No animal compares to humans when it comes to studying populations over time. Easy to track and occasionally living in relative isolation, Homo sapiens is the only species that keeps detailed records. That is why biologist Virpi Lummaa of the University of Sheffield in England started in 1998 to comb through Finnish church records from two centuries ago for clues about the influence of evolution on reproduction.

"I always wanted to work on primates," Lummaa says. "But if I wanted to collect a similar data set on wild chimps, I would be struggling. I've decided to study another primate in the end."

The 33-year-old Finnish biologist, aided by genealogists, has pored through centuries-old tomes (and microfiche) for birth, marriage and death records, which ended up providing glimpses of evolution at work in humanity's recent ancestors. Among them: that male twins disrupt the mating potential of their female siblings by prenatally rendering them more masculine; mothers of sons die sooner than those of daughters, because rearing the former takes a greater toll; and grandmothers are important to the survival of grandchildren. "I'm trying to understand human reproductive behavior from an evolutionary perspective," Lummaa says.

Most recently, Lummaa and her colleagues studied the effect of males on their female twins. Of 754 twins born between 1734 and 1888 in five towns in rural Finland, girls from mixed-gender pairs proved 25 percent less likely to have children, had at least two fewer children, and were more than 15 percent less likely to marry than those born with a sister. This impact remained the same regardless of social class and other cultural factors and even if the male twin died within three months of birth, leaving the female twin to be reared as if she was an only child, the researchers reported in Proceedings of the National Academy of Sciences USA.

Lummaa speculates that the findings may be prompted by the male hormone testosterone crossing the womb and masculinizing the female twin, as has been seen in animal studies. Whatever the cause, there's no question of the outcome: Data shows that mothers of opposite-sex twins end up with 19 percent fewer grandchildren than moms of same-sex twins, meaning evolution would seem to favor the latter.

The results are somewhat puzzling, says Ken Weiss, biological anthropologist and geneticist at Pennsylvania State University, noting that "if twinning is genetic, then there should be a slight selection bias against it, so that twinning would be kept rare. But some animals twin routinely." Given the seeming conflict, he says, "there are dangers in over-interpreting 'fitness' effects, even if the observation is correct."

Modern medicine and nutrition tend to obscure these kinds of results as well, hence the need to go back to the preindustrial Finns, before the advent of birth control and the easing of periodic famine and high child mortality rates. "It's almost a shock when you realize that 100 to 150 years ago, 40 percent of babies died before they reached adulthood," even when adulthood was defined as age 15, Lummaa notes.

"In the absence of cultural practices such as contraceptives and assisted reproduction, humans are subject to the same evolutionary forces as are other organisms," says biologist Tobias Uller of the University of Wollongong in Australia. "Given that Virpi's data is extraordinarily detailed compared to what we have available for most other animals, the human data can profitably be used to address key issues in evolutionary theory."

The evolutionary biologist has also used this historical data set to ponder the conundrum of grandmothers. That is, why human women often live long after they are able to reproduce (on average around the age of 50), unlike almost all other animals. "If your ultimate purpose in life was to create as many offspring as possible or pass off as many genes," Lummaa says, "it's kind of strange that human women stop halfway."

One possible explanation is that having a grandmother around somehow improves the reproductive potential of her grandchildren. In fact, that is exactly what the researchers found when they reviewed stats on 537 Finnish women who had a combined total of 6,002 grandchildren. Adding in data from more than 3,000 French Canadians (who had a modest 100,074 grandchildren) confirmed that having grandma around to help enabled younger women to have more children sooner and with improved chances of surviving into adulthood. "That suggests that perhaps one reason why women do carry on living is because they are able to help," Lummaa says.

Of course, studying humans requires teasing out the confounding cultural effects. For example, the Finnish data indicated that child mortality was much higher in mainland towns than on the islands of Finland's Archipelago Sea. This can be traced back to the fact that mainland women were responsible for farm work, leading to earlier replacement of mother's milk with cow's milk. "That led to infections," Lummaa notes. "In the archipelago this was not the case." Birth rates in both areas also tended to cluster roughly nine months after the period when Finns traditionally married: after the fall harvest.

Studying humans has other pitfalls, most notably that it's very easy to become involved with your subjects, Lummaa says. "We have thousands of people. I can't say I know every one of them but there are some families who pop out," she recalls. "One woman had 18 children and every single one died before adulthood while she lived into her 90s without any of these children." She adds: "If you are studying humans, you can't help feeling more connected to whatever you find out."

Lummaa is learning that first hand these days, having recently given birth to a three-month-old son, Eelis. "It's your own child, you can't have a scientific attitude," she admits, "but you are thinking, 'Well, what in the patterns I see is genetic and is it coming from the mum or dad?' I'm always trying to see my parents' traits in my son." She is thrilled, of course, though her research warns it bodes ill for her life expectancy. Premodern Finnish mothers among the Sami people (famous for their reindeer herding) who bore sons had shorter life spans than those who gave birth to daughters. This has to do with birth weight—male babies are typically larger—but also with that dreaded male hormone, testosterone. "Testosterone can compromise your immune system, it can affect your health," Lummaa says. "Boys are a little bit more costly than girls" to raise, because they drain more physical resources from their mothers. Sons also are not as likely as daughters to stick around to help their mothers out later in life.

Fortunately for Lummaa, she has the benefit of modern medicine. But "I can certainly see that it's taking a lot of energy and I'm sure it's aging me," she says, chuckling. "How on earth these women managed to give birth every year is truly amazing."

Lummaa has now turned her attention to the effect of grandfathers on grandchildren. If grandmothers improve survival odds, what do elderly males contribute? "If anything there's a negative effect," she says. This could be because of the cultural tradition of catering to men, particularly old men. "Maybe if you had an old grandpa, he was eating your food," she speculates. Or it could be that because men can continue to reproduce, they are less vested in anyone other than their own children. Another possible reason is that women can be sure that a grandchild is their genetic descendant, but it is more difficult for grandfathers. This may also have spurred them to seek second and even third wives rather than focusing on their children. "We are comparing men who married once in their lifetime[s] with men who are married several times," Lummaa says.

Lummaa is not alone in using human history to try to enhance evolutionary understanding. A recent study by ethnologist Dustin Penn of the Austrian Academy of Sciences in Vienna and population scientist Ken Smith of the University of Utah, using Mormon church records from the 19th century, discovered that having more children upped women's chances of dying prematurely. Anthropologist Kristen Hawkes, also at Utah, reached the same conclusions as Lummaa about the utility of grandmothers after studying populations of human hunter-gatherers in Africa and South America. "It's most interesting to find out what's causing the differences between human populations," Lummaa says. "How do those general evolutionary theories actually explain the patterns we see in humans? And how much is due to other reasons?" As Lummaa says, "We've got more data than we've got time to analyze." Meaning Lummaa, her colleagues and her scientific descendants will have plenty to study until she is a grandmother herself.