Human Locomotion and Heat Loss:

An Evolutionary Perspective
Daniel E. Lieberman*1

ABSTRACT
Humans are unique in many respects including being furless, striding bipeds that excel at walking
and running long distances in hot conditions. This review summarizes what we do and do not
know about the evolution of these characteristics, and how they are related. Although many
details remain poorly known, the first hominins (species more closely related to humans than to
chimpanzees) apparently diverged from the chimpanzee lineage because of selection for bipedal
walking, probably because it improved their ability to forage efficiently. However, because bipedal
hominins are necessarily slow runners, early hominins in open habitats likely benefited from
improved abilities to dump heat in order to forage safely during times of peak heat when predators
were unable to hunt them. Endurance running capabilities evolved later, probably as adaptations
for scavenging and then hunting. If so, then there would have been strong selection for heat-
loss mechanisms, especially sweating, to persistence hunt, in which hunters combine endurance
running and tracking to drive their prey into hyperthermia. As modern humans dispersed into
a wide range of habitats over the last few hundred thousand years, recent selection has helped
populations cope better with a broader range of locomotor and thermoregulatory challenges, but
all humans remain essentially adapted for long distance locomotion rather than speed, and to
dump rather than retain heat. ⃝C 2015 American Physiological Society. Compr Physiol 5:99-117,

2015.

Introduction multiple lines of evidence suggest that there was strong

selection on early hominins (species more closely related to
Humans are distinctive compared to other mammals in humans than to chimpanzees) to stand and walk efficiently,
numerous respects including being habitually bipedal, and and that the origins of bipedalism was later followed by
the ability to walk and run long distances at relatively fast additional selection for long distance walking and then for
speeds in hot, arid conditions. The optimal walking speed endurance running. In turn, it is reasonable to hypothesize
for an average-sized human is 1.2 m/s, about 20% faster and that selection for long distance walking and running created
four times more efficient compared to our closest relatives, a selective advantage for hominins to dump heat effectively
chimpanzees, and about 20% faster and approximately as in hot, arid conditions. After modern humans evolved and
efficient as a pony’s optimal walking speed (117, 133, 161). dispersed all over the globe, further selection occurred to
In addition, although maximum running speed in humans help different populations adapt to a wide range of climatic
is unimpressive, about half that of most equivalent-sized conditions, but all human populations are variants of a basic
quadrupeds (52), humans are among the few mammals—and adaptive pattern for long-term aerobic exertion in hot habitats.
the only species of primate—that can repeatedly run very I first summarize the evidence for the evolution of human
long distances at relatively fast speeds under aerobic capacity. locomotion and then heat loss, in both cases drawing on the
Other cursorial (running adapted) quadrupeds can run long fossil record as well as comparisons between humans and the
distances at a trot but not a gallop, yet the preferred trotting African great apes. I then evaluate alternative hypotheses for
speed of a pony, approximately 3 m/s, is only half the speed how these two distinctive systems evolved, and the extent to
at which a fit human can run a marathon with nearly equal which they are linked. I conclude with a discussion of the
efficiency (63, 133). Finally, most mammals are able to walk contemporary relevance of the evolutionary bases of these
or run long distances only in relatively cool conditions, but adaptations.
humans are the sole species of mammal that excels at long
distance trekking and running in extremely hot conditions. No
horse or dog could possibly run a marathon in 30◦ C heat (40). * Correspondence to danlieb@fas.harvard.edu
The purpose of this review is to summarize the evidence 1 Department of Human Evolutionary Biology, Harvard University,
for the origins of the special nature of human locomotion 11 Divinity Avenue, Cambridge MA 02138, USA
and heat loss mechanisms, and to make the argument—first Published online, January 2015 (comprehensivephysiology.com)
elucidated by Carrier (24)—that these systems share a linked DOI: 10.1002/cphy.c140011
evolutionary history. Although some details remain murky, Copyright ⃝ C American Physiological Society.

Volume 5, January 2015 99

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

Bipedalism locomote quadrupedally while maintaining adaptations for

arboreal climbing such as short hind limbs, long upper limbs,
Fossil evidence discovered over the last few decades indicates and long, curved manual phalanges (for review, see Ref. 46).
that selection for bipedalism may have been the initial spark Given the unique nature of knuckle walking in combination
that set the human lineage on a different evolutionary path with additional similarities between chimpanzees and goril-
from the African great apes. In fact, the presence of adapta- las, it was assumed before molecular data indicated otherwise
tions for bipedalism is a major reason for classifying many that these species were closely related cousins who shared
of these fossils as hominins (a problem of circular logic if a common knuckle-walking ancestor, and that the LCA of
bipedalism evolved more than once). Bipedalism, however, humans and the African great apes was unlikely to have been a
is an unusual form of locomotion that has a long, complex knuckle walker, but instead a generalized quadruped or possi-
evolutionary history involving several stages, each of which bly even some sort of brachiator with an orthograde (upright)
was contingent upon previous stages, and likely driven by trunk (see Ref. 60).
selective pressures caused by changing climatic conditions. The fact that chimpanzees and humans are more closely
Therefore, before reviewing the evolutionary transformations related to each other than to gorillas has necessitated a re-
that led to modern human walking and running, it is useful to evaluation of hominin origins. From a phylogenetic perspec-
begin with a consideration of the phylogenetic and ecological tive, the most parsimonious scenario is that knuckle walking
contexts in which hominin bipedalism first evolved. and other similarities between the African great apes evolved
just once in the LCA of chimpanzees, humans, and gorillas,
and that the more recent LCA of just humans and chimpanzees
Evolutionary context was also a knuckle walker that resembled chimpanzees and
As Figure 1 illustrates, molecular data unambiguously gorillas in many respects (125). If not, then the many simi-
indicate that humans and chimpanzees share a last common larities between chimpanzees and gorillas must have evolved
ancestor (LCA) that diverged approximately 5 to 8 million independently, which is highly unlikely.
years ago (Ma), and that gorillas diverged from the human- Reconstructing the LCA as a knuckle-walker has impor-
chimpanzee clade approximately 8 to 12 Ma. Molecular tant implications for hypotheses about the origins of bipedal-
evidence that humans and chimpanzees are sister taxa was ism, but this reconstruction is unsubstantiated and is the sub-
initially a surprise to paleontologists because chimpanzees ject of much debate for three reasons. First, although the
and gorillas share many similarities in their cranial and LCA of humans and chimpanzees has never been discovered,
postcranial anatomy, with many differences attributable to there are many species of fossil great apes from the Miocene
the effects of size (9, 54, 55, 151). Among their many sim- (23-5 Ma), and few of them closely resemble either chim-
ilarities, chimpanzees and gorillas habitually knuckle walk, a panzees or gorillas, especially in terms of their locomo-
distinctive form of locomotion that involves resting the fore- tor anatomy. Instead, some such as Proconsul are gener-
limbs on the dorsal surface of the middle phalanges of a flexed alized quadrupeds, and others such as Morotopithecus are
hand. Knuckle walking is commonly interpreted as a way to orthograde climbers (97, 171). Second, chimpanzees and
gorillas knuckle walk in a slightly different manner, lead-
ing some scholars to speculate that this mode of locomotion
H. sapiens P. troglodytes P. paniscus G. gorilla
evolved independently in the two species (77). Third, although
0– there is almost no fossil record of chimpanzee and gorilla evo-
lution, paleontologists have discovered a number of putative
1–
early hominins, and some scholars have argued that these
2– early members of the human lineage do not resemble chim-
panzees or gorillas (4, 181). Although this view cannot be
Time (millions of years)

3– discounted entirely, the balance of evidence suggests that the

4– LCA was a knuckle-walker. Most importantly, interpretations
of the earliest hominins as unlike chimpanzees or gorillas
5– LCA of humans
are problematic because these fossils actually resemble chim-
and chimps
6– panzees and gorillas in most aspects for which they are unlike
later hominins (91, 186). In addition, while Miocene apes are
7– diverse, most of these species are unlikely to be closely related
8– to the LCA of humans and chimps, which means that their
LCA of humans, locomotor adaptations do not refute reconstructions of the
9– LCA as similar to the extant African great apes (114). Finally,
chimps and gorillas
although adult gorillas do knuckle walk with more vertical
Figure 1 Evolutionary tree showing the relationships among forelimbs than chimps, this difference is exactly what one
humans, chimpanzees, and gorillas, as well as the Last Common
Ancestors (LCA) of humans and chimps, and of humans, chimps, and expects in larger bodied animals, which use less crouched pos-
gorillas. The dates of the divergences are only approximate. tures for reasons of scaling (10). As with many aspects of their

100 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

0 Holocene hominins needed to travel longer distances to find fruit (131).

Chimpanzees rely on fruit for more than 75% of their diet,
and travel on average only 2 to 3 km per day (119). Yet,
1 the highly flexed limb postures required for knuckle walk-
Time (millions of years ago)

ing are highly inefficient, costing four times more energy per
Pleistocene unit body mass per unit distance than humanlike bipedalism
2 (Ice age) (117, 154). If early bipeds were able to reduce their cost of
locomotion, they would have reaped substantial energetic ben-
efits if and when they needed to travel longer distances than
3 chimps to forage. Testing this hypothesis, however, requires
Pliocene
reliable reconstructions of the locomotor repertoire of not just
the LCA but also the first hominins.
4

