Evolution of group living

Why do lions live in groups? Lions are spectacularly sociable: they hunt together, raise their cubs in nursery groups and defend joint territories. We have studied the evolutionary basis of lion sociality for more than 30 years, and we have tested a number of alternative hypotheses.

The traditional explanation for lion sociality has been cooperative hunting, but we published a series of papers in the 1990s showing that lions do not hunt as cooperatively as believed. In fact lions only hunt cooperatively when they need to. If their best chance for a meal is a large dangerous prey like a Cape buffalo, lions certainly do pull together. But if the prey is relatively easy for a singleton to capture, the rest of the pride is more likely to cheer on their hunting companion rather than join the hunt. Hence we often see a lone female pursuing a warthog while the rest of the pride merely watches. Similarly, lions don’t generally cooperate when hunting wildebeest but are somewhat more likely to jointly pursue a zebra (a more difficult prey).

Consequently, individual lions don’t always feed better in larger groups. Food availability is highly seasonal in the Serengeti, and during the “lean season” solitary females (who can catch warthogs by themselves) feed just as well as individuals in groups of five females (which are large enough to subdue buffalo); but groups of two to four females suffer lower food intakes. The evolution of sociality would have required a strong advantage for ancestral lions to start living in pairs or trios, so group hunting isn’t the answer.

Nursery groups (we call them “crèches”) form whenever pridemates give birth within a few months of each other. After hiding their cubs for the first few weeks of life, mothers bring them to the rest of the pride and nursing mothers become almost inseparable for the next two years. But crèche-mates suffer lower feeding success, so the behavior does not derive from any advantage of group foraging. Instead, the crèche is a defensive formation.

Resident males are the fathers of all the cubs sired during their tenure. Outside males are always looking for mating opportunities. If they encounter a mother with small cubs, they will kill the cubs to induce her to return to mating condition. Resident males patrol the edges of the pride territory, and outside males sometimes encounter the crèche unaccompanied. Cubs in a crèche are far more likely to survive these encounters than cubs raised by singleton mothers. Groups of mothers can successfully counter-attack the invading males, whereas a female has no chance one-on-one against a fully grown male. Thus crèche formation clearly increases female reproductive success. However, most other felids also show infanticide by incoming males (including leopards, cougars, tigers, bobcats and lynx), yet these species are all solitary—so defense against infanticide doesn’t explain why lions are the only social cats.

The true hallmark of lion behavior is their joint defense of a territory. We conducted a series of experiments in the late 1980s and early 1990s that revealed important aspects of cooperative defense. Karen McComb measured the responses of one to seven Serengeti females to the recorded roars of one or three females. A roar is a territorial display, so if a stranger roars in the middle of your territory, it’s like coming home and finding a stranger sitting comfortably in your living room. The females responded according to the odds: if lone females heard the roar of a single female, they would sit tight and try to recruit their distant pridemates; but groups of three females would immediately approach the loudspeaker. When exposed to a roaring trio, a real trio would again try to recruit help; but a quintet would quickly approach. As long as “us” outnumbered “them” by at least two individuals, “we” would move to oust the invaders. In a set of experiments to resident males, Jon Grinnell found a similar sense of “numeracy” but the males sometimes approached even when outnumbered three to one—probably because male lions only have a brief opportunity to father offspring and are more likely to be suicidal in protecting their pride.

Possessing a high quality territory is essential for successful reproduction, and as a pride grows it is able to annex particularly valuable landscape features. Anna Mosser used the long-term ranging and reproductive data from the Serengeti study area to develop a map of lion “real estate.” Lions need continuous access to three key resources: food, water and shelter. Lions do not rove at will over the landscape, looking for food—they center their activities at sites where prey are likely to pass. During the dry season, water is restricted to only a few areas. Small cubs must be hidden in dense vegetation where they are safe from hyenas and leopards. River confluences turn out to be the most important landscape features in the Serengeti. Tributaries join in a geometric pattern that is reminiscent of a series of funnels: migratory herbivores are reluctant to cross rivers because of the risk of ambush, so they often end up trapped in the junction of a confluence—most kills occur inside these bottlenecks. Water flow also scours the riverbeds most deeply at confluences, leaving the most persistent waterholes during the dry season. And the added moisture permits the densest vegetative growth at a confluence. So lion real estate value is highest at the confluences and prides that control enough confluences leave the most descendants. Prides in high quality areas recruit more daughters who then help them to retain the most valuable sites. Over 40 years of data, we find that the size of the pride largely determines the quality of the habitat that they can control.

Lions are the dominant species of the savannas. The savannas are the most heterogeneous of felid habitats; lions also live at higher densities than other big cats and thus face the most intense competition for good habitat. Thus only lions can benefit from banding together and defending highly valuable landscape features.