Humans are pretty odd animals. Any list of our odd features would be a long one, but menopause would have to come in right near the top.

Where men’s capacity to reproduce diminishes steadily at about the same rate as our bodies age, women stop reproducing around the age of 50. That’s well before the diseases of ageing really kick in.

The fact that women stop producing babies when they still have many years to live challenges a simple reading of evolutionary theory. Surely women who kept on reproducing would have left more descendants – all bearing their genes for late-life mothering?

A lot of evolutionary thinking about menopause centres on the costs of dying in childbirth and the benefits provided by grandmothers. By stopping reproduction and thus living long enough to help raise their grandchildren, grandmothers contribute to their ongoing evolutionary fitness.

But can they contribute enough to make it worth foregoing their own reproduction? Especially since grandchildren each carry only a quarter of their grandmother’s genes, whereas sons and daughters carry half their mother’s genes. The genetic calculations suggest women should keep on making babies of their own rather than retiring to become full-time grannies.

But theoretic modelling by Michael Cant and Rufus Johnstone suggested in 2008 that conflict within the family provides the missing ingredient in our understanding of menopause. A study published today in Ecology Letters provides compelling support for this idea.

When a woman’s first children grow up and start to produce children of their own, those adult children compete for the same resources that the woman might herself use to produce more babies of her own.

Mirkka Lahdenpera, Duncan Gillespie, Virpi Lummaa and Andrew Russell analysed 200 years of data from Lutheran parish records in pre-industrial Finland. I have written before about the extraordinary work done by Lummaa and her collaborators, testing ideas about evolution in historic populations. In this case, they studied records of 653 women born between 1702 and 1823 who had 4703 offspring and 9164 grandchildren in total.

For these women, there were a relatively small number who, between their mid-thirties and menopause, were giving birth to children while their older children were already reproducing. Reproducing at the same time as a daughter didn’t reduce the survival of either of the newborns (although a study in contemporary rural Gambia found an effect of this type). But reproducing at the same time as a daughter-in-law reduced the survival of both newborns by as much as 66 percent.

Resources were scarce in these populations, as they have been for most people throughout most of history. When the mother and her daughter-in-law have to compete for those resources, both end up faring worse. The daughter-in-law has no genetic interest in the mother’s offspring’s success. Unlike the mother’s daughter who is as closely related to the mother’s new offspring (her full sibling if the mother was still married to her father) as she is to her own babies.

When the genetic relationships and the costs of conflict are combined with the benefits provided by grandmothers and the chances of dying in late-life childbirth, the evolutionary models suggest that it makes plenty of sense for women to retire from childbearing and get busy grandmothering. In evolutionary terms at least, conflict with younger women forces mothers-in-law into menopause.

This is just one of the many evolutionary conflicts that permeate family life. I hope to return to that special relationship between mother and daughter-in-law in the near future.

Comments and Tweets appreciated as always. If you tweet, include @Brooks_Rob