Major transitions in hominin bipedalism

5
Although the first hominins appear to have been bipeds, they
5 4 3
probably stood, walked, and ran very differently from modern
δ18O benthic carbonate (0/00)
humans. Instead, hominin bipedalism appears to have evolved
Colder Warmer via three major stages (summarized in Figs. 3 and 4), each of
which contributed in different ways to the suite of locomotor
Figure 2 Record of the temperature of the earth’s oceans based on
δ 18 O from benthic carbonates. Note the general cooling trend over
the last 5 million years, but with much variation.
0

H. neanderthalensis

H. floresiensis
H. sapiens

H. heidelbergensis
anatomy and diet, locomotor differences between gorillas and

H. erectus
chimpanzees are best explained as consequences of size.
Debates over the nature of the LCA remain resolved, 1
but the hypothesis that they were knuckle walkers has far-

P. robustus
reaching implications for hypotheses about why and how

P. boisei
Homo habilis
bipedalism evolved in the rainforest contexts in which the
LCA, like extant great apes, almost certainly lived. Africa 2
Au. africanus

during this period, the late Miocene, was warmer and wet-
Au. sediba

Paranthropus
aethiopicus
Millions of years before the present

ter than today, but was generally trending (but with many
fluctuations) toward becoming cooler and drier as a result
of global and regional changes (Fig. 2) (11, 26, 76). Major
3
Australopithcus

consequences of these changes were fractionation of the rain

afarensis

forests and the expansion of more open woodland habitats,

which appear to be the dominant ecological context in which
the earliest hominins have been unearthed (see below). More
4
Ar. ramidus

open habitats must have been a source of stress for primar-

ily frugivorous apes because the patches of fruit on which
they relied would have become smaller, more dispersed, and
more seasonal. Although there are many proposed explana-
Ardipithecus
kadabba

tions for why bipedalism initially evolved (for reviews, see 5

Refs. 46, 60, 91), the two leading hypotheses propose that
these stresses favored incipient bipedalism in ways relevant
to food acquisition. The first hypothesis is that bipedalism
initially evolved as a postural adaptation for more effective 6
Sahelanthropus
tchadensis

upright feeding. Studies by Hunt (68) and by Crompton and

tugenensis
Orrorin

colleagues (32, 33, 163) have shown that apes often stand
upright when either feeding on the ground or in trees. If
hominins with anatomical variations that improved their abil- 7
ity to stand upright had a foraging advantage as fruits became
more rare, they might have been at a selective advantage. The Figure 3 Dates of species of hominins with representative skeletons
from each of the major stages of human evolution. Species in the genus
second hypothesis, which is not exclusive of the first, is that Homo are in black, species in the genus Australopithecus are in white,
bipedalism was selected to improve locomotor efficiency as and early hominin species and genera are in gray.

Volume 5, January 2015 101

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

Ancestral adaptations for arboreality Derived adaptations for walking Derived adaptations for running
(partial) (partial) (partial)

Enlarged ant. & post.

Superiorly-oriented semicircular canals
shoulder
Vertically-oriented Nuchal ligament
Long arms neck
Narrow waist
Short, stiff Long, curved Enlarged cranial
lumbar region lumbar spine gluteus maxmius
Tall, posteriorly- Sideways-facing
oriented ilia Larger hip joint
pelvis
Long, curved
Large hip joint
phalanges Larger knee joint
Knees Angled
Short achilles under hips Elongated Achilles
tendon tendon
Large knee joint
Mobile ankle Short toes
Large heel bone
Long, curved Full arch
Partial arch
phalanges
Chimpanzee Au. afarensis H. erectus

Figure 4 Comparison of major locomotor features in chimpanzees, australopiths, and humans. Highlighted are ancestral
features for arboreal locomotion in chimpanzees, derived features for walking in Australopithecus, and derived features for running
in Homo.

adaptations observable in modern humans. To understand conjectural. Ar. ramidus also has numerous chimp-like adap-
human bipedalism therefore requires considering each of tations for climbing trees such as a very divergent hallux, long
these stages, beginning with the earliest known hominins. and curved phalanges, a slightly inverted subtalar joint, and
relatively long arms. A partial foot that is similar to Ardi’s,
but dated to 3.4 Ma, suggests that this type of foot persisted
Stage 1: Earliest hominins for at least a million years (57).
To what extent and how the earliest hominins were bipedal
The oldest proposed hominin species is Sahelanthropus is difficult to reconstruct. Without postcranial evidence, we
tchadensis, whose remains have been found in Chad in sedi- can establish only that Sahelanthropus was a postural biped;
ments dated to between 6.0 and 7.2 Ma (17, 18, 82, 168). The in addition, Ar. ramidus lived approximately 1.5 million years
only material described so far for Sahelanthropus is a partial after the first putative bipedal hominins. If we assume that Ar.
cranium, several mandibular fragments, and numerous teeth, ramidus was representative of earlier hominin species, then
but the cranium is almost surely that of a biped because its it is not unreasonable to reconstruct the earliest hominins as
inferiorly oriented foramen magnum is nearly parallel relative being a combination of arboreal climbers and bipedal walk-
to the long axis of the orbits, indicating a vertical upper neck, ers. Analyses of their feet suggest they probably walked like
hence habitual upright posture (192). A second species from chimps on the lateral margin of the foot (95), but there are not
Kenya, Orrorin tugenensis, is dated to 6 Ma, and includes a yet enough data to infer reliably whether they had inefficient
partial femur that has several features typical of later bipedal bent-hip bent-knee gaits or more extended lower limbs dur-
hominins (4, 113, 126). Finally, two early species assigned ing walking. In other words, they may have been occasional
to the genus Ardipithecus have been found in Ethiopia: Ar. or facultative bipeds. Even so, whatever form of bipedalism
kadabba (dated to 5.2-5.8 Ma) and Ar. ramidus (dated to 4.3- they practiced was clearly different from their more habitually
4.5 Ma) (58,150,181,182). Little is known about Ar. kadabba, bipedal descendents who are classified in the genus Australop-
but among the fossils assigned to Ar. ramidus is a partially ithecus, and who represent the second major stage in hominin
complete skeleton (nicknamed “Ardi”) that has numerous fea- locomotor evolution.
tures indicative of bipedalism. The most important of these
is the pelvis, which, although distorted, was short and prob-
Stage 2: Australopiths
ably had laterally-oriented ilia, permitting the small gluteal
muscles to function as hip abductors (in apes, these muscles The australopiths (the colloquial term for species in the genus
function primarily as extensors) (96). The Ar. ramidus foot Australopithecus) comprise a diverse group of hominins that
also has several indications of a partially rigid midfoot, and lived in Africa between 4 and 1 Ma (see Fig. 3). Much of the
as well as hyperextensible phalanges (95). It has been argued variation among australopith species is craniodental, reflect-
that Ar. ramidus has other adaptations for bipedalism such as ing adaptations to diverse diets, but recently recovered evi-
a long, lordotic lumbar region and an inwardly angled femur, dence reveals that these species also varied appreciably in
but these features are not preserved in the skeleton and remain terms of locomotion. The three species for which we have the

