Based on these criteria, several lines of fossil evidence suggest that ER capabilities first emerged in the genus Homo. These features are discussed at length in Bramble and Lieberman (2004), but a few that are illustrated in Fig. 8.1 merit brief mention here. First, while there are some indications in the skeleton of morphological specializations related to the mass-spring mechanics of running, features related to stabilization are more prevalent. In terms of trunk stabilization, the cranial portion of the gluteus maximus, which plays a critical role in running but not walking, has a considerably expanded origin in H. erectus relative to Australopithecus (Rose, 1984; Lieberman et al., 2006). The gluteus maximus also acts in concert with the erector spinae to stabilize the trunk, and the sacroiliac trough in which the latter originates may be considerably expanded in Homo compared to Australopithecus (see Lovejoy, 1988). Even more concrete evidence of derived mechanisms for stabilization relevant only to running is in the head. As shown by Spoor et al.
Fig. 8.1 Illustration of basic body shape differences between A. afarensis (left) and H. erectus (right) highlighting features discussed in the text that are derived in H. erectus and which would improve endurance running performance. Features in parentheses are as yet unknown (in the foot) or hypothetical reconstructions (e.g., Achilles tendon length). Note that shoulder position (indicated with an *) in H. erectus is unresolved (Modified from Bramble and Lieberman, 2004).
(1994), the diameters of the anterior and posterior semicircular canals relative to body mass, which influence their sensitivity to head pitching accelerations, are first expanded in early Homo compared to Australopithecus and Pan. The vestibular system is fully formed prior to birth (Jeffery and Spoor, 2004), and is not significantly challenged during walking. It is difficult to think of any human activity other than running that would have selected for increased sensitivity to head pitching.
The anatomical relationships between the shoulder and the head comprise another set of derived features of Homo that are absent in Australopithecus and which have key roles in head stabilization during running. During walking, the head is stabilized in minor ways by inertia, the viscoelastic properties of the ligaments and muscles that connect the head to the axial skeleton, and by contractions of the head extensors (Hirasaki et al., 1999; Winter et al., 1990). During running, however, the heelstrike transient imparts such a rapid and substantial pitching impulse to the head that it needs to be stabilized almost instantly to avoid vestibular overload. Humans do so by a novel mechanism (a mass-damping system), in which the long axis accelerations of the arm counter pitching accelerations of the head via an out-of-phase elastic linkage (Bramble et al., 2009). A critical component of this system is an almost complete decoupling of the head and shoulder so they can act as linked masses. In chimpanzees, the shoulder and head are tightly connected by a massive trapezius, the rhomboideus, and the atlanto-clavicularis (Aiello and Dean, 1990). These connections have all been lost in humans with the exception of the cleidocranial portion of the trapezius (CCT). This muscular strap between the shoulder and midline occiput interdigi-tates with another novel feature in humans, the nuchal ligament (NL). This tendon-like structure originates on the midline of the occiput and connects with the upper trapezius as well as a deeper fascial septum that attaches to the cervical spines (Mercer and Bogduk, 2003). A NL is present in other cursors such as canids, equids and bovids, as well as in a few species with massive heads (Dimery et al., 1985; Bianchi, 1989). In running, but not walking, the CCT fires before HS on the stance side arm, linking the mass of this arm with the head in the midsagittal plane via the NL. Critically, evidence for this linkage is first present in the fossil record of early Homo (all H. erectus skulls as well as KNM-ER 1813) because the NL leaves a trace on the skull in the form of a sharp, everted, median nuchal line that that is not present in Australopithecus or Pan. While apes and australopiths sometimes have a rounded ridge in the midline of the occipital, this ridge lacks the everted contour indicative of a NL.
Other derived changes in Homo relevant to stabilization during running but not walking may be evident in the shape of the waist, thorax and neck. Leg swing during the aerial phase of running causes substantial angular momentum, which, unchecked, would cause the body to rotate around a vertical axis before heel strike. Humans counteract this momentum not only by swinging the arms in opposition to the legs, but also by rotating the thorax independently of the hips and head (Hinrichs, 1990). Such rotations, which are neither important nor particularly marked during walking, are made possible by two zones of separation: a relatively narrow, tall waist; and a relatively tall neck with low, wide shoulders. Although the waist in Australopithecus was probably as tall as in Homo, it was relatively wider as judged by the greater bi-iliac breadth of the australopithecine pelvis (Lovejoy, 1988; Schmid, 1983). A narrow waist in Homo may reflect smaller guts (Aiello and Wheeler, 1995), but it would also have improved running performance by reducing resistance between the pelvis and ribcage, and decreasing inertial moments for thorax rotation.
