I'm so glad you asked. Seriously. An answer to this question is one of the most illuminating and disturbing stories in human evolutionary biology, and almost nobody knows about it. And so, O my friends, gather close, and hear the extraordinary tale of:
HOW THE WOMAN GOT HER PERIOD
Contrary to popular belief, most mammals do not menstruate. In fact, it's a feature exclusive to the higher primates and certain bats*. What's more, modern women menstruate vastly more than any other animal. And it's bloody stupid (sorry). A shameful waste of nutrients, disabling, and a dead giveaway to any nearby predators. To understand why we do it, you must first understand that you have been lied to, throughout your life, about the most intimate relationship you will ever experience: the mother-fetus bond.
Isn't pregnancy beautiful? Look at any book about it. There's the future mother, one hand resting gently on her belly. Her eyes misty with love and wonder. You sense she will do anything to nurture and protect this baby. And when you flip open the book, you read more about this glorious symbiosis, the absolute altruism of female physiology designing a perfect environment for the growth of her child.
If you've actually been pregnant, you might know that the real story has some wrinkles. Those moments of sheer unadulterated altruism exist, but they're interspersed with weeks or months of overwhelming nausea, exhaustion, crippling backache, incontinence, blood pressure issues and anxiety that you'll be among the 15% of women who experience life-threatening complications.
From the perspective of most mammals, this is just crazy. Most mammals sail through pregnancy quite cheerfully, dodging predators and catching prey, even if they're delivering litters of 12. So what makes us so special? The answer lies in the way human embryos develop.
In many mammals, the placenta, which is part of the fetus, just interfaces with the surface of the mother's blood vessels, allowing nutrients to cross to the little darling. Marsupials don't even let their fetuses get to the blood: they merely secrete a sort of milk through the uterine wall. However, other mammal groups, including the higher primates, have retained a more direct connection, termed a hemochorial placenta. Among humans, chimpanzees and gorillas, its development is especially invasive.
Inside the uterus we have a thick layer of endometrial tissue, which contains only tiny blood vessels. The endometrium seals off our main blood supply from the newly implanted embryo. The growing placenta literally burrows through this layer, rips into arterial walls and re-wires them to channel blood straight to the hungry embryo. It delves deep into the surrounding tissues, razes them and pumps the arteries full of hormones so they expand into the space created. It paralyzes these arteries so the mother cannot even constrict them.
What this means is that the growing fetus has direct, unrestricted access to its mother's blood supply. It can manufacture hormones and use them to manipulate her. It can, for instance, increase her blood sugar, dilate her arteries, and inflate her blood pressure to provide itself with more nutrients. And it does. Some fetal cells find their way through the placenta and into the mother's bloodstream. They will grow in her blood and organs, and even in her brain, for the rest of her life, making her a genetic chimera**.
This might seem rather disrespectful. In fact, it's sibling rivalry at its evolutionary best. You see, mother and fetus have quite distinct evolutionary interests. The mother 'wants' to dedicate approximately equal resources to all her surviving children, including possible future children, and none to those who will die. The fetus 'wants' to survive, and take as much as it can get. (The quotes are to indicate that this isn't about what they consciously want, but about what evolution tends to optimize.)
There's also a third player here – the father, whose interests align still less with the mother's because her other offspring may not be his. Through a process called genomic imprinting, certain fetal genes inherited from the father can activate in the placenta. These genes ruthlessly promote the welfare of the offspring at the mother's expense.
How did we come to acquire this ravenous hemochorial placenta and aggressive implantation process, which give our fetuses and their fathers such power? The full answer is lost in the mists of time. Uteri do not fossilize well. However, evolutionary trees suggest that the hemochorial placenta is an ancestral trait among mammals that many groups have discarded. Not ours though.
It’s possible that allowing fetuses to drink from the (blood) firehose may help with brain development. Brains are energetically expensive, so it might not be a coincidence that some of the species with the most extensive placental invasion (humans, chimps and gorillas) also have some of the biggest brains. On the other hand, non-invasive placentae can also be highly efficient, so this is not entirely convincing.
The consequences seem clearer than the causes. Mammalian pregnancy is a well-ordered affair when the mother is a despot. Her offspring live or die at her will; she controls their nutrient supply, and she can expel or reabsorb them any time. Pregnancy with an invasive hemochorial placenta, on the other hand, is run by committee – and not just any committee, but one whose members can have very different, competing interests and share only partial information.
It's a tug-of-war that, in some situations, can deteriorate to a tussle and, occasionally, to outright warfare. Many potentially lethal disorders, such as ectopic pregnancy, gestational diabetes, and pre-eclampsia can be traced to mis-steps in this intimate game.
What does all this have to do with menstruation? We're getting there.
From a female perspective, pregnancy is a huge investment, and the more invasive the placenta, the bigger that investment. When a human embryo implants, its mother not only loses full control of her own hormones, she also risks hemorrhage when it comes out. So it makes sense that females want to screen embryos carefully. Going through pregnancy with a weak, inviable or even sub-par fetus isn't worth it. And the more costly and risky pregnancy becomes, the more conservative females should be in this respect.
That's where the endometrium comes in. You've probably read about how the endometrium is this snuggly, welcoming environment just waiting to enfold the delicate young embryo in its nurturing embrace. In fact, it's quite the reverse. Researchers, bless their curious little hearts, have tried to implant embryos all over the bodies of mice (who also have hemochorial placentae, though theirs are much less invasive than ours). The single most difficult place for them to grow was – the endometrium.
