Sex ratios: Bad times wallop extra sons.

Nearly 30 years of deer-watching on a Scottish island have revealed a quirk of male-female sex ratios. It may explain why a famous biological hypothesis has been maddeningly hard to prove.

The Trivers-Willard hypothesis, named for the scientists who devised it in 1973, predicts that to maximize their chances for heirs, the flossiest, fittest females in some species give birth to more sons than daughters. Struggling, underfed moms should bear extra females.

In decades of searching for this effect, "there have been a lot of contradictory results," says Loeske E.B. Kruuk at the University of Edinburgh. In the June 3 NATURE, she and her colleagues report that the Trivers-Willard effect once seen among red deer on the Isle of Rum disappeared in one area where culling stopped and the population tripled. Crowding and harsh weather, argue the scientists, killed more male than female fetuses, canceling the predicted effect--that top moms would conceive extra sons.

The mechanism could be simple. Perhaps demands of the bigger, faster-growing male fetuses are more likely to overwhelm the resources of an already overstressed mom than the smaller demands of a female fetus are, Kruuk speculates.

Whatever the cause, this male vulnerability could be confounding studies of the Trivers-Willard hypothesis, Kruuk suggests. The only documentation for its effects come from populations not stressed by overcrowding, she points out.

The Trivers-Willard hypothesis applies to species in which a mother's condition affects her sons' breeding success more than her daughters'. Among red deer, for example, a female in glowing health tends to bear robust sons that grow into big males defending large harems. These sons reproduce more prolifically than daughters, who bear one calf a season, even when in the best of health.

If a weak female produces a runty son, he may not mate at all because the big guys monopolize the females. For fit moms, sons offer a bonus. For frailer moms, a daughter is a safer bet.

Just what internal mechanism could skew sex ratios has fueled considerable speculation. Current views suggest that the mother's hormonal state affects whether more male or female zygotes implant in her uterus, Kruuk says.

In the same issue of NATURE, Andrew Cockburn of the Australian National University in Canberra praises the "superb long-term data" from Rum. Fully demonstrating the Trivers-Willard hypothesis in wild animals has been "pretty difficult," he told SCIENCE NEWS. He notes that the first success was on Rum, where biologist Fiona E. Guinness of the University of Cambridge in England learned to recognize hundreds of deer by sight.

The idea that bad times slam sons more than daughters "has been widely seen as an alternative to the Trivers-Willard hypothesis," Cockburn says. Researchers had mused that the factors emphasized by the two views might work together, but teasing out interactions seemed as daunting as proving the original hypothesis. "That was an area that inspired a bit of hopelessness," he says.

After the latest analysis, Cockburn predicts, "people will see the two hypotheses can be married."