If you’ve ever chased a cat that’s trying to avoid a bath, you have every right to conclude that, for our size, we humans are pretty poor runners. But chasing a cat is sprinting. Where we excel is endurance running. Moreover, we run long distances at fast speeds: many joggers do a mile in seven-and-a-half minutes, and top male marathoners can string five-minute miles together for more than two hours. A quadruped of similar weight, about 150 pounds, prefers to run a mile at a trot, which takes nine-and-a-half minutes, and would have to break into a gallop to keep pace with a good recreational jogger. That same recreational jogger could keep up with the preferred trotting speed of a thousand-pound horse.

Good endurance runners are rare among animals. Although humans share the ability with some other groups, such as wolves and dogs, hyenas, wildebeest, and horses, we alone among primates can run long distances with ease.

But what evidence can support the idea that endurance running by itself gave early humans an evolutionary advantage, and that it wasn’t just “piggybacking” on our ability to walk? Many traits, after all, are useful for both activities; long legs, for instance, and the long stride they enable, are helpful to walking as well as to running. But running and walking are mechanically different gaits. A walking person, aided by gravity, acts as an inverted pendulum: the hip swings over the planted foot [see “The Biomechanist Went over the Mountain,” by Adam Summers, November 2004]. In contrast, a runner bounces along, aided by tendons and ligaments that act as springs, which alternately store and release energy.

Bramble and Lieberman point to a number of features, preserved in fossils, that imply Homo adapted to a bouncy gait—whereas Australopithecus stuck with walking. Fossils lack tendons and ligaments, of course, but traces of their attachment points are sometimes present, and the characteristics of the missing tissue can be inferred by comparing how the attachments fitted with the rest of the animal’s anatomy. For example, the Achilles tendon, attached to the heel bone, is one of the most important elements in a human’s bouncy gait. In Australopithecus, however, the attachment point of the tendon is distinctly chimpanzee-like. Another spring occurs in the foot itself: tendons in the sole of a human’s foot keep it arched. The arch flattens and springs back with each step. In contrast, Lucy had only a partial arch. Homo habilis had a full arch. Chimpanzees have no arch at all.

In addition to springs, endurance running requires more stabilization of the trunk than walking does. Members of the genus Homo have substantial gluteus maximus (butt) muscles. Those muscles have numerous large attachments from the hip to the base of the spine. In Australopithecus fossils, though, the muscle has a much more limited area of attachment. If you’ve seen a chimpanzee in trousers, you know how baggy they look. Chimpanzees are gluteally challenged as well. Large butt muscles are not only better looking in pants; they also make for efficient energy transfer during running by stabilizing each hip. But the muscles are not used for walking on level ground.

In contrast with the trunk, the shoulder of the chimpanzee is well stabilized, tied to the spine and the head by several strong muscles. Lucy retained the stabilized shoulder, but in humans those muscle connections are less robust—and for good reasons. When we walk, our shoulders don’t move much, but when we run, because of the relatively loose attachment, the shoulders rotate strongly one way while the hips rotate the other. The counterrotations help keep us in balance. And because only one part of the trapezius muscle attaches to the head, we can swing the upper body without inadvertently rotating the head—which enables us to see where we’re going.

In spite of the loose attachment between head and shoulders, running joggles the head more than walking does. Homo therefore has several “antibobblehead” adaptations that other apes and Australopithecus lack. The first is a modification of the semicircular canals, the organs in each inner ear that tell the brain which way is up. Three such canals sit at right angles to one another in each inner ear. Two are enlarged in Homo, and the size makes it easier to sense, and presumably to counteract, a nodding head. An elastic ligament that runs from a ridge at the base of the skull to the base of the neck, damps the bobbing effect. Analogous ridge structures, to which damping ligaments can be attached, occur in dogs and horses, the other long distance runners, but not in Lucy.

Bramble and Lieberman’s wide-ranging analysis makes important corrections to the scientific picture of early humans. Our ancestors may have ranged across large distances in the heat of the African savanna in relatively short spurts of long-distance running, as well as by walking. They may have been trying to maximize the chance of encountering carrion before other scavengers did, or perhaps they were adapted to running down prey before spear throwers or bows were invented.

In any case, our current appetite for jogging is made possible by the early selective pressures that made humans one of the most accomplished endurance-running animals. For myself, though, I imagine another adaptation. The heat and the running must have been powerful motivators for our ancestors to sit in the shade and ponder how to affix a rock to a stick.