The second rotational zone of separation, between the upper thorax and neck, is harder to assess in early Homo. Whether the thorax ofAustralopithecus was funnel-shaped, as in apes, or barrel-shaped, as in humans, is debated (Schmid, 1991; Ward, 2002), but most evidence suggests that a barrel-shaped upper thorax is first present in the KNM-WT 15000 skeleton (Jellema et al., 1993). The primitive condition of a narrow upper thorax in combination with more muscular connections between the shoulder and head would have no effect on walking performance capabilities. They are useful for helping generate torque in the shoulder for orthograde climbing (Larson, 1993), but would decrease the ability to stabilize the head during running. It is interesting to speculate that selection for running capabilities may have come at the expense of adaptations for climbing, explaining why Homo is the first non-arboreal primate. However, Larson et al. (2007; see also Larson, 2009) have suggested that the KNM-WT 15000 shoulder was somewhat narrow with a relatively short clavicle and a glenoid fossa that faced anteriorly in order to accommodate a low degree of humeral torsion. A humerus from Dmanisi (D2700) also has a low degree of humeral torsion (Lordkipanidze et al., 2007). It is hard to interpret these data in part because both KNM-WT 15000 and D2700 are juveniles. The clavicle (which grows intramembranously) is the last bone in the human body to attain adult size, and both skeletons have clavicles that fall in the range of humans at equivalent ontogenetic stages (Scheuer and Black, 2000). Without better reconstructions of the upper thorax itself, it is difficult to assess the relative breadth and position of the shoulders in these specimens, one of which (KNM-WT 15000) has evidence for axial pathologies that may have affected upper thoracic anatomy. Regardless, low humeral torsion in H. erectus would have compromised its ability to throw effectively (Larson et al., 2007; Larson, 2009), raising questions about how the species was able to hunt (see below).
As noted above, it is much harder to document elastic structures in the skeleton relevant to the mass-spring form of energy exchange used in running but not walking. The most important anatomical components of the system are extensive tendons, especially the Achilles, which are substantially longer in humans relative to chimpanzees or gorillas. The size of the Achilles tendon insertion in the Hadar calcanei (Susman et al., 1984) suggests that they had an ape-like configuration, but such inferences must remain speculative without evidence of some relationship between tendon length and tendon insertion morphology. A more promising anatomical region for evidence of mass-spring anatomy that requires further study is the foot, especially the plantar arch. Some form of arch is useful in bipedal walking in order to act as a windlass to stiffen the foot for effective toe-off (KappelBargas et al., 1998), but in running the arch functions quite differently as a spring, storing and releasing approximately 17% of the energy generated during each impact of the foot with the ground (Ker et al., 1987). Although australopith-ecines clearly had some form of plantar arch, there are several indications that the arch had a different configuration in Homo. In particular, the navicular in apes and australopith-ecines has a relatively expanded medial tuberosity, suggesting that it was a weight-bearing element (Harcourt-Smith, 2002). In addition, the first hominin fossil with a close-packed calcaneo-cuboid joint (as evident by an expanded medial flange on the proximal cuboid) is OH 8, a specimen attributed to early Homo (Lewis, 1989; Susman, 2008). Together, these novel features - along with an unequivocally adducted big toe and a relatively shorter forefoot (see Susman et al., 1984; Aiello and Dean, 1990) - hint that elastic storage mechanisms in the foot necessary for running may be derived features of the genus Homo.
Finally, it is important to note that there are more than a dozen other derived skeletal features of the genus Homo, particularly in H. erectus, which improve performance for both walking and running (summarized in Bramble and Lieberman, 2004). Given that hominins were habitual bipeds for at least 4 million years before the origin of H. erectus with little evidence for any major change in postcranial anatomy (reviewed in Ward, 2002), it is difficult to imagine that selection for walking alone was responsible for the derived features of Homo. The most likely scenario is that H. erectus was the first hominin with a substantially expanded diurnal day range made possible by both walking and running. Indeed, both gaits are important ways to travel long distances, and one can expect that hominins would have walked rather than run whenever possible (see below). Thus, the extent to which selection acted on running versus walking is impossible to assess, as both would have been important. That said, it is worthwhile noting that the considerably more extreme thermoregulatory and mechanical demands of running might have imposed a greater selective benefit on performance capabilities in running than walking. In addition, many ancestral features of australopithecines that improve climbing performance, such as long forearms and heavily muscled shoulders, do not conflict with the biomechanical demands of walking, but may impede the ability to stabilize the head. Selection for running capabilities may thus have selected against arboreal capabilities in Homo.
Put together, there is much evidence that H. erectus but not Australopithecus was capable of ER. However, this inference does not imply that H. erectus was necessarily as good as modern humans or even later archaic Homo at ER. Some modern human features that improve ER performance may have evolved since early H. erectus. In addition, there are some hints that H. habilis may have possessed some ER capabilities, but the evidence is sparse and equivocal (see Bramble and Lieberman, 2004). While it is possible that ER capabilities had evolved by the time of H. habilis, it is premature to be definitive, and there are theoretical reasons to hypothesize that such capabilities, if they existed, were not as developed as in H. erectus. Natural selection tends to take advantage of existing variations in the context of some fitness benefit. Thus, it is unlikely that selection would have favored the evolution of ER-related features if hominins had not already been engaged to some extent in a form of ER. One potential scenario is that early Homo during the Oldowan started to scavenge and/or hunt to a limited extent. At some point, hominins that were better at ER for various reasons (longer legs, larger anterior and posterior semicircular canals, and so on) had a slight fitness benefit, leading to the evolutionary changes that we observe in H. erectus.
Was this article helpful?