Quantitative Analysis of Social Networks for Two Zebra Species, Grevy’s Zebra (Equus grevyi) and Plains Zebra (Equus quagga)

Abstract

Grevy’s zebra (Equus grevyi) and plains zebra (Equus quagga) are two closely related species that exhibit vastly different social systems. Grevy’s zebra exhibit resource defense polygyny in which males defend territories with critical resources in hopes of obtaining access to unstable open membership groups of females. Formation of stable groups is hindered by the contrasting priorities of females based on reproductive state: females in peak lactation must remain within two kilometers of water, while non-lactating females may travel further from water in search of better forage. In contrast, plains zebra live in closed membership groups in which a single male stallion defends a harem of females and dependent offspring. These harems often aggregate together along with non-breeding bachelor males to form spatially cohesive herds, constituting the upper level of their two-tiered social structure. This study uses social network analysis to explore the individual incentives behind the association decisions that ultimately culminate in the vastly different social systems of these two zebra species. For analysis of Grevy’s zebra, interactions are evaluated on an individual basis, whereas for plains zebra, each harem is treated as a cohesive decision-making unit and only the herd-level interactions among harems and bachelor males are examined. The results of this analysis provide quantitative evidence that individuals act to maximize their own reproductive success consistent with their varying priorities based on sex and reproductive status. For Grevy’s zebra, we found that lactating females associate assortatively with other lactating females. We also obtained evidence that lactating females generally form a relatively long-term monandrous bond with a single territorial male, while non-lactating females tend to form more brief polyandrous bonds with multiple territorial males. As a result of associating with multiple spatially isolated territorial males, polyandrous females serve to connect the Grevy’s zebra network as a whole. For plains zebra, we found that bonds between bachelors are more persistent than those between harems or those between bachelors and harems, suggesting that bachelors form cohesive cliques. We also obtained evidence that harems form more strong bonds with other harems in the face of greater bachelor harassment, but that harems with more adult females form fewer strong bonds with other harems. Stallions appear to favor maintaining strong alliances with other stallions to cooperatively defend against intruding bachelors. Our preliminary data suggest that as the number of adult females in a harem increases, it becomes more difficult for the stallion to dictate his harem’s movements and associations. We hypothesize that females may collectively assert control by breaking bonds with other harems in favor of searching for greener pastures. Finally, our data suggest that the plains zebra network has the potential for more rapid transmission of disease and other types of information throughout the population than does the Grevy’s zebra network, which is consistent with differences in the social systems exhibited by both species.