Swedish records stretching back to 1751 have allowed a couple of statisticians to put to the test competing arguments about why fewer male babies are born in times of stress. Previous research into the ratio of male to female infants after events ranging from London's killer smog in the 1950s to the earthquake that rocked Kobe, Japan, in 1995, showed this to be the case. Some scientists argued that a pregnant woman's stress response affects all male fetuses, damaging them for life. Others reasoned that a pregnant woman's stress response merely winnows out those babies unlikely to themselves reproduce during difficult conditions.

"It's better to have a female than a male in stressful times," explains statistician Ralph Catalano of the University of California Berkeley. He argues that weak males are unlikely to survive to reproductive age or, if they do, are unlikely to be able to win mates over more robust males. "If you have a daughter, [her] reproductive success is not contingent on robustness because males are not as picky," he adds.

Under stress--whether environmental or psychological--human beings release the hormone cortisol, which helps prepare the body for the proverbial fight or flight. This hormone can cross the placenta into the developing baby. But researchers had failed to hit on a method that could resolve the conflicting theories for why stressed mothers give birth to fewer boys. Catalano and Tim Bruckner of U.C. Berkeley realized that the two theories made different predictions for the long term viability of male babies. If the cortisol response damages all baby boys equally, then in times of relative ease the average male life expectancy should go up. If, on the other hand, the cortisol response is preventing weak males from being born, then average life expectancy for men should go down in times of relative ease because, overall, more weak men are being born.

Looking at records from 1751 to 1912--the last group of Swedes born whose overall lifespan can be reliably calculated--the researchers found that there was an inverse relationship between the sex ratio and life expectancy: "When you have more males, those males don't live as long," Catalano explains. "When you get fewer males, they tend to live longer." This supports the hypothesis that stress just prevents weaker male fetuses from being carried to term, the statisticians argue in a paper presenting the findings in the current issue of the Proceedings of the National Academies of Science.

This inverse relationship persisted as the statisticians corrected for various possible problems, such as the long-term rise in human longevity, seasonality and even the better record keeping since 1861. But it is still just information from one country that may not be broadly applicable. Similar data from Denmark, England and Wales, which has recently become available, will be critical. "This has to be replicated elsewhere," Catalano says. "We'll see what the results of those studies are."

Nevertheless, this finding may help scientists get to the root of another question: why humans evolved this stress mechanism in the first place. "One argument is that it was conserved to help us lift buses off of pedestrians who just got hit," Catalano remarks. "My belief is that nature conserved the stress mechanism because of its effect on gestation. Natural selection doesn't care about individuals, it just wants poundage and conservation of genes. It found a way to maximize the pounds and the genes."