Far from offering a nurturing embrace, the endometrium is a lethal testing-ground which only the toughest embryos survive. The longer the female can delay that placenta reaching her bloodstream, the longer she has to decide if she wants to dispose of this embryo without significant cost. The embryo, in contrast, wants to implant its placenta as quickly as possible, both to obtain access to its mother's rich blood, and to increase her stake in its survival. For this reason, the endometrium got thicker and tougher – and the fetal placenta got correspondingly more aggressive.
But this development poses a further problem: what to do when the embryo died or was stuck half-alive in the uterus? The blood supply to the endometrial surface must be restricted, or the embryo would simply attach the placenta there. But restricting the blood supply makes the tissue weakly responsive to hormonal signals from the mother – and potentially more responsive to signals from nearby embryos, who naturally would like to persuade the endometrium to be more friendly. In addition, this makes it vulnerable to infection, especially when it already contains dead and dying tissues.
Many mammals, including lemurs and lorises among primates, have solved this problem by evolving ways to share nutrients without deep invasion of the mother’s blood supply. But - possibly because restricting blood flow to the fetus never became adaptive in a lineage with very hungry brains - not the higher primates.
It appears that the solution, for higher primates, was instead to slough off the whole superficial endometrium – dying embryos and all – after every ovulation that didn't result in a healthy pregnancy. It's not exactly brilliant, but it works, and most importantly, it's easily achieved by making some alterations to a chemical pathway normally used by the fetus during pregnancy.
In other words, it's just the kind of effect natural selection is renowned for: odd, hackish solutions that work to solve proximate problems. It's not quite as bad as it seems, because in nature, women would experience periods quite rarely – perhaps no more than a few tens of times in their lives between lactational amenorrhea and pregnancies***.
Humans invest more in pregnancy than even chimpanzees and gorillas. So it makes sense that we should also invest in more stringent quality control for embryos. As with all evolutionary story-telling, it’s dangerous to assume causality here. But it certainly looks like our unusually heavy menstruation is part of this stringent quality control.
Links / References:
The evolution of menstruation: A new model for genetic assimilation
Genetic conflicts in human pregnancy.
Menstruation: a nonadaptive consequence of uterin... [Q Rev Biol. 1998]
The role of invasive trophoblast in implantation and placentation of primates
References for the mouse implantation studies:
Runner, M. N. (1947) Development of mouse eggs in the anterior chamber of the eye. Anatomical Record 98: 1-17.
Kirby, D. R. S. (1965) The "invasiveness" of the trophoblast. Pages 68-73, in W. W. Park (ed.), The early conceptus, normal and abnormal. University of St. Andrews, St. Andrews.
* mouse blastocysts implanted in kidney, spleen, testis, brain, liver
McLaren, A. (1965) Maternal factors in nidation. Pages 27-33, in W. W. Park (ed.), The early conceptus, normal and abnormal. University of St. Andrews, St. Andrews.
"The fact that successful implantation and subsequent development can occur in the testis emphasises the most striking and, to me, unexpected feature of implantation outside the uterus, namely, that it is entirely independent of the sex or hormonal status of the host. Pregnant females, non-pregnant females and males all provided an apparently equally good substrate for implantation in Kirby's experiments …"
"What a contrast this makes to the situation inside the uterus! The experiments so far might lead one to think that the rodent blastocyst's capacity to implant and develop was so strong that only the most unfavourable of circumstances could stop it. Yet the uterus, which one might picture as especially welcoming and receptive, is in fact so inimical to implantation that there is only a brief period of time during which a blastocyst placed within it can hope to survive."
Credits: During my pregnancy I was privileged to audit a class at Harvard University by the eminent Professor David Haig, whose insight underlies much of this research. Thanks also to Quora User, who reminded me of crucial details. All errors are mine alone.
*Dogs undergo vaginal bleeding, but do not menstruate. Elephant shrews were previously thought to menstruate, but it's now believed that these events were most likely spontaneous abortions.
** Scientists Discover Children’s Cells Living in Mothers’ Brains (Thanks to Quora User for the link).
***One older published estimate for hunter gatherers was around 50, but this relied on several assumptions, including 3 whole years of menstruation before reproduction (36 periods) for no obvious reason.
A study of the Dogon of Mali based on 57 women in natural-fertility cycles estimates the median number of life-time menses at 109: http://www.jstor.org/stable/2744446
To get a feel for life history parameters, we can look at data from the Hadza of Tanzania, who reach puberty around 18, bear an average of 6.2 children in their lives (plus 2-3 noticeable miscarriages) starting at 19, and go through menopause at about 43 if they survive that long (about 50% don't). Around 20% of babies die in their first year; the remainder breastfeed for about 4 years. So this is 25 years of reproductive life, of which about 20 are spent lactating, and 4.5 pregnant. That would leave only about 6 periods, but amenorrhoea would cease during the last year of lactation for each child, so this figure is too low. On the other hand, this calculation ignores the ~50% of women who died before menopause, miscarriages, months spent breastfeeding infants who would die, and periods of food scarcity, all of which would further reduce lifetime menstruation. Stats from: http://www.fas.harvard.edu/%7Ehbe-lab/acrobatfiles/who%20tends%20hadza%20children.pdf