102 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

best data are Au. afarensis, which lived in East Africa from species: H. habilis and H. erectus. Limited postcranial remains
3.9 to 3.0 Ma; Au. africanus, which lived in South Africa from attributed to H. habilis, which may be as old as 2.3 Ma and
3 to 2 Ma; and Au. sediba, which is found in South African persisted until 1.4 Ma, suggest that body size and proportions
sites dating to 1.8 to 2.0 Ma (for reviews, see Refs. 46 and in this species were similar to Australopithecus, but an isolated
89). As summarized in Figure 4, all of these species retain partial foot (OH 8) that might be from H. habilis has numerous
adaptations evident in great apes and earlier hominins for tree derived features including a definitive longitudinal arch (59).
climbing including relatively short legs, long arms, and long, H. erectus, in contrast, is much more like modern humans in
curved pedal phalanges; but they also had more modern feet terms of its postcranial anatomy. Although body size varies
than Ardipithecus including an adducted hallux, and a partial enormously in the species (body mass estimates range from
longitudinal arch; in addition, they had a long lumbar region 40-70 kg), they had relatively long limbs (118), relatively large
with a strong lordosis that positioned the body’s center of joints in the lower extremity and spine (72, 135, 139), basin-
mass above the hips; a wide pelvis with efficient hip abduc- shaped pelves with relatively large gluteal muscles (92), and
tors; medially positioned knees; and enlarged lower extremity narrow waists. Footprints that were likely made by H. erectus
joints to withstand the higher stresses caused by bipedalism are extremely modern, indicating a full arch with short toes,
(see Refs. 2, 60, 172). The greatest variation appears to be and a long striding gait (8, 41). In addition, many adaptations
in the foot and ankle. Whereas the posterior calcaneus (heel for arboreal locomotion that were present among australopiths
bone) in Au. afarensis is large and inferiorly flat like a modern are absent in H. erectus. Although H. erectus postcrania differ
human’s to stabilize the foot during heel strike, it is small and in some respects from those of modern humans, most notably
triangular in cross-sectional shape in Au. sediba, which prob- in having more flared ilia, their overall locomotor anatomy is
ably walked on a more inverted foot (39). Other features in similar enough to infer that they walked and ran like living
Au. sediba such as its narrow thorax and upwardly-oriented humans (Fig. 4).
shoulder joint also would have benefited climbing perfor- There are several alternative hypotheses to explain the
mance (28). shift to modern locomotor anatomy that occurred across the
The general picture of the australopiths is that they were transition from Australopithecus to Homo. One hypothesis
habitual and effective walkers and climbers, but that some is that australopith bipedalism was partially apelike with an
species such as Au sediba may have been more arboreally inefficient bent-hip bent-knee gait that was necessitated by
adapted than others such as Au. afarensis and Au. africanus. A retained adaptations for arboreal locomotion, and that the
group of species termed the “robust” australopiths (some pale- transition to Homo was driven by selection to improve walk-
ontologists group them in a separate genus, Paranthropus), ing efficiency (147,160). A corollary of this hypothesis is that
which are characterized by craniodental adaptations for chew- selection for long-distance walking efficiency came at the
ing very mechanically demanding food, also appear to have expense of adaptations for arboreality such as short legs, and
been effective, habitual bipeds, but there are hints they might long, curved toes. This view, however, has been challenged
have differed from earlier, more gracile species such as Au. by several lines of evidence including the orientations of tra-
africanus and Au. afarensis in subtle ways (2, 158, 159). The beculae in fossil distal tibia of Au. africanus, which indicate
likely overall diversity among australopiths makes sense given that this species loaded its ankles, hence its knees, in extended
what is known about ongoing climate change in Africa during postures unlike those used by chimps (6). In addition, foot-
this period, the Pliocene (5.3-2.8 Ma) (see Fig. 2). As Africa prints dated to 3.6 Ma from Laetoli, Tanzania that were likely
continued to become cooler and drier during the Pliocene, made by Au. afarensis are consistent with modern striding
albeit with many swings back and forth, there was consider- gaits (33, 121). Finally, some other features characteristic of
able variation within and between regions, often because of australopiths such as long toes and long femoral necks would
intense tectonic activity (26,38,76,111). Although fruits must not have compromised walking performance (132), and there
have remained an important part of their diet, the australop- is no evidence that selection for improved walking perfor-
iths were probably under intense pressure to exploit non-fruit mance would have selected against upper limb features that
plant foods, many of them very tough and fibrous, and which are useful for climbing but which have little to no effect on
required more travel time to acquire in perilous open habitats walking such as long forearms, curved manual phalanges, and
(for review, see Ref. 90). It, therefore, makes sense that many superiorly oriented shoulder joints.
australopith species were under selection to be efficient at A second hypothesis regarding the locomotor differences
long-distance walking but also to retain arboreal adaptations between Australopithecus and Homo is that this transition
for access to fruits as well as protection from predators. was partially driven by selection for endurance running (15).
In terms of skeletal anatomy, several lines of evidence sup-
port this hypothesis. Most importantly, species in the genus
Stage 3: Homo
Homo, especially beginning with H. erectus but also including
The third and last major stage in hominin locomotor evolution H. sapiens, have numerous features that would have improved
occurred in the Genus Homo (see Figs. 3 and 4). Substantial performance in running but not walking. Since the predomi-
variation has led to much confusion over the taxonomy of nantly mass-spring gait mechanics of running differ markedly
early Homo, but most experts recognize at least two major from the primarily pendular mechanics of walking, a large

Volume 5, January 2015 103

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

proportion of these features include potential adaptations for walking possibly came at the expense of some performance
storing and releasing elastic energy such as a long Achilles abilities in the trees. Finally, running was probably selected
tendon (inferred indirectly from the small apelike insertion for as open grassland habitats continued to expand as the Ice
on the posterior calcaneus in Australopithecus), and evidence Age began between 3 and 2 Ma (see Fig. 2). Expansion of
for a close-packed calcaneo-cuboid joint, which helps the lon- these habitats favored the evolution of many herbivores, in
gitudinal arch function as a spring (2, 84). Many additional turn leading to selection for a variety of carnivores, of which
novel features in the genus Homo may be adaptations for Homo appears to be just one example (166). Since bipedal
stabilization during running, which is a more serious chal- hominins are necessarily slow (two legs can generate half the
lenge than during walking. These features include relatively power of four legs), endurance running may have been an
larger anterior and posterior semicircular canals than in apes adaptation to enable bipedal hominins to hunt herbivores in a
and australopiths, which help the head sense and adjust to novel way known as persistence hunting, which also requires
the rapid pitching forces generated by running (56); nar- adaptations for dumping heat. However, before evaluating
row waists and shoulders that are largely decoupled from hypotheses for how selection may have favored the transitions
the head, and which allow the torso to rotate independently of described above, it is first necessary to review the derived ther-
the pelvis and head; a nuchal ligament, which helps passively moregulatory features that enable active humans to keep cool,
stabilize the head during running (90); and an expanded cra- and how and when these adaptations might have evolved.
nial portion of the gluteus maximus to counter high pitching
forces on the trunk (92). Another set of derived skeletal fea-
tures relevant to running performance may be adaptations to Heat Exchange
cope with the higher internal and external forces generated by
running. These features including relatively larger joint sur- Compared to most mammals, humans have an impressive
face areas in the lower extremities and lower spine (72, 135); ability to keep cool during strenuous physical activity in hot
short toes, which decrease moments around the metatarsopha- conditions. In fact, as argued below, these capabilities may
langeal joints (132); and a shorter femoral neck (94). have played an important role in hominin evolution, first help-
Below we will consider why selection might have favored ing slow, unsteady bipeds forage over long distances when
adaptations for long-distance running in Homo, but there is predators were less likely to hunt them, and then helping
reason to speculate that climbing performance must have been hominins become diurnal predators themselves. Before evalu-
compromised by some of these features, notably low shoul- ating these hypotheses, it is useful to first review the evolution
ders that have fewer muscular connections to the upper spine of four derived sets of adaptations for preventing hyperther-
and head, relatively shorter forearms, shorter and straighter mia in humans: an increased ability to sweat; an external nose;
toes, and relatively longer legs (2). Selection for endurance enhanced ability to cool the brain; and an elongated, upright
running may therefore explain why humans are the only pri- body.
mate that is poorly adapted for arboreal locomotion. That said,
not all derived features in Homo are adaptations for running;
some of them would also have benefited walking, and some Sweating
might be related to other functional tasks such as throwing or One of the most distinctive aspects of human anatomy and
tool-making (127). physiology is an increased capacity to cool through sweating.
Altogether, each of the three major transitions in hominin Sweating cools via evaporative heat loss when secreted water
locomotion (facultative bipedalism and climbing in early (sweat) vaporizes on the surface of the skin. Since it requires
hominins, habitual walking and facultative climbing in aus- 580 calories (2426 J) to transfer 1 g of water at 35◦ C to water
tralipiths, and walking and running in the genus Homo) was vapor at the same temperature, this phase transition transfers
contingent on previous events, was driven by climatic change, considerable energy in the form of heat from the organism
and involved trade-offs. It is worth reiterating that selection to the atmosphere, thus cooling the body surface where the
for bipedalism was probably spurred by fragmentation of rain phase shift occurs.
forest habitats and might not have been advantageous had the Although sweating is an extremely effective method of
LCA not been relatively inefficient (a hypothesis that is depen- cooling, it is a specialized form of heat exchange that derives
dent on how one reconstructs the LCA). Whatever its initial from other forms of evapotranspiration, notably panting. Con-
benefits, habitual bipedalism resulted in a loss of stability sequently, before discussing the origins and evolution of
and speed among early hominins. Selection for more dedi- sweating, it is useful to first consider panting, which is the
cated long distance walking in Australopithecus took place in primary mechanism of cooling in all non-human mammals.
the context of increasing cooling and drying of Africa over Panting occurs from the evaporation of water in the upper res-
the Pliocene (26, 111). Such conditions would have favored piratory tract, primarily in the oral cavity, the oropharynx, and
hominins better able to walk long distances to forage for the upper portion of the trachea. Since the respiratory epithe-
widely dispersed foods in increasingly open, arid, and hot lium of the pharynx is highly vascularized, evapotransipra-
habitats. Although the australopiths retained many adapta- tion during panting efficiently cools blood, hence core body
tions for arboreal locomotion, selection for more efficient temperature. Panting is extremely efficient and effective, but

