The most dramatic illustration of bipedalism is the pelvis, and the most dramatic specimen demonstrating pelvic morphology is the relatively complete skeleton from Hadar, Lucy, AL 288-1. This important fossil preserves a nearly complete innominate (hip) bone and sacrum, which have been used to reconstruct the pelvis of this individual. The upper portion of the innominate bone, called the ilium, is short and curving compared to the long, flattened ilium of chimpanzees and other apes. This curvature provides an anterior attachment for the muscles that pull the femur forward during walking or running. Although the ilia are short in length compared to a chimpanzee, they extend more broadly to the side, resulting in a pelvis that is very broad overall. Indeed, this pelvis is nearly as wide as that of a human female, despite the small body size of the australopithecine. This pelvic width contributes to a very different body shape for early hominids than for humans.
The width of the pelvis affects the muscular requirements of walking. Whenever one leg supports the body, the trunk of hominids tends to fall away from the supporting leg. The muscles that prevent the body from falling over attach to the lateral part of the ilium and to the femur, pulling the trunk upward around the hip joint. A wide ilium tends to increase the efficiency of these muscles. Their effectiveness is also aided by a long femur neck, just as long handles on a pair of scissors greatly increase the force with which they can cut. This configuration of muscles is similar to the condition found in living people, but it is much more extreme. In living people, the pelvis is much narrower in proportion to overall body height, and femur necks are much shorter in proportion to femur length.
A number of hypotheses compete to explain why the bipedal adaptation of early hominids should be different from modern human bipedalism in this way. One explanation is that the pelvic width contributes to the length of the stride, by rotation of the pelvis during walking. This rotation would increase stride length without increasing the length of the swing of the legs, allowing an increased stride without lowering the mass of the body--which if lowered would subsequently have to be raised again through greater muscular exertion (Rak 1991). Alternatively, the widely spaced femora may allow a greater mechanical advantage for the muscles that draw the legs medially, toward the midline. Such a configuration would be advantageous for certain leg motions, especially the style of climbing that requires the legs to clamp around a branch or trunk. This kind of climbing would be more necessary to bipeds who lacked the prehensile feet of living apes.
The wide pelvis of early hominids had one consequence beyond those related to bipedalism, by setting the stage for a major evolution of the birth process. During birth, the infant must pass through the birth canal, entering it through the pelvic inlet. Although relatively small compared to the pelvic inlets of the apes, the oval shape of the australopithecine pelvic inlet probably did not create any problems during birth because australopithecine infants likely had head sizes about the same as those of chimpanzees. Even so, Robert Tague and Owen Lovejoy (1986) have speculated that birth in australopithecines may have involved a more complex series of events than chimpanzee births, with the possibility that the infant head may have had to rotate a quarter-turn to be born in line with the sideways oval in a manner similar to the half-rotation of a human infant during birth. Karen Rosenberg and Wenda Trevathan (2002) suggest that at this time, hominid females began to require assistance with the birthing process. The pelvic constraints associated with bipedal locomotion would come to determine much more about the birth process in later hominids.
Rak Y. 1991. Lucy's pelvic anatomy: its role in bipedal gait. J Hum Evol 20:283-290.
Rosenberg KR, Trevathan W. 1996. Bipedalism and human birth: the obstetrical dilemma revisited. Evol Anthropol 4:161-168.
Tague R, Lovejoy C. 1986. The obstetric pelvis of A.L. 288-1 (Lucy). J Hum Evol 15:237-255.