More Recent Comments

Friday, December 19, 2008

How Women Got Their Menopause

The December 13 issue of New Scientist has an article about the evolution of menopause in humans [Are daughters-in-law to blame for the menopause? ]. The author is Alison Motluk, the Toronto corresponent for New Scientist.

She begins with ..
IT FLIES in the face of natural selection, yet in humans it seems fixed and universal: at around age 50, not far past the midpoint of life, normal healthy women lose their capacity to bear children. Following a decade of gentle winding down, the whole reproductive system screeches to a halt. It is as though, after a few years of wearing bifocals, all women suddenly went blind.

Menopause is a mystery. It leaves women with 20, 30, perhaps even 50 years of life - squandered time in evolutionary terms, because no further genes can be passed on. Yet the selection pressure for menopause must have been strong: there are no known pockets of women around the world who do not go through it. All the evidence suggests menopause has been around a long time, and that the age at which it hits has changed little. Increased longevity seems not to have budged our closing hours. Nor, apparently, has lifestyle; it hits hunter-gatherers at pretty much the same age as hip New Yorkers.
The search for an adaptive explanation for menopause has been going on for over fifty years.

The most common just-so story is called the "Grandmother Hypothesis." It imagines a time in the distant past when humans females were fertile until the day they died, which for 75% of women who survived childhood was before the age of 30 according to a recent study. A mutation arose in one individual and the effect was to induce menopause, or sterility, at about age 50. This new mutation proved to be so beneficial that it spread throughout the species. Today every woman undergoes the pain and frustration of menopause.

Why was it so beneficial? Because post-menopausal women whose partners were still alive could invest their time in looking after their grandchildren instead of having more children of their own.

To its credit, the New Scientist article reviews the latest work on the Grandmother Hypothesis and concludes, correctly, that the effect on ancient hunter-gather societies could not possibly have been significant enough to be adaptive.

The same reasoning applies to the "Mother Hypothesis," which claims that by going through menopause a mother will avoid the risks of future childbirths enabling her to concentrate on raising her existing children. Both of these hypotheses assume that "just saying no" was not a reasonable strategy in ancient societies and that's why menopause was necessary.

So what's the alternative if you are committed to an adaptive just-so story? You are going to be shocked ... menopause evolved to benefit daughters-in-law!!! (Our daughter-in-law will be so pleased.)

I'm not even going to dignify such a stupid idea by pointing out the obvious flaws.

In case you haven't clued in by now, the bottom line is that all these hypotheses are flawed from the get-go because they all make the unnecessary assumption that the pain and suffering of menopause are adaptive. What if they aren't? What if menopause is neutral with respect to evolution or even maladaptive? What's the point of making up irrational just-so stories when there's no evidence to suggest that menopause has any positive effect of fitness?


24 comments :

Anonymous said...

"Today every woman undergoes the pain and frustration of menopause. . . the pain and suffering of menopause"

What pain and frustration and suffering? I had no pain and suffering; I got rid of pain and suffering from fluctuating hormone levels. I think menopause is wonderful and I'd vote for having it at 30.

Andrew said...

all this crap was gospel in BIO150.

Larry Moran said...

Postdiluvian says,

all this crap was gospel in BIO150

That makes me very sad.

Anonymous said...

“What if menopause is neutral with respect to evolution or even maladaptive?”

A loss of reproductive capacity is a severe fitness loss, not very likely to be neutral. It is comparable with dropping dead at that age (well…from an evolutionary perspective, that is). If it is maladaptive, the question is why it persists. Something like that screams for an explanation. If we have a hypothesis, we can test it.
I concur that the cause may well be non-adaptive, for example an evolutionary constraint, but we are not aware of such causes. Until we do, an adaptive explanation makes as much sense as any other, I guess.

Larry Moran said...

Corneel says,

A loss of reproductive capacity is a severe fitness loss, not very likely to be neutral.

Sounds like you should have no trouble identifying the fitness cost.

Given that 25% of hunter-gatherer women survive past 30 years old, and given that of those who survive past 50 years old many will not have a spouse (no sex, no children). How many +50 year old women were having children on a regular basis in societies that lived in Africa about 100,000 years ago?