104 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

has several constraints. First, panting exchanges body heat

only on the surface of the respiratory tract, whose area is lim-
ited even in animals that expand respiratory surface area with
nasal turbinates, elongated snouts, or protruded tongues. Sec-
ond, since panting primarily occurs in the dead space of the Epidermis
respiratory tract during shallow breaths, panting efficiency is
elevated primarily by increasing the frequency of respiration
while decreasing tidal volume (129, 146). However, rapid, Dermis Eccrine
shallow breaths cause CO2 buildup in the lungs, risking alka- gland
losis and requiring the alternation of panting breaths that ven-
tilate only dead space with deep, inspiratory breaths. This Apocrine Sebaceous
gland Hair
alternation, moreover, is prevented in galloping quadrupeds gland
follicle
in which fore-aft tilting of the body causes the viscera to
oscillate forcefully in phase with stride frequency, inhibiting
Blood vessels
diaphragmatic contractions between strides (14). As a result,
most quadrupeds cannot gallop for long distances in hot con- Figure 5 Schematic of differences between eccrine and apocrine
ditions because they cannot pant while galloping, and thus glands (see text for details)
rapidly overheat (162).
Given the limitations of panting, it is hardly surprising
that natural selection has favored alternate means of heat Humans, like most mammals, have both apocrine and
exchange through evapotranspiration on the skin’s surface. eccrine glands, but differ in the relative distribution and num-
A few mammals such as rodents and kangaroos apply saliva ber of these gland types. Almost all mammals have apocrine
to their skin (36), but the most efficient strategy is to sweat, glands in certain regions of the skin in association with hair
which takes advantage of large surface areas. Heat exchange follicles but develop eccrine glands solely in the palms of
through sweating has only evolved in a few mammals because the hands and soles of the feet to increase frictional gripping
it requires at least three factors, all of which are derived in capabilities (1). Although apocrine glands probably evolved
humans: the ability to sweat enough but not too much water to produce odorants for olfactory and pheromonal commu-
on the skin’s surface, air convection at the skin’s surface, and nication, several groups of mammals evolved the ability to
effective conduction of blood below the skin’s surface. use apocrine glands for cooling by evapotranspiration. The
best-studied mammals are tropical ungulates such as cows,
sheep, camels, goats, and horses, which are known to sweat
Eccrine glands
when thermally stressed (19,145,183,190). Not all ungulates,
Sweating in which evapotranspiration occurs on the body however, use apocrine glands for thermoregulation, a capac-
surface is not unique to humans, but humans have a specially ity which appears to correlate with short fur (see below) and
elaborated system of cutaneous sweat glands that differs in body size; smaller, furrier mammals rely almost entirely on
several important respects from other mammals, including panting, while larger, short-haired animals have higher densi-
those that also sweat. To understand these differences it is ties of sweat glands to dump more heat (130). One interesting
necessary to distinguish between apocrine and eccrine glands exception to these trends is the Bedouin black goat, whose
(Fig. 5). Apocrine glands consist of a large, spongy-shaped black fur absorbs radiant heat, posing a special thermoregula-
secretory duct that is located deep within the dermis, com- tory challenge that may have triggered selection for copious
bined with a long, more tubular excretory duct that leads to sweat production by a relatively small number of apocrine
a hair follicle, often in conjunction with a sebaceous gland glands (13, 70). Marsupials such as kangaroos also use apoc-
(leading to the alternate term epitrichial gland, which means rine sweat for cooling, but only during vigorous physical
“by the hair”). Apocrine glands are controlled by sympa- activity (37).
thetic adrenergic nerves, and secrete in association with seba- The evolution of cutaneous glands in primates followed a
ceous glands a viscous fluid that includes lipids, proteins, and very different path. Like most small-bodied mammals, strep-
steroids. In contrast, eccrine glands are smaller than apocrine sirrhine primates (lemurs, lorises, and galagos) and New
glands, lie only in the outer portion of the dermis, and consist World Monkeys (platyrrhines) have mostly apocrine glands
of a coiled secretory tube and a relatively straight excretory over the general body surface and lose evaporative heat only
duct that extends to the skin’s surface. The excretory duct by panting (65). Although reliable data from most primate
of eccrine glands is formed by a double layer of epithelial species are not available, enough evidence exists to indi-
cells that are capable of resorbing some of the constituents of cate that Old World Monkeys (catarrhines) uniquely evolved
sweat, especially salt, thereby allowing eccrine sweat to aver- an elaborated system of eccrine glands, comprising approxi-
age about 99 percent water. Unlike apocrine glands, eccrine mately 50 percent of subcutaneous glands (53, 61, 71, 80, 104,
glands are not directly associated with hair follicles, and they 105, 141, 183). Data on gland distribution from apes are also
are innervated by sympathetic cholinergic nerves. limited, but several studies report that gibbons and orangutans

Volume 5, January 2015 105

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

resemble Old World Moneys, and that approximately two-

thirds of the cutaneous glands in chimpanzees and gorillas
are eccrine (43, 48, 105). Assuming these data are correct, the
fact that human cutaneous glands are nearly 100% eccrine Terminal hair
indicates that humans are most appropriately viewed as the
extreme of a trend among monkeys and apes. To what extent Vellus hair
the high density of human eccrine glands results from more
eccrine glands relative to other primates as well as from fewer
apocrine glands is poorly established but both factors may be Epidermis
important. Eccrine gland density in chimps is reported to be
50% of that in humans (105). In addition, although human
Sebaceous
embryos develop apocrine gland placodes throughout the gland
skin, these incipient glands atrophy and disappear everywhere
except the axillary and pubic regions (12, 142), where apoc-
Hair
rine glands still perform ancient communicative functions.
follicle
Although eccrine glands are likely denser and more
numerous in humans than in other primates, including apes,
their evapotranspirative function is nonetheless similar. Sev-
eral studies have demonstrated that non-human primates
sweat in response to heat and/or exercise stress (71, 98, 157)
and that acclimatization to thermal stress increases this
response in monkeys and apes as in humans (65, 141, 157). Figure 6 Schematic of differences between vellus and terminal hair
These studies did not separate the relative contributions of (see text for details).
apocrine and eccrine secretions to sweat, but the limited evi-
dence available suggests that total sweat rates in non-human
primates are much lower than in humans. Hiley (65) reported one famous experiment, Schmidt-Nielsen (144) found that
maximum secretion rates of 97 and 80 g/m2 /h in baboons and shearing a camel (fed ad libitum water) doubled its rate of
chimpanzees, respectively, similar to heat-induced apocrine water loss after being shorn.
secretion rates reported in ungulates (32-150 g/m2 /h), and far Humans are typically described as mostly hairless, but a
below the maximum values recorded in humans, which vary more accurate description is that they are mostly furless. The
between 366 to 884 g/m2 /h (47). It is therefore unlikely that average human has approximately 2 to 5 million hairs on the
non-human primates can match maximum sweating rates in body surface, with a density of 500 to 1000 follicles/cm2 in
humans, which typically exceed 1 L/h (48). It also unknown a neonate and 55 to 800 follicles/cm2 in adults with much
how effective non-human primate sweating is in terms of heat variation depending on body region, sex, hair color, and other
exchange. factors (170). However, with the exception of the scalp, axilla,
and pubic regions, almost all hair follicles in humans pro-
duce vellus rather than terminal hair (Fig. 6). Vellus hairs,
Fur loss
which are extremely fine and less than 2 mm long, grow from
One critical factor that influences the effectiveness of evapo- shallow follicles that are not connected to sebaceous glands
transpiration for heat exchange is hair, especially dense hair, (99). During ontogeny, vellus hair follicles are converted by
colloquially known as fur. Almost all mammals have fur, androgens (primarily dihydrotestosterone) to terminal folli-
which has multiple functions that include protection (e.g., cles, which are deeper, larger and connected to sebaceous
from bites and thorns), visual communication, camouflage, glands. Terminal hair is thus thicker, longer, and more pig-
reflecting radiant heat, and acting as an insulator by limiting mented. It is, therefore, incorrect to state that humans are
thermal conduction (heat transfer from the body surface to hairless. Further, it is possible that the density of vellus hair
the outside) and convection (the movement of air relative to follicles in humans follows predictions of scaling. As noted by
the body surface). However, the thickness and density of fur Schwartz and Rosenblum (149), terminal hair density in pri-
results in two trade-offs for sweating animals. First, evapo- mates scales with negative allometry relative to body surface
transpiration cools the body only if the phase transition from area, with larger bodied primates such as chimps and gorillas
water to vapor occurs along the skin’s surface, where it can having sparser terminal hair than gibbons or macaques. To test
cool underlying blood. Fur thus hinders effective cooling by whether humans fit this trend will require comparative data
moving the location of most evaporation away from the skin’s on scaling of hair follicle density rather than just terminal hair
surface. Second, fur inhibits or reduces air convection at the density.
skin’s surface, limiting the rate of evaporation where it has Despite much speculation over why and when humans
the highest cooling potential. As one would predict, mammals were selected to be generally furless, the most popular
that use sweat to cool tend to have short, sparse fur (128). In hypotheses focus on thermoregulation because a transition

106 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

from terminal to vellus hair removes the insulation that fur

normally provides and thus markedly increases air convec- Human
tion near the skin, allowing more heat exchange (3). One
hypothesis is that selection on early hominins in more open
habitats was made possible by the fact that large-bodied apes
such as chimpanzees already have relatively sparse hair and
thus lacked sufficient pelage for reflecting exogenous heat or Sphenoid
Meati
protecting the skin from harmful UV radiation (148). Under
such conditions, improved sweating efficiency led to selec-
tion for further fur loss as well as higher melanin content.
Soft Hard palate
A related hypothesis is that once hominins became bipedal,
palate
their bodies (except the tops of their heads) were exposed to
significantly less solar radiation, decreasing the benefit of fur
for reflecting radiation while simultaneously increasing the
benefit of fur loss to increase the rate and efficiency of evap-
orative cooling by sweating (176-179). Another hypothesis Chimpanzee
is that selection for either trekking or long distance running
promoted selection for fur loss because of the elevated need
to dump endogenous heat production caused by running and
walking (15, 136). These and other hypotheses are difficult to Sphenoid Meati
evaluate, however, in large part because we do not yet know
when hominins transitioned from having predominantly vel-
lus rather than terminal hair, and when eccrine gland distri- Soft palate Hard palate
bution was elaborated.

