Reproductive aging patterns in primates reveal that humans are distinct

Abstract

Women rarely give birth after ∼45 y of age, and they experience the cessation of reproductive cycles, menopause, at ∼50 y of age after a fertility decline lasting almost two decades. Such reproductive senescence in mid-lifespan is an evolutionary puzzle of enduring interest because it should be inherently disadvantageous. Furthermore, comparative data on reproductive senescence from other primates, or indeed other mammals, remains relatively rare. Here we carried out a unique detailed comparative study of reproductive senescence in seven species of nonhuman primates in natural populations, using long-term, individual-based data, and compared them to a population of humans experiencing natural fertility and mortality. In four of seven primate species we found that reproductive senescence occurred before death only in a small minority of individuals. In three primate species we found evidence of reproductive senescence that accelerated throughout adulthood; however, its initial rate was much lower than mortality, so that relatively few individuals experienced reproductive senescence before death. In contrast, the human population showed the predicted and well-known pattern in which reproductive senescence occurred before death for many women and its rate accelerated throughout adulthood. These results provide strong support for the hypothesis that reproductive senescence in midlife, although apparent in natural-fertility, natural-mortality populations of humans, is generally absent in other primates living in such populations.

Fig. 1.

Fertility completion (red symbols and lines) compared with mortality (black lines) in nonhuman primates and humans. In four primate species, so few individual experienced fertility completion before death that model parameters for fertility completion were not estimable (muriqui, capuchin, chimpanzee, gorilla); in these cases, the red diamonds indicate the age-specific cumulative incidence functions for fertility completion (adjusting for the competing risk of death). For three primate species a Gompertz model was estimable (baboon, blue monkey, sifaka); in these cases, Gompertz models of fertility completion are plotted as red lines. Note that initial rate (Gompertz a) for fertility completion was significantly lower than initial rate for mortality, so that even though fertility completion accelerated at a faster rate than mortality in these three species, relatively few individuals experienced fertility completion before death. For the !Kung population of humans, the red line showing the Gompertz model (and the thin curved red line showing the Gompertz–Makeham model, a slightly better fit) reveals that fertility completion occurs prior to death for many women in this population.

Fig. 2.

All-cause fertility cessation (red) occurred at the same rate as mortality (black) for seven nonhuman primate populations. All lines represent Gompertz models. The strong similarity between the fertility cessation models and the mortality models indicates that most fertility cessation was due to death, revealing that somatic senescence eclipsed reproductive senescence in these species. Note the strong contrast with the human data in Fig. 1.

Fig. 3.

The life history pacing of fertility completion is different in nonhuman primate species than in humans. The 90th percentiles of the distributions for age at death (in years) are plotted against the 90th percentiles of the distributions for age at last live birth (in years). The distributions include both censored and uncensored data. The black line represents the line of identity between the two distributions. The 90th percentiles of these two distributions are quite similar for the nonhuman primates, but they are quite different for the !Kung population of humans.