Big African Animals Are Pickier Eaters Than We Imagined

To the surprise of ecologists, plant-eaters manage to coexist on the savanna by each choosing different favorite foods

The Grevy's zebra (left) and the plains zebra may be tough to tell apart—until you examine their dietary preferences via their poop. Both images by Winfried Wisniewski/Minden Pictures/Corbis

What do the diets of elephants, zebras, antelope and cows have in common? Most would answer correctly that all of these animals eat plants. But according to high-tech analysis of their poop, that is where the similarities end. Africa’s herbivores have surprisingly different favorite foods, according to research published this week in Proceedings of the National Academy of Sciences.

This study only tallied up what the animals ate, not which plants are essential for maintaining their health, so scientists can't yet say for sure whether eliminating a particular plant species would have a significant impact on the herbivores involved. But armed with this knowledge, wildlife managers can ensure that efforts to protect animals also include protection for their preferred types of grass, leaves or brush, hopefully helping to stave off ecological turmoil.

“When I talk to non-ecologists, they are stunned to learn that we have never really had a clear picture of what all of these charismatic large mammals actually eat in nature,” says Tyler Kartzinel, a postdoctoral research associate at Princeton University and lead author of the paper. “What we show is that the dietary differences among species are far greater than everyone has assumed.”

To arrive at these findings, Kartzinel and his colleagues from Princeton and the Smithsonian Institution traveled to Kenya’s Mpala Research Center and Conservancy. They focused on seven species that represented 99 percent of the large herbivore population at their study site, including elephants, plains zebras, Grevy’s zebras, impalas, Cape buffalo, a tiny antelope called the dik-dik and domesticated cattle. 

Figuring out what these animals eat in the wild was a challenge—the researchers couldn’t just follow them around and tally up whatever they put in their mouths, trunks or snouts. “These animals are difficult and dangerous to observe up close,” Kartzinel says. “They move long distances, they feed at night and in thick bush, and a lot of the plants they feed on are quite small.”

To overcome the observational obstacle, the team turned to a method developed several years ago called DNA metabarcoding—essentially, a high-tech way of analyzing dung samples. DNA metabarcoding identifies genes from digested food found in feces and then matches those sequences to a database of plants, pinpointing what the animal last had for breakfast. “We spent a lot of time in Kenya watching animals out of car windows, waiting for them to defecate,” Kartzinel says. “When they did, we rushed out, grabbed a sample and brought it back to the lab.”

Based on an analysis of nearly 300 fecal samples, the team found that the seven species’ diets all differed significantly. Even the two zebra species—the most closely related animals and the ones whose ranges overlap—had very different favorites. In total, the two zebra species at about 45 species of plants, but 15 species differed significantly between their diets—10 on the Grevy zebra’s list of favorites, and five on the plains zebra’s. “This may not sound like much at first, but it’s a third of all of the food types that we detected in the diets of either species,” Kartzinel says.

The findings help explain how it’s possible that so many large, plant-eating animals can coexist on the African savanna. The results also have important conservation implications, showing that animal diversity likely hinges on plant diversity—a piece of the wildlife management puzzle that until now had been poorly understood. “There has been a lot of anguish in ecology about the fact that our models often fail, and experimental results tend not to be the same from time to time and place to place,” says study co-author Robert Pringle, also at Princeton. “We end up making simplifying approximations, and those can sometimes be very misleading.”

As this study shows, the new DNA metabarcoding method “enables us to be precise where previously we have been fuzzy,” Pringle continues, telling a more detailed story of who eats what on the savanna.  

Get the latest Science stories in your inbox.