Subcutaneous heat convection Figure 7 Midsagittal comparison of the external and internal nasal
cavities in humans and chimpanzees. Note the lack of a rostrum in
Sweating exchanges heat only if evapotranspiration is able humans, combined with the presence of an external nasal cavity, which
to cool blood near the skin’s surface and then circulate the alters the orientation of flow through the nostrils, producing more tur-
bulent flow.
blood to the core, replacing it with more hot blood to be
cooled. Consequently, another variable that affects efficient
heat exchange from evapotranspiration is the extensiveness
and control of blood circulation in the dermal layer of the proportional to body mass). Although many primates includ-
skin. According to Montagna (Ref. 104: p. 16) “In no other ing chimpanzees have a generally shorter snout (rostrum) than
animal is skin so abundantly vascularized, not even in the great other mammals, the genus Homo lacks a snout altogether (see
apes.” Published evidence to support this statement, however, Fig. 7). As a result, non-human primates have significantly
is descriptive rather than quantitative (5) highlighting the need smaller nasal epithelial surface areas relative to body mass
for detailed studies of differences in dermal vascularization than most mammals, reflecting the generally warm, humid
between humans and non-human primates. environments in which they live (90). This ratio, however, is
considerably more extreme in humans, whose short, retracted
midface results in a nasal epithelium surface area one-tenth
Nasal cooling the expected value for a mammal of the same body mass, as
Most heat exchange occurs through evapotranspiration, but shown in Figure 8 (90). Because of a relatively short neck,
another set of derived adaptations for heat exchange in humans humans also have a similarly small tracheal surface area rel-
is in the nose, even though these adaptations are not regulated ative to body mass. Since the genus Homo appears to have
by reflexes and account for a smaller percentage of cooling. evolved in hot, arid habitats (26), it is unlikely that midfacial
Like most terrestrial vertebrates, humans use the respiratory shortening evolved in humans as a thermoregulatory adapta-
epithelium in the internal portion of the nose to warm or tion. Instead, the small surface area/volume ratio of the nasal
cool inspired air to approximately 37◦ C, and to add moisture epithelium must have imposed special challenges by limit-
to attain approximately 75% to 80% humidity by the time ing the ability to exchange heat and moisture. As one might
it reaches the lungs; about one-third of this heat and mois- expect, humans appear to have evolved several related adap-
ture is then recaptured during expiration in temperate con- tations for heat exchange to cope with these constraints.
ditions (30). One important difference between humans and The first set of adaptations is to increase turbulence. In
other mammals, however, is the ratio of the surface area of most mammals, airflow through the internal nose is primarily
nasal epithelium in the internal nose, where heat and moisture laminar, which has the advantage of generating less resis-
exchange occur, relative to the volume of airflow (which is tance but also creates a velocity gradient in which flow rates

Volume 5, January 2015 107

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

106 nose generates more turbulent airflow during exhalation, thus

helping to conserve water, because of two additional derived
features: the right angle formed between the nasopharynx
Turbinate surface area (mm2)

105
TSA = 2460Mb0.77 and the internal nose, and the aperture created by the internal
choanae (posterior nasal apertures) and the inner nose. The
104 extent to which these features increase turbulence is untested.
Another important thermoregulatory adaptation in
103 Human humans is obligate oral breathing during vigorous exercise.
As described by the Hagen-Poisseuille equation, resistance
in a tube increases in proportion to volumetric flow rate, and
102
Non-human primates ( ) inversely to the tube’s radius to the power of 4 during laminar
flow and to the power of 5 during turbulent flow (70). Increases
101 in flow rate and turbulence thus become a significant con-
10–3 10–2 10–1 100 101 102 103 straint during vigorous exercise when breathing rates more
Body mass (kg) than double to 40 breaths/min and tidal volume can triple to
106 1.5 L/breath (31). Although exercising humans decrease air-
flow resistance in the nasal cavity by increasing nostril diame-
ter and expanding the nasal cavity through vasoconstriction of
Tracheal surface area (mm2)

105
the respiratory mucosa (109,138), resistance rapidly becomes
TSA = 954Mb0.79 too great for the lungs to overcome, requiring humans to
104 switch to either oral breathing or combined oronasal breath-
ing during vigorous exercise (107, 175). Intriguingly, humans
103 Human are the only mammal species that switches to obligate oral
or oronasal breathing during vigorous activity (103) raising
102
the possibility it is an adaptation for dumping heat endurance
running, but at the expense of greater rates of water loss (15).
Unlike sweating and fur loss, nasal adaptations can be
101 partially traced in the fossil record, as shown in Figure 9.
10–3 10–2 10–1 100 101 102 103 The most important line of evidence is eversion of the lateral
Body mass (kg) margins of the nasal (piriform) aperture, which first appear in
early Homo fossils ascribed to H. habilis and H. erectus that
Figure 8 Scaling of surface area of the turbinates (top) and trachea
(bottom) relative to body mass in a wide range of mammals. Humans are dated to about 2 million years ago (51). This outward ori-
fall significantly below the line (Data courtesy of T Owerkowicz, sum- entation of the piriform margins, unique to Homo, indicates
marized in Ref. 90). that the nasal cartilages did not lie flat in the same plane as
the rest of the midface, as in apes such as chimpanzees, but
instead formed a protruding external nasal vestibule. Another
approach zero along the wall of the nose, forming an inert line of evidence for enhanced nasal turbulence is the nasal sill
boundary zone estimated to be 0.25 mm in humans (56). Since (also shown in Fig. 9), a discontinuity between the nasal aper-
this boundary is enough to reduce the respiratory epithelium’s ture and the inner nasal chamber that is absent in the African
capacity for heat and moisture exchange, it is not surprising great apes, but first appears in some species of Australop-
that the genus Homo evolved several features to increase tur- ithecus such as Au. afarensis, and is particularly pronounced
bulent flow in the nasal cavity, thus eliminating any laminar in Homo (75). More recent selection for nasal shape related
boundary and causing more air to flow across the epithelial to turbulence is evident among modern humans populations
surface. The most distinctive nasal feature that enhances tur- that originated in Africa but then adapted to diverse environ-
bulence is the external nose, unique to humans, in which a mental conditions over the last few hundred thousand years.
series of cartilages create a vestibular space that is mostly Many studies have documented that modern human popula-
devoid of epithelium, but which orients the nares (nostrils) tions which have long been living in more arid climates (cold
nearly 90◦ relative to the airway of the internal nose, increas- or hot) have significantly taller nasal apertures that are rel-
ing turbulence (29,143). A related set of features that increase atively more narrow and that have higher epithelial surface
turbulence are several pairs of valve-like constrictions, 30 area to nasal volume ratios (23, 50, 67, 185, 189).
to 40 mm2 in area, formed by the nostrils and between the
external and internal nasal chambers. These valves function
as Venturi throats (nozzles) that generate turbulence when Brain cooling
air transitions from low velocity and high pressure in the A major thermoregulatory challenge for all animals is to
external nasal cavity to high pressure and low velocity in the maintain a stable temperature in the brain, and it is com-
internal nose (169). There are also indications that the human monly argued that human brain cells can only briefly tolerate

108 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

Flat nasal Everted nasal

margin margin

Nasal
sill

Australopithecus africanus Homo habilis

Everted nasal
margin
Nasal
sill

Homo erectus Homo heidelbergensis

Figure 9 Comparison of the face in four hominin species showing variation in the nasal sills
and margins. Top left, Australopithecus africanus; Top right, Homo habilis; Bottom left, early
African Homo erectus; Bottom right, Homo heidelbergensis. In all species of Homo, the lateral
margin of the nasal aperture is everted, indicating the presence of an external nose.

