1637. Why Some Mammal Species Kill Their Infants?

By Carl Zimmer, The New York Times, November 13, 2014

In the early 1970s, Sarah Blaffer Hrdy, then a graduate student at Harvard, traveled to India to study Hanuman langurs, monkeys that live in troops, each made up of several females and a male.

From time to time, Dr. Hrdy observed a male invade a troop, driving off the patriarch. And sometimes the new male performed a particularly disturbing act of violence. He attacked the troop’s infants.

There had been earlier reports of infanticide by adult male mammals, but scientists mostly dismissed the behavior as an unimportant pathology. But in 1974, Dr. Hrdy made a provocative counter proposal: infanticide, she said, is the product of mammalian evolution. By killing off babies of other fathers, a male improves his chances of having more of his own offspring.

Dr. Hrdy went on to become a professor at the University of California, Davis, and over the years she broadened her analysis, arguing that infanticide might well be a common feature of mammalian life. She spurred generations of scientists to document the behavior in hundreds of species.

“She’s the goddess of all this stuff,” said Kit Opie, a primatologist at University College London.

Forty years after Dr. Hrdy’s initial proposal, two evolutionary biologists at the University of Cambridge have surveyed the evolution of infanticide across all mammals. In a paper published Thursday in Science, the scientists concluded that only certain conditions favor the evolution of infanticide — the conditions that Dr. Hrdy had originally proposed.

“My main comment is, ‘Well done,'” said Dr. Hrdy. She said the study was particularly noteworthy for its scope, ranging from opossum to lions.

The authors of the new study, Dieter Lukas and Elise Huchard, started by plowing through the scientific literature, looking for evidence of infanticide in a variety of mammalian species. The researchers ended up with data on 260 species, and in 119 of them — over 45 percent — males had been observed killing unrelated young animals.

Dr. Lukas and Dr. Huchard then looked at how infanticide might have evolved over the past 160 million years of mammal evolution. “When we started, we weren’t sure if infanticide was present in some ancestral mammal and is just more pronounced in some species and is lost in other species,” said Dr. Lukas. “Or maybe it just evolved in those species where the conditions were right.”

Their research supports the second scenario. The common ancestor of living mammals did not practice infanticide: The behavior evolved later and independently in a number of separate animal lineages.

Dr. Lukas and Dr. Huchard found that the species that evolved infanticide had a few things in common. For instance, the females tended to give birth year-round, rather than at just one time of the year. Another key factor was that males and females lived in groups where females greatly outnumber males.

The scientists concluded that conditions such as these fostered the evolution of infanticide. If only a few males get to mate with females, there will be a lot of other males facing the prospect of dying childless. Natural selection will favor males that can take over groups of females.

But these males are pressed for time. It won’t be too long before another male threatens them with exile. Meanwhile, many females will be busy nursing infants. Killing the offspring of other males can free up females for reproduction and widen the window of opportunity for new males, leading to more offspring for them.

The researchers also noted that infanticide generally did not evolve in species where females only give birth once a year. “There’s no sense for a male to kill the offspring in the previous year, because he has to wait anyway,” said Dr. Lukas.

Dr. Hrdy and other early infanticide researchers not only investigated how infanticide might have evolved. They also suggested that mothers and fathers had evolved defenses against infanticidal males.

Dr. Lukas and Dr. Huchard found support for this idea, too. In species in which infanticide appeared, females gradually mate with more and more males. This behavior may protect infants by introducing uncertainty: Males might be more reluctant to kill infants if one of them could be his own.

“In effect, whenever promiscuity is high enough, it does not pay for males to commit infanticide,” said Carel P. Van Schaik, a primatologist at the University of Zurich. “This new study beautifully confirms the major role of sexual behavior.”

Despite its sweep, , the new study doesn’t settle every debate over infanticide. Dr. Van Schaik, for example, suspects that the true number of infanticidal species is higher. In some of them, defenses against the practice have become so effective that infanticide has become rare, and therefore hard for scientists to observe.

The new study also clashes with earlier work finding that infanticide altered the social structure of some species. Last year, for example, Dr. Opie and his colleagues reported that monogamy — in which one male and one female lived together — has evolved in primates as a defense against infanticide.

In their new study, Dr. Lukas and Dr. Huchard found no such pattern. Dr. Opie believes that is partly because they didn’t take into account the full diversity of how mammals mate.

“What we’re saying is that infanticide is driving monogamy, and we don’t think anything they’ve shown in this paper contradicts that,” said Dr. Opie.

Among the mammal species that Dr. Lukas and Dr. Huchard didn’t include in their study was our own. Dr. Lukas said the cultural diversity of humans means it is hard to make blanket statements about them. “For many social behaviors, it’s very difficult to define a human universal,” said Dr. Lukas.

That’s not to say that adults never harm — or even kill — children. But Dr. Hrdy argued that the infanticide she documented in langurs — in which males stalked babies for days —was fundamentally different from what happens in our own species.

“A male would have to pull out a butcher knife and stab the baby, but that’s not what you’re seeing,” said Dr. Hrdy. “I don’t think langur infanticide tells us much. I think it’s a very different phenomenon.”