Chimpanzees are mostly peaceable creatures, spending much of their time foraging for food and grooming each other. But occasionally they kill their own kind. Why they engage in these lethal bouts of aggression has been uncertain. One theory holds that killing is an evolved strategy for reducing competition for resources; another posits that human disturbance—including hunting and deforestation—has triggered the behavior. Now a large study of killings in chimp communities across Africa has cast new light on the dark side of our closest living relatives.

Michael Wilson of the University of Minnesota and his colleagues amassed data on chimps in 17 communities at 10 sites in west, central and east Africa whose members had long been observed by researchers. For each community the team obtained data on demography, population size, ranging behavior of the group, and so forth. The team also compiled data on bonobos from four communities, to determine whether bonobos really are less aggressive than chimps, as is commonly asserted. Wilson presented the study results on April 12 at the annual meeting of the American Association of Physical Anthropologists in Portland, OR.

Focusing on 86 cases in which scientists had observed the killings or had inferred them based on compelling evidence, Wilson and his collaborators found that kills occurred in most of the chimpanzee communities and that victims tended to be infant and adult males outside the killer’s social group. Most of the killings were conducted by groups of males.

Interestingly, human disturbance did not appear to be a factor in the kills. Indeed the community with the most kills—a group at the site of Ngogo in Uganda’s Kibale National Park—had the least human disturbance. Neither was the ratio of males to females in a group a factor. What did appear to be a factor was the number of males in a group: the higher the number of males in a group, the higher the number of kills. Ngogo, the community with the highest rate of kills per year, also had the highest number of adult males.

“The number of males is important because the more males there are, the more competition there is for mates in the community,” Wilson explains. The number of males also equals the community’s fighting strength for defending their territory and the food resources in it. Males in communities with more males can afford to be more aggressive because they have backup. The researchers did not identify a particular number of males that triggered killing. Rather, Wilson offers, the key may be the relative numbers of males in neighboring communities—that is, the balance of power.

“This tells us something about human evolution,” Wilson comments. He notes that although scientists do not know whether humans and chimps inherited their capacity for lethal aggression from a common ancestor or whether it arose in both species through convergent evolution, “lethal aggression is related to power asymmetries where members of one group can kill others with low cost.”

As for the bonobos, this study bolsters the claim that they are less aggressive than chimpanzees: there were no clear-cut homicides in any of the bonobo communities. Another presentation given at the meeting provided a possible clue to the apparent absence of male aggression among these apes: Victoria Wobber of Harvard University and her colleagues studied testosterone levels in chimpanzees and bonobos from infancy to adulthood and found that whereas chimpanzee testosterone levels surged during adolescence (particularly among males), bonobo testosterone production remained consistent over the course of development.