temperatures above 41◦ C (e.g., Refs. 21 and 45). Some trop- counter-current exchange system in which cooled blood from
ical ungulates and carnivores evolved a carotid rete, in which the superficial cortical and ophthalmic veins passes around
cooled venous blood from the nasal cavity exchanges heat hotter arterial blood arising from the core. A final hypothesis
with carotid blood through a counter-current flowing anas- is that the expanded thickness of spongy bone (diplöe) in the
tamotic network, but retes never evolved in primates. Fur- cranial vault of Homo acts as a thermal insulator keeping the
ther, maintaining thermal homeostasis in the brain is a special brain cool (90).
problem in humans for two reasons. First, humans must cool Further research is needed to test if humans evolved spe-
relatively more brain tissue than any other terrestrial mammal cial mechanisms for brain cooling and, if so, when they
because our brains are approximately five times larger than evolved. Falk (45) has shown that the frequency of emis-
expected for a mammal of the same body mass (101). Sec- sary veins more than doubled in genus Homo compared to
ond, humans are the only primate that regularly engages in Australopithecus. If reverse flow occurs in these veins, then
prolonged vigorous activity in hot conditions. Are humans it is possible this cooling mechanism underwent selection as
adapted to cooling the brain simply by cooling the core humans became more active runners or trekkers (see below).
through sweating, or did hominins evolve additional mecha- An alternative or additional hypothesis is that the evolution
nisms for neural cooling, and what role did these adaptations of these cooling mechanisms released constraints on the evo-
play in human brain evolution? lution of larger brains in active hominins (44).
One uniquely human adaptation may be enhanced heat
exchange in the head. According to several studies, the scalp
has one of the highest densities of eccrine glands in the body Posture and body shape
(21,80), and it has been proposed that blood cooled by sweat- A final category of features relevant to how hominins coped
ing in the scalp actually flows backward into the brain through with heat stress is posture and variations in body shape. The
tiny emissary veins, thus acting as a specialized, regional cool- most influential hypothesis, proposed by Wheeler, is that
ing system (20, 45, 191). This hypothesis, however, is contro- upright posture and locomotion was an adaptation to reduce
versial, and has yet to be supported by in vivo data. Another exogenous heat gain from solar radiation (176). When the sun
untested hypothesis is that humans have an expanded cav- is at its zenith, a bipedal human exposes only 7% of its sur-
ernous sinus compared to other primates (22). Such expan- face area to maximal radiation, approximately one-third the
sion would also be advantageous because it functions as a maximally exposed surface area of a similar-sized quadruped

Volume 5, January 2015 109

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

Cold-adapted Heat-adapted

Figure 10 Schematic of Bergmann and Allen’s rules, comparing cold- versus hot-adapted body
forms (modified, with permission, from Ref. 134). The cold-adapted body form has a much lower
ratio of surface area to volume.

(177, 179). Wheeler also argued that bipedalism is adaptive which the surface area to volume ratio remains constant
for sweating by elevating the torso, forelimbs and head higher independent of height as long as the radius is constant (134).
above the ground surface where wind speed is greater and tem- Among human populations, the correlation between latitude
peratures are lower (178). Wheeler’s model clearly predicts and hip width is approximately 0.90, independent of variation
that furless, sweating hominins would have had a thermoreg- in stature (134), but body mass correlates with latitude
ulatory advantage if they were upright bipeds, but it is unclear only for extreme comparisons (49). These findings help
if early hominins were furless, how much radiation they were explain why species in the genus Homo such as early African
exposed to, and to what extent they were active at midday, H. erectus that lived in open, hot habitats have relatively
when most animals in open habitats seek shade. It is there- narrow pelves, whereas archaic Homo species such as
fore unclear whether bipedal posture evolved to facilitate heat Neanderthals that lived in Ice Age Eurasia have relatively
exchange among the first hominins, or whether sweating and wide pelves (2). Allen’s rule has also been shown to explain
fur loss evolved later in human evolution, perhaps in the genus much ecogeographical human variation, with populations
Homo. from relatively warmer climates having longer limbs relative
Related factors that influence heat exchange are body size to body mass than those from colder habitats (164). Since
and limb length (see Fig. 10), both of which affect surface most variation in upper and lower extremity length is caused
area to volume ratios because the volume of an object such by variation in humerus or femur length relative to body
as a sphere scales to its radius to the power of three (4/3πr3 ) mass, the ratio of the radius to humerus (brachial index) and
but the surface area scales to its radius to the power of two the tibia to femur (crural index) are strongly correlated with
(4πr2 ). As classically formulated, Bergmann’s rule states that latitude in both extant and fossil human populations (165).
within homeothermic species, populations in colder regions
will be larger than those in warmer regions in order to retain
heat by minimizing surface area to volume ratios. Because Evolutionary scenarios linking
extremities such as limbs, tails and ears have high surface area
to volume ratios, a related ecogeographical trend is Allen’s
locomotion and heat loss
rule, that populations in colder regions have relatively shorter Although the first hominins were probably either occasional
extremities than populations in warmer regions. Although or habitual bipeds, special human abilities to walk and run
Bergmann’s rule has been shown to apply to many birds and long distances evolved subsequently over the course of sev-
mammals (103), it is only partly relevant to hominins because eral major transitions, with more efficient walking probably
bipedal body shape is better approximated as a cylinder in being selected for in some species of Australopithecus and

110 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

endurance running then evolving in Homo possibly along its benefits for feeding outweighed any costs for locomotion
with additional improved walking capabilities. As summa- (32, 33, 163). In either case, incipient bipedalism is likely to
rized above, we do not know when hominins developed have been intermediate in efficiency, and would not have nec-
the ability to cool effectively through sweating by increas- essarily entailed selection for enhanced abilities to cool. Put
ing the density of eccrine glands and losing fur, but other differently, selection for early bipedalism should not to be
improved abilities to lose heat can be seen to a limited extent in confused with later selection for more efficient bipedalism
Australopithecus (notably body shape) and then to a consid- and possibly thermoregulation when hominins did eventually
erable extent in the genus Homo. Given the many functional occupy very open habitats. Testing this hypothesis will require
roles of hominin locomotor and thermoregulatory capabilities, better data on the habitats and locomotor capabilities of early
these adaptations must have evolved partly independently as hominins.
a result of multiple selective pressures. However, there are Natural selection often favors efficiency, but an additional
several reasons to speculate that selection for long distance important selective factor for hominins in open habitats must
walking and then running also drove selection for more effec- have been predation. Bipedal hominins are necessarily slow
tive cooling. Two hypothetical scenarios best fit the evidence. because two legs generate approximately half as much power
as four legs. The world’s fastest sprinters can achieve top
speeds slightly in excess of 10 m/s for only short durations,
Hypothesis 1: Predator avoidance while walking roughly half the speed of equivalent-sized quadrupeds with
in open habitats much shorter legs (52). It is thus reasonable to infer that
One longstanding and very common hypothesis is that selec- hominins who had to forage by walking long distances in
tion for living in more open, hot and arid habitats drove open habitats were easy prey for predators such as lions and
selection for hominins to become more efficient walkers in sabertooth tigers. In this context, enhanced thermoregulatory
order to forage for more widely distributed food and simul- abilities might have been strongly favored by natural selection
taneously drove selection for more efficient heat dumping to because they would have enabled hominins to preferentially
cope with the thermoregulatory challenges of trekking in the forage during the hottest times of the day when predators
heat. There are several problems with this popular hypothe- are generally constrained to rest and cannot run very far at
sis (often termed the savanna hypothesis). First, although the high speeds. In other words, increased abilities to dump heat
earliest hominins may have evolved in less densely forested might have coevolved with increased walking efficiency to
habitats in which fruits were scattered in smaller, more dis- increase foraging safety and efficiency in hot, open habitats
tant patches and thus required more walking, the anatomy with sparsely distributed resources. Testing this hypothesis
of the partial Ardipithecus skeleton and other early hominin will be difficult, because its major prediction is that high
fossils suggest that they were adapted for a combination of densities of eccrine glands and lower densities of terminal
climbing plus bipedal walking. More data are needed on early hair evolved in the genus Australopithecus or possibly earlier.
hominin anatomy, but many adaptations for increased bipedal Also needed to test this hypothesis are better data on the
efficiency such as an adducted big toe or long legs, do not habitats in which these hominins lived, as well as reliable
appear until later. In addition, although there is debate over estimates of their day ranges (see Ref. 116).
how open the habitats of early hominins such as Ardipithecus,
Sahelanthropus and Orrorin were, they were not open savan-
nas without a substantial degree of tree cover (25, 26, 180). Hypothesis 2: Hunting and gathering
Another problem with the savanna hypothesis as an expla- Another hypothesis that has received considerable attention is
nation for the origins of bipedalism and fur loss is the inabil- that selection for increased locomotor and thermoregulatory
ity to explain why these derived features evolved in just efficiency occurred primarily in the genus Homo because
hominins and not in other mammals in the same habitats. of the evolution of hunting and gathering. The hunting and
As noted previously, some African ungulates did evolve elab- gathering economic system is complex and multifactorial,
orated apocrine sweat capabilities along with carotid retes and involving a combination of long distance foraging, hunting,
expanded nasal turbinates (34, 128). However, bipedalism is tool-making, and intense cooperation such as food-sharing
an unusual form of locomotion with substantial costs in terms and division of labor (74, 100). Although foraging and
of speed and stability, and which has evolved contingently some degree of tool-using were almost certainly part of
only a few times. It is hard to imagine how other African the australopith behavioral repertoire, there are multiple
mammals in open habitats would have benefited from switch- lines of archaeological and paleontological evidence that the
ing to bipedal locomotion had they not been descended from hunting and gathering system evolved as a whole in the genus
African great apes. As discussed above, a reasonable hypoth- Homo (see Refs. 78 and 91). Species of Homo, especially
esis is that bipedalism was selected in early hominins primar- starting with H. erectus, also differ morphologically from
ily because they evolved from a knuckle-walking ancestor australopiths in many ways likely relevant to hunting and
whose cost of transport was approximately four times greater gathering.
than most other mammals, including humans (154). Alterna- Since the primary gait of hunter-gatherers is walking, the
tively or additionally, bipedalism may have evolved because most common hypothesis invoked to explain these differences