You must know the answer since you are so certain that menopause caused a significant loss of reproductive capacity.

Please share it with us.

crf said...

I found these articles on whether apes may go through menopause.

This one says maybe. They are studying it:

http://www.associatedcontent.com/article/37918/do_apes_go_through_menopause.html?cat=53

This one says no, and mentions the grandmothering adaptive hypothesis:

http://www.newscientist.com/article/dn13052-menopause-sets-humans-apart-from-chimps.html

Wavefunction said...

Spandrels?

Anonymous said...

You´re argument is circular, in that you assume that an "old woman" will not find a spouse. A women that shows no sign of (reproductive) ageing, should have no problem finding a spouse. Those women that DO live to that age suffer a fitness loss, by not being able to reproduce. If hardly any women lived to that age, as you suggest, then you are right, and menopause was neutral. But then the question is why demographic ageing does not proceed at a similar pace as reproductive ageing. Why don´t we all just die at the age of 40?

I admit, that´s not really an answer. But still, can you suggest an alternative scenario that causes demographic and reproductive ageing to proceed at a different pace?.

Anonymous said...

What if it was a benefit with a hidden cost?
The same gene that causes menopause might be responsible for something extremely beneficial in women who live to be only 30.
For example, if it's the gene that makes women fertile for all the year, instead of short periods of fertility she got from her ancestors, and the woman dies at ~30 anyway, it'll be selected quite well, wouldn't it.

Larry Moran said...

Corneel asks,

I admit, that´s not really an answer. But still, can you suggest an alternative scenario that causes demographic and reproductive ageing to proceed at a different pace?

What's the problem? There's nothing in evolutionary theory that I know of that says you have to keep reproducing up until the day you die.

I notice that you didn't answer my question. Do you think that in the past there were many women who had children after they turned 50? That's what you have to assume if you think that menopause was adaptive.

Furthermore, you have to assume that the women had no choice in the matter. The only way to get them to stop having children was to evolve a mechanism that forced them to become sterile at 50.

Anonymous said...

What is the selective advantage of extending menopause? If a woman can bear a child every two years and starts at 14, she would have mothered 18 children by the time she hits 50. She'd get an extra 10 if she could delay menopause until 70. That sounds like a big advantage until you try to find women who actually have had 10 or 20 children. They are rare, and the number who have that many children, all of whom have had children of their own, are even rarer. The bottleneck is not the ability to conceive children. The bottleneck is elsewhere. An extra 20 years of fertility is not likely to be as much as an advantage as naive analysis suggests.

Timothy V Reeves said...

OK let's assume the menopause mutation is neutral. What then happens to the unmutated gene? Does it die out by chance?

Random drift seems to have the concomitant of a spread or mix of traits - perhaps the heterosexual/homosexual mix is a sign of that?

I think we have been here before. See
http://sandwalk.blogspot.com/2008/01/teaching-idiots-about-evolution.html

Larry Moran said...

Timothy V Reeves asks,

OK let's assume the menopause mutation is neutral. What then happens to the unmutated gene? Does it die out by chance?

Yes. That's what random genetic drift does. It fixes alleles in a population by chance.

Joe Pickrell said...

presumably relevant:
http://biology.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pbio.0040007

"Rather than showing the direct effects of selection characterizing other life-history traits, post-reproductive lifespan in these fish appears to be a random add-on at the end of the life history."

DiscoveredJoys said...

I believe that it is significant that the age of the menopause is so similar across all the human populations.

It suggests to me that the age of the menopause is not adaptive in itself but constrained by natural selection to a narrow range.

I don't have any data (I know, I know) but I expect natural selection will tend to 'optimise' the age of the menopause - too soon and not enough copies of the woman's genes will be passed into the later generations. Too late, and although the woman may bear more children, but more will fail to breed - either because the children will be less fit because of poor quality 'old DNA' or because they, and the earlier children, will suffer from parental care spread too thinly.

On the other hand, the onset of puberty does appear to react quickly to environmental factors...

Anonymous said...

“What if menopause is neutral with respect to evolution or even maladaptive?”

