Trying to Make Evolutionary Sense of Menopause

This is a bit of a deviation from neuroscience (although neuroscience and evolution are fundamentally related), but I stumbled upon an article in PNAS about menopause that I found interesting, and I wanted to comment on it. I never really thought much about the evolution of menopause, and now that I have, it is a very unusual biological process (as well as a very unpleasant one for women). I don’t know if Darwin ever considered menopause in reference to his theories. If any readers know of any instances where he did, I’d be grateful if you could give me some page numbers or quotes.

According to evolutionary theory, the goal of any organism is to procreate—to pass on its genes. Thus, it has always been an enigma to evolutionary theorists why human females live so long after they have lost the ability to reproduce. They are the only ones in the primate family that have a long postreproductive life. It is confusing in and of itself that they lose the ability to reproduce during their lifetime, as it is quite rare in the rest of the animal kingdom.

This anomaly has caused some to suggest that menopause is not an adaptive trait, but a byproduct of the medical advancements we’ve made that have resulted in an extended human lifespan. Proponents of this argument assert that women now simply outlive their supply of egg follicles. This hypothesis is contradicted, however, by the fact that even in contemporary hunter-gatherer societies that have no access to modern medicine, women still experience menopause and live into their sixties.

Thus, evolutionary biologists have continued to try to develop an evolutionary explanation for menopause and a long postreproductive lifespan. In order to do so, it is necessary to figure out why these things may have conferred an adaptive advantage to our ancestors. This has given evolutionary theorists fits.

The first reasonable explanation was espoused by Dr. George C. Williams in 1957. He hypothesized that menopause is adaptive because it keeps older women from being exposed to the risks associated with childbirth (which were much higher for our ancestors). This allows them to remain alive long enough to ensure their children are raised to maturity to have their grandchildren (thus continuing the original mother’s gene line). This became known as the grandmother hypothesis.

Dr. Kristin Hawkes and colleagues elaborated on this hypothesis in 1997, when they studied a contemporary hunter-gatherer society in Tanzania called the Hadza. Hadza grandmothers are among the most assiduous workers in the society. They spend up to eight hours a day gathering food, which they bring home to feed their grandchildren. When Dr. Hawkes’ group saw how important the role of the grandmother was in Hadza society, they suggested that a long postreproductive life allowed those women to focus on the health of their grandchildren. This ability to provide for grandchildren and encourage the continuation of their genetic heritage, Hawke asserted, could have led to natural selection for menopause and living long afterwards.

Evolutionary biologists, however, were still not satisfied with this hypothesis. In order for menopause to outweigh the advantage of a continued ability to reproduce, a postmenopausal woman’s children would have to have twice as many children themselves. Infant mortality among those grandchildren would also have to be virtually nonexistent. Thus, when one crunches the numbers, the grandmother hypothesis doesn’t seem to quite add up. Certainly grandmotherly assistance in raising a daughter’s children is adaptive, but a quantitative analysis can’t justify the loss of childbearing ability to begin with.

Researchers at the Universities of Cambridge and Exeter published a paper last week in PNAS that they hope may help to resolve the debate over the evolutionary origins of menopause. They suggest that previous models have focused on personal-fitness and kin-selected fitness, or in other words, the health risks of reproduction for an older woman and the assistance she provides to the survival of her kin (like seen in the Hadza). What is ignored, they assert, is competition—namely reproductive competition between new females introduced into a group and older females who already have offspring in the group.

Their model is based on what is called “female-biased dispersal”, which simply means that in the hunter-gathering times of our species, females were more prone to move between social groups than men. A female newcomer in a group would have no genetic ties to anyone else in the group, and would have to compete with other women in the group for chances to reproduce. Older women who already had children could still continue their genetic heritage through grandchildren, instead of having more children themselves. Thus, the newcomer female would win the competition because procreating with a male in the group was her only reproductive option, while the older female had the option of gaining a grandchild and helping it to survive—an option that results in a greater chance of promoting her genetic heritage than procreating herself.

As evidence for this model, the authors present data that shows the reproductive overlap in humans is extremely low when compared to other primates. On average, women have their first baby at age 19, and their last at age 38—exactly when their first-born has reached normal breeding age. They also state that the rate of attrition of the initial oocyte stock in human females should allow for reproduction up to an age of about 70. At around age 38, however, there is an increased rate of ovarian follicular hazard. By age 50 (the average start of menopause), follicle stocks have dropped below reproductive levels. Thus, according to the researchers, at the age when they can begin to expect reproductive competition from the next generation of females, reproductive senescence is accelerated.

The authors suggest this hypothesis as a complement to, not a replacement for, the grandmother hypothesis. While the assistance a grandmother provides in caring for her grandchildren has an adaptive quality, they argue it cannot explain a female’s loss of reproductive ability. A combination of both theories, they state, best explains menopause and a long postreproductive lifespan.

While this is certainly a feasible hypothesis, it is far from the last word on the topic. In order for this theory to develop a stronger foundation, more studies must be conducted on other species that have cooperative breeding societies. It will be important to find if the reproductive overlap in these groups is similar to that seen in humans. And, although the reproductive overlap data is intriguing, menopause still seems to me to be an awfully complex mechanism to have evolved to reduce reproductive competition. Of course, evolution has resulted in many other inexplicably complex mechanisms.

If I were a woman I would be a little disgusted with evolution for forcing me to go through all of these uncomfortable processes like monthly cycles, pregnancy, and menopause. If evolution were fair, men would have to deal with at least one of these burdens. On the other hand, as a man I’m glad evolution isn’t fair—sorry!


Cant, M.A., Johnstone, R.A. (2008). Reproductive conflict and the separation of reproductive generations in humans. Proceedings of the National Academy of Sciences DOI:10.1073/pnas.0711911105