Volume 5, January 2015 111

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

is that the genus Homo was selected for long distance walking generate relatively more body heat when running. Persistence
(trekking), which would have also favored sweating and fur hunts usually involve alternating chasing and tracking phases.
loss in open habitats. Evidence to support this hypothesis During the chasing phase, runners run after their prey at a
includes many derived features in H. erectus that would have speed that makes the animal gallop and thus gain heat; during
improved walking performance such as long legs, relatively the tracking phase, which occurs after the prey has galloped
larger lower extremity joints, and essentially modern feet (for away from the hunter, the hunter tries to follow and locate
review, see Ref. 2). Although we do not know when enhanced the animal, usually at a walking pace, while the prey seeks
sweating and the loss of fur evolved, external noses, which shade and rests. If the hunter can resume chasing the ani-
play vital roles in heat and moisture exchange during walking mal before it has recovered a normal core body temperature,
(but not running) also first appear in the genus Homo (51). In then the prey’s core body temperature will keep rising until
addition, there is no question that the habitats in which early it reaches a hyperthermic state, at which point the hunter can
Homo evolved in Africa were hot, arid and open. Regardless dispatch it from a close distance without danger or sophisti-
of debates whether australopiths were inefficient, bent-hip- cated weapons. Persistence hunting thus requires the hunter
bent-kneed bipeds, it seems reasonable to hypothesize that to able to run at speeds that make quadrupeds gallop, to track,
there would have been strong selection among hunting and and to keep cool without dehydrating. Further, contrary to
gathering hominins to walk as efficiently as possible and to some misconceptions (112, 156), PH does not require the
tolerate thermal stress, especially if they were active during ability run 35 to 40 km without stop, but instead involves
the midday to avoid predators (see above). roughly equal proportions of walking and running at moderate
An additional, related hypothesis that is not entirely speeds over distances ranging from 15 to 40 km. The greatest
exclusive with selection for trekking is that there was strong physiological constraint for humans is water. Yet according
selection in Homo for long distance running. As noted above, to ethnographic accounts by Liebenberg (86, 87), Kalahari
humans have superlative abilities to run long distances, in Bushmen are able to hunt this way without dehydrating in
large part because of an extensive suite of adaptations for run- part by drinking copiously before starting a PH. When and to
ning that have little or no effect on walking performance. Such what extent runners carried water in containers such as ostrich
adaptations include elongated tendons such as the Achilles, eggshells or gourds is unknown. Note also that endurance run-
relatively short toes, an expanded cranial portion of the glu- ning is only 30% to 50% more costly than walking (133), but
teus maximus, a narrow waist, a nuchal ligament, and enlarged that the energetic returns from hunting are typically orders of
anterior and posterior semicircular canals (15, 92, 132, 155). magnitudes higher (100), especially for persistence hunting,
In addition, running is considerably more thermogenic than which has a much higher success rate than bow and arrow
walking, yet novel heat exchange adaptations make humans hunting (87). Consequently, persistence hunting would have
unique among mammals in being able to run long distances been a beneficial strategy among even energy-limited hunter
in hot conditions (89). It follows that although walking is gatherers.
unquestionably important in human evolution, walking alone Another hypothesized advantage for the evolution of
is unable to explain the combination of unique locomotor derived endurance running and thermoregulatory capabilities
and thermoregulatory adaptations that evolved in the genus in hominins is scavenging. All carnivores including modern
Homo. hunter-gatherers sometimes scavenge, but competition for
A potential explanation for these evolved capabilities is carcasses is intense and also requires speed and fighting
that there was also selection for the ability to hunt by chasing (167). One hypothesis is that hominins first started to incor-
animals at running speeds over long distances in the heat, porate meat in their diet by scavenging for carcasses in open
a strategy known as Persistence Hunting (PH). To appre- habitats, perhaps through cues such as circling vultures in
ciate the likely significance of PH, it is useful to consider the distance (15). Since hyenas, like other quadrupeds avoid
that there is abundant evidence that early Homo was hunting running during the midday heat, hominins with the ability to
large, mature bovids more than 2 million years ago without run without overheating would have had an advantage scav-
any of the deadly projectile technologies employed by recent enging for carcasses at times of low competition with other
hunter-gatherers (16,42). Before the invention of stone points carnivores. This behavior is still practiced by modern hunter-
500,000 years ago (184) and the bow and arrow less than gatherers (100, 108, 153), and might have been an important
100,000 years ago (152), the most lethal weapons available to intermediate stage between foraging and hunting and
early Homo hunters were rocks and untipped wooden spears. gathering.
Consequently, hunters would have needed to kill prey at close Like many evolutionary hypotheses, the PH and scav-
range, which is extremely dangerous and thus avoided (27). enging hypotheses are difficult to test rigorously, but they fit
PH, however, takes advantage of human abilities to run long many lines of evidence. Most obviously, selection for scav-
distances in hot conditions at speeds that require their prey to enging and PH helps explain why many derived adaptations
gallop, thus driving them into hyperthermia (15, 24). As doc- for endurance running appear at about the same time as the
umented by Liebenberg (86, 87), persistence hunts usually oldest evidence for meat-eating in human evolution (15). In
occur during peak heat, often in temperatures above 30◦ C, addition, these adaptations also appear approximately when
and focus on large prey, presumably because larger animals both meat-eating and bigger brains evolved. Perhaps the

112 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

ability to hunt released a constraint on selection for relatively Recent evolution and contemporary
larger brains, which require plentiful energy and fat, which
are scarce resources. Additionally or alternatively, tracking
relevance
during persistence hunting employs complex cognitive skills Although species in the genus Homo spread over much of
that would have benefited from more encephalization. The the Old World during the Pleistocene, H. sapiens (modern
origins of the final component of this behavioral strategy— humans) evolved in Africa between 200,000 and 300,000
the ability to dump heat effectively from increased densities years ago, and then subsequently dispersed into Eurasia,
of eccrine glands and loss of fur—remains elusive. Without Oceania, and the New World over the last 100,000 years
knowing the genes underlying these adaptations and when (78). As modern humans moved into new habitats with differ-
they evolved, it is not yet possible to test whether selection ent ecologies and climatic conditions, they interbred a little
for improved heat exchange was driven by PH in Homo, with archaic human species such as Neanderthals (H. nean-
was made possible by previous selection for long distance derthalensis) that had inhabited these regions for long peri-
walking and predator avoidance in Australopithecus, or some ods of time, but for the most part modern humans replaced
combination of the two. Further, while selection for the archaic humans (124, 140). Since the first modern humans
ability to lose heat would have been important for male and were all African hunter-gatherers, primarily adapted to cope
female hunter-gatherers, it is not possible to test whether with walking and running long distances in hot, arid condi-
selection for endurance running was solely or primarily in tions, these pioneers must have faced serious thermoregula-
males. Males who hunt more effectively have been shown tory challenges as well as the need to change their locomo-
to have higher reproductive success than less skilled hunters tor behaviors as they dispersed into different environments.
(see Ref. 100), but females can also run well. Although Moreover, because this period was the most intense period
hunting is generally done by males not females (74), women of the Ice Age, the Late Pleistocene, the most extreme chal-
who are not mothers (e.g., teenagers) might have benefited lenges must have been those faced by early modern humans
from the ability to scavenge or persistence hunt. Alternatively in highly seasonal temperate habitats.
or additionally, adaptations for endurance running are not Modern humans in different regions coped in two ways.
sex-linked. The most potent adaptations were obviously cultural, made
possible by modern human abilities and proclivities to inno-
vate, communicate, and cooperate. Cultural evolution led
Other selective forces to a stunning variety of technologies and behaviors that
Despite the important interdependence between locomotion have enabled hunter-gatherers to live in almost every habi-
and heat exchange, other selective forces almost certainly tat, including the arctic. Although innovations such as com-
played roles in the evolution of human bipedalism, fur loss, plex clothing and shelters (and eventually the agricultural and
sweating and related adaptations. In addition to selection for industrial revolutions) partially buffered humans from natural
feeding and foraging (discussed above), hypothesized selec- selection, they accelerated natural selection in other respects
tive forces on bipedalism have included food carrying and by enabling people to live in novel habitats and by increas-
food-provisioning (96, 102), tool-using (35, 64, 173), the abil- ing population sizes, hence the number of available mutations
ity to see over tall grasses, and even swimming or wading on which selection can act (62). For example, humans in the
(106, 187). Similarly, in addition to selection for dumping arctic would not have been selected to have relatively shorter
excess heat during locomotion, the evolution of fur loss (but limbs had not their clothing, harpoon technology and other
not sweating) has been proposed to have arisen from selection innovations enabled them to survive in cold conditions in the
against ectoparasites (110, 122, 123), sexual selection (35), first place. Although approximately 86% of the genetic vari-
and aquatic habitats (106, 187). It is beyond the scope of ation in H. sapiens is within rather than between populations
this review to evaluate these hypotheses, which have varying (7, 85), the combined result of cultural evolution and natural
degrees of merit (for reviews, see Refs. 46, 60, 81, 91, 172). selection has been an integrated and interrelated combina-
Although some of these selective forces (and probably others) tion of cultural and physiological adaptations that contribute
were undoubtedly important factors in human evolution, it is to modern human diversity. Most of these regional variations
important to emphasize that selection for efficient long dis- reflect selection for disease and diet, but a few reflect selection
tance walking and running must have been partly contingent for thermoregulation (92, 137).
on and entailed additional selection for improved abilities to In terms of adaptation for thermoregulation, the most
dump heat. In the case of walking, bipedal hominins would conspicuous source of diversity is skin pigmentation, which
have been at a severe disadvantage walking long distances is strongly correlated with UV levels, probably reflect-
if they could not have done so during times of peak heat ing a trade-off between protecting the skin from radiation
when their inability to sprint rapidly would have placed them with inhibiting vitamin D synthesis (69). Another source of
at high risk of predation from lions, saber-toothed cats and climate-related variation is body shape and size. The earliest
other carnivores. In addition, the ability to persistence hunt is modern humans who moved into Europe came from Africa,
fundamentally based on the ability to keep cool while running, and thus it should be unsurprising that they were generally
which is extremely thermogenic. tall and relatively narrow, but over millennia selection drove

