Nature is unsentimental. In Herbert Spencer's famous phrase, it is a question of the survival of the fittest. You are born. If you are lucky, you reproduce. Then you die. Indeed, once you can no longer reproduce you are, evolutionarily speaking, dead anyway. That, at least, is the conventional wisdom.

So it has been a bit of a mystery why women, who normally become infertile in their mid-40s with the arrival of the menopause, often live on and on. Indeed, the average lifespan of women in almost all societies exceeds that of men¡ªand most men remain fertile well into old age.

The presumed explanation depends on a more sophisticated understanding of what ¡°fittest¡± means. It is not enough to survive and have children. Your line has to continue through your grandchildren. In this context, the survival of women beyond menopause makes sense if it translates into extra grandchildren. Proving that hypothesis, though, is difficult. But Mirkka Lahdenpera, of the University of Turku, in Finland, and her colleagues, think they have managed to do so. Their work, just published in Nature, draws on records made in the 18th and 19th centuries in Finland and Canada. The results are striking. In both countries, a woman gained two extra grandchildren for every decade she survived beyond the age of 50.

The post-reproductive survival of humans women in particular is truly unusual. Non-reproductive helpers of individuals who are breeding are found in many species. But they are usually young animals that have yet to establish themselves, rather than relics from previous generations. The post-reproductive elderly just die. Chimpanzees, for example, have a similar pattern of fertility to people. A female chimp's fertility peaks in her late 20s, and is more or less extinguished by her mid-40s. But in chimpanzees, mortality rises as fertility declines.

Nor, despite the increase in average lifespan in the rich world over the past few centuries, is post-reproductive survival a modern phenomenon. Until very recently, most of the increase has been due to better survival in childhood and youth, rather than the increased prolongation of middle age into old age. Even in pre-industrial populations, around a third of women were over 45.

All this suggests that the prolonged old age of people women in particular is indeed an evolved phenomenon, rather than being a consequence of better living conditions. And the quality of the records Dr Lahdenpera was working with allowed her to dig into the details of how this came about.

First, it was nothing to do with a long-lived woman having had more offspring in the first place, and the number of her grandchildren thus merely being a function of the number of her children. Nor was there any detectable effect due to the socio-economic status of the women involved. Rich or poor, high- or low-born, all showed the same trend. And, contrary to the beliefs of some workers in the field, both sons and daughters benefited from the survival of a matriarch, not merely daughters alone.

Nor was the increase due to a single effect. Instead, Dr Lahdenpera was able to disentangle a range of beneficial phenomena which, when added together, resulted in the increased reproductive success that she observed. People whose mothers were still alive both gave birth to more offspring and raised a higher proportion of those offspring to adulthood. They also gave birth to their first children at a younger age than those whose mothers had died, and the births of their children were more closely spaced. All of these things contribute to reproductive output.

Moreover, the physical presence of the matriarch was vital. Children who lived more than 20km from their mothers produced significantly fewer offspring than those who lived in the same village. That suggests the increase in the number of grandchildren was due not to some subtle genetic effect, but rather to help whether physical or in the form of advice that matriarchs were contributing.

Perhaps the most evolutionarily significant finding, though, is the age at which the matriarchs in the study died. The average lifespan of post-menopausal Finns was 68. Of Canadians, it was 74. These ages correspond to the points where the matriarch's children themselves had stopped reproducing. At that point, a woman's fitness plummets. And so, sadly, does her life expectancy.

(Agencies via Xinhua  March 30, 2004)