Your dinner or your life: What determines the sprint speed of gazelles, zebras, giraffes … and ostriches?

2010 076 Masai Mara b 24

Thomson gazelle, on the run. Photo by ngari.norway

In the evolutionary history of big herbivores and the carnivores that prey upon them, the phrase “arms race” is only technically a metaphor. Antelope and zebras are literally born to run, and many of the things that chase them, like wild dogs or cheetahs, are either masters of endurance or champion sprinters. The evolutionary story almost writes itself: over millions of years of chasing, and being chased, whenever the predators evolved to become faster, the prey were selected to run even faster—until a cat evolves that can go from 0 to 60 faster than my Volkswagen Rabbit.

Except of course there’s more to life than running for your life. An antelope’s frame is under more demands than evading cheetahs—it also needs to travel long distances to follow food availability with the shifting rainy season. In fact, the North American fossil record suggests that big herbivores on that continent evolved long legs for distance running millions of years before there were predators able to chase after them. And then again, not all predators run their prey down; lions, for instance, prefer to pounce from ambush.

In a paper recently released online ahead of print in the journal Evolution, Jakob Bro-Jørgensen sets out to disentangle exactly these competing explanations.

Bro-Jørgensen compiled data on sprint speed and diet in a set of fifteen large herbivores from the African savannah, including Thomson gazelle, impala, plains zebra, giraffe, and even warthog and ostrich. For each herbivore species, he calculated an index of vulnerability to each of five savannah predators: cheetah, spotted hyena, wild dog, leopard, and lion. The first three of those predators run down their prey; the last two prefer to ambush from short distance.

Cheetah Run 4

Predation pressure! Photo by ShootNFish

Bro-Jørgensen then constructed analytic models using the risk of predation by each predator, diet composition, and body mass as predictors for the the herbivores’ sprint speeds. He constructed the predictive models with the phylogenetic relationships among the herbivores as a covariate—to account for the fact that sable antelope and roan antelope may be similar to each other just because they’re a lot more closely related to each other than either is to ostriches.

The model that best explained variation in herbivore sprint speed incorporated vulnerability to all five predators—but neither diet nor body mass. Not all predators had the same effect on herbivore speed, though: increased risk of predation by wild dogs, hyaenas, or cheetahs was correlated with increased sprint speed—but greater risk of predation by lions or leopards was correlated with reduced speed. That’s consistent with the idea that running fast may not be a very helpful defense against a predator that attacks from ambush.

Bro-Jørgensen’s results are only correlative; this isn’t the same quality of evidence we might get from a comparative study of sprint speeds in, say, two populations of the same herbivore species that are under selection by predators using different attack strategies. But it does seem like a clear-cut demonstration that, at a very broad scale, the big herbivores on the African savannah have been shaped by running for their lives, not for their dinners.


Bro-Jørgensen, J. n.d. Evolution of sprint speed in African savannah herbivores in relation to predation. Evolution 1–17. doi: 10.1111/evo.12233.

Janis, C. M., and P. B. Wilhelm. 1993. Were there mammalian pursuit predators in the tertiary? Dances with wolf avatars. Journal of Mammalian Evolution 1:103–125. doi: 10.1007/BF01041590.