuals which changed sex at one of two ages, and each type has essentially the same reproductive potential. This suggests that we should see a rather sharp break in age-specific sex ratios in protogynous or protandrous populations. Some hermaphroditic populations exhibit such sharp changes, but others do not. The expected pattern would tend to be obscured by differing growth rates among individuals.