You ask "What if." But aren't you assuming that it is not adaptive? Why not test whether or not it is adaptive, or maladaptive vs neutral? Why assume, and assert, that it is non- or maladaptive has been fixed by drift?

If there are adaptationist just-so stories, are there not also driftist just-so stories?

Why not develop testable hypotheses and investigate the range of viable explanations, both ultimate and proximate? Why deride every approach but one's own favored untested and hunch-driven assumptions?

Tupaia

Anonymous said...

Tupaia: "... are there not also driftist just-so stories? ... Why deride every approach but one's own favored untested and hunch-driven assumptions?"

You must be new here. :)

Larry Moran said...

anonymous says,

You ask "What if." But aren't you assuming that it is not adaptive? Why not test whether or not it is adaptive, or maladaptive vs neutral? Why assume, and assert, that it is non- or maladaptive has been fixed by drift?

You raise some good points.

Since we don't know whether menopause is neutral, adaptive, or maladaptive, it would seem like a good idea to reserve judgment until we have evidence, right?

You must be just as mad as I am over articles that make the uncritical assumption that menopause must be adaptive just because it exists.

Please join me in the fight against bad science.

Carlo said...

Can I ask a few questions (following a statement)? In the field of evolutionary biology, particularly molecular evolution, Kimura's Neutral Theory has become the de-facto standard against which hypotheses about selection must be tested - this is because of neutral evolution's excellent explanatory power.

Why is it that adaptationist human evolution fields (e.g., Darwinian Anthropology, Evolutionary Psychology) apply 'evolutionary theory' as though this fundamental revolution in our understanding of how features can change over time had never happened? Why is it that when our lab submits papers showing rigorous evidence of selection at the genic level, our reviewers are always mindful of our having used proper statistics in rejecting the null hypothesis of neutral evolution - yet few of these 'just so' stories even acknowledge the possibility of neutrality?

Anonymous said...

Actually, my own hunch is that menopause is not itself an adaptation but is a developmentally linked side effect of other processes that were positively selected during human evolution, perhaps related to heterochronic change.

But I don't know that for sure. That's why I think that the scientifically rigorous approach is to make predictions and test hypotheses rather than assert default assumptions.

Tupaia

Anonymous said...

Generic driftist just-so story:

There once was a complex, salient yet phenotypically neutral trait that just so happened to be fixed in a populations. The end.

How do I know that the trait is neutral? Just take my word on it.

Generic constraintist just-so Story:

There once was a complex, salient trait that just so happened to be a spandrel. The end. The end as in no need to test some frivolous adaptationist just-so story.

How do I know it's a spandel? Because it sounds more sophisticated than a vulgar adaptation.

The perils of dismissing adaptationism: how the whale got its flipper tubercles.

http://www.sciencedaily.com/releases/2004/05/040512044455.htm

Without this research, it would be very tempting to assume that tubercules on the humpback whale's flippers are neutral phenotypic variants of no adaptive significance. How Gouldian that would be. But clearly not the case.

(Sure, flipper tubercles are probably not "optimal" in an idealized engineering sense. After all, they had to develop out of a preexisting structure. Well, no duh.)

Tupaia

Anonymous said...

To be fair, there is definitely a lot of mediocre pop science journalism about one academic or another's half-assed theory. Heck, even in scientific meetings you see folks trying to hype their overblown interpretations of their own meager data.

So, point taken. But it doesn't just apply to adaptationists.

Tupaia

Larry Moran said...

Carlos says,

Can I ask a few questions (following a statement)? In the field of evolutionary biology, particularly molecular evolution, Kimura's Neutral Theory has become the de-facto standard against which hypotheses about selection must be tested - this is because of neutral evolution's excellent explanatory power.

Could I suggest a slight modification of this statement. It's not "Kimura's Neutral Theory" that has become the de facto standard. It's evolution by random genetic drift.

It's true that neutral alleles are fixed by random genetic drift so there's a connection between Neutral Theory and drift. But there's more to the story than that.

When you understand drift you also understand that beneficial alleles are not necessarily fixed by natural selection and you understand that deleterious alleles can be fixed by drift.

It's better the say that the default explanation is random genetic drift and not "neutral theory."

Carlo said...

Thanks! Duly noted.