Volume 5, January 2015 113

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

European populations towards more cold-adapted shapes with remarkably little about the recent evolution of these anatomi-
larger body masses and relatively shorter limbs (66). Other cal and physiological features within different modern human
evidence for regional selection related to climate includes populations. Overall, an integrated, evolutionary approach to
variations in nasal shape (189), and eccrine gland density human locomotion and thermoregulation will help illuminate
(79, 174). Another candidate for ecogeographical selection how and why humans are the way we are.
is EDAR370A, an allele that arose in East Asia approxi-
mately 30,000 years ago that is associated with more numer-
ous eccrine glands and thicker hair, but which also affects Acknowledgements
other traits such as incisor shape and breast size (73).
I am grateful to Dennis Bramble, David Carrier, Yana Kam-
There is much less evidence for regional selection on loco-
berov, Bruce Morgan, Pardis Sabeti, Cliff Tabin, Sijia Wang,
motion. One well studied candidate is ACTN3, which affects
and Sara Wright for their insights and for many long discus-
the relative percentage of fast-twitch and slow twitch fibers,
sions about the topics reviewed here.
and which has a variant, R577X, that has been associated
among some (but not all) populations with high percentages of
fast twitch fibers better adapted for power and speed (188). It
is possible that as some human populations experienced more
References
selection for power relative to endurance as they moved into 1. Adelman S, Taylor CR, Heglund MC. Sweating on paws and palms:
novel habitats or (more likely) became farmers. Additional What is its function. Am J Physiol 229: 1400-1402, 1975.
2. Aiello L, Dean MA. An Introduction to Human Evolutionary Anatomy.
candidates for selection on locomotion include relative heel London: Academic Press, 1990.
length, toe length, and limb length (83, 115, 120, 132). Future 3. Al-Ramamneh D, Gerken M, Riek A. Effect of shearing on water
turnover and thermobiological variables in German Blackhead mutton
research is needed to test these and other related hypotheses. sheep. J Anim Sci. 89: 4294-4304, 2011.
4. Almécija S, Tallman M, Alba DM, Pina M, Moyà-Solà S, Jungers WL.
The femur of Orrorin tugenensis exhibits morphometric affinities with
both Miocene apes and later hominins. Nat Commun 4: 2888, 2013.
Conclusion 5. Baccaredda-Boy A. A note on the cutaneous arterial vessels of some
primates. Angiologica 1: 209-212, 1964.
6. Barak MM, Lieberman DE, Raichlen D, Pontzer H, Warrener AG,
In conclusion, through a series of contingent events, many Hublin JJ. Trabecular Evidence for a human-like gait in Australopithe-
cus africanus. PLoS One 8: e77687, 2013.
driven by climate change, hominins evolved a series of adapta- 7. Barbujani G, Magagni A, Minch E, Cavalli-Sforza LL. An apportion-
tions for bipedal walking and then running, which were proba- ment of human DNA diversity. Proc Natl Acad Sci U S A 94: 4516-4519,
1997.
bly the main impetus for additional selection for sweating, fur 8. Bennett MR, Harris JW, Richmond BG, Braun DR, Mbua E, Kiura
loss, and other adaptations for effective heat loss during vig- P, Olago D, Kibunjia M, Omuombo C, Behrensmeyer AK, Huddart
D, Gonzalez S. Early hominin foot morphology based on 1.5-million-
orous activity. Regardless of differences within and between year-old footprints from Ileret, Kenya. Science. 323: 1197-1201, 2009.
populations, all humans have remarkable abilities to walk 9. Berge C, Penin X. Ontogenetic allometry, heterochrony, and interspe-
cific differences in the skull of African apes, using tridimensional Pro-
and run long distances, and as a species we are generally well crustes analysis. Am J Phys Anthropol 124: 124-138, 2004.
adapted for endurance. In addition, because of the endoge- 10. Biewener AA. Scaling body support in mammals: Limb posture and
muscle mechanics. Science. 245: 45-48, 1989.
nous heat gain caused by endurance activities, combined with 11. Bobe R. Fossil mammals and paleoenvironments in the Omo-Turkana
our tropical origins, all humans are well adapted to lose heat. Basin. Evol Anthropol 20: 254-263, 2011.
12. Bolognia JL, Jorizzo J, Schaffer J Dermatology (3rd ed). Philadelphia:
Despite some recent selection and much sophisticated tech- WL Saunders, 2012.
nology, this unique and integrated suite of endurance and heat 13. Borut A, Dmi’el R, Shkolnik A. Heat balance of resting and walking
goats: Comparison of climatic chamber and exposure in the desert.
loss adaptations remain fundamental to human physiology. Physiol Zool 52: 105-112, 1979.
Although we have learned much over the last few decades 14. Bramble DM, Jenkins FA Jr. Mammalian locomotor-respiratory inte-
gration: Implications for diaphragmatic and pulmonary design. Science
about the evolutionary history of human locomotion and ther- 262: 235-240. 1993.
moregulation, considerable additional research is needed. For 15. Bramble, DM, Lieberman DE. Endurance running and the evolution of
Homo. Nature 432: 354-352, 2004.
one, there is a strong need for more fossils not just of early 16. Braun DR, Harris JW, Levin NE, McCoy JT, Herries AI, Bamford MK,
hominins, but also from the African great apes in order to Bishop LC, Richmond BG, Kibunjia M. Early hominin diet included
diverse terrestrial and aquatic animals 1.95 Myr ago in East Turkana,
test hypotheses about the anatomy of the LCA of humans Kenya. Proc Nat Acad Sci USA 107: 10002-10007, 2010.
and chimpanzees. We also need to develop better tools to test 17. Brunet M, Guy F, Pilbeam D, Lieberman DE, Likius A, Mackaye HT,
Ponce de León MS, Zollikofer CP, Vignaud P. New material of the
hypotheses about the nature of early hominin locomotion from earliest hominid from the Upper Miocene of Chad. Nature 434: 752-
fossils (e.g., from the structure of trabeculae within joints), 755, 2005.
18. Brunet M, Guy F, Pilbeam D, Mackaye HT, Likius A, Ahounta D,
and on the ecological contexts in which they were walking Beauvilain A, Blondel C, Bocherens H, Boisserie JR, De Bonis L,
and running. The evolution of human thermoregulation is Coppens Y, Dejax J, Denys C, Duringer P, Eisenmann V, Fanone G,
Fronty P, Geraads D, Lehmann T, Lihoreau F, Louchart A, Mahamat A,
even more murky, largely because the really key physiologi- Merceron G, Mouchelin G, Otero O, Pelaez Campomanes P, Ponce De
cal features that underlie them leave no fossil traces. The most Leon M, Rage JC, Sapanet M, Schuster M, Sudre J, Tassy P, Valentin
X, Vignaud P, Viriot L, Zazzo A, Zollikofer C. A new hominid from
promising avenues for research in this field will be to under- the upper Miocene of Chad, central Africa. Nature 418: 145-151, 2002.
stand the genetic and developmental bases for the loss of fur 19. Bullard RW, Dill DB, Yousef MK. Responses of the burro to desert
heat stress. J Appl Physiol 29: 159-166, 1970.
and elaboration of sweat glands in humans, and then study 20. Cabanac M. Selective brain cooling in humans: “fancy” or fact? Fed
the evolutionary history of these genes. And finally, we know Am Soc Exp Biol 17: 1143-1146, 1993.

114 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

21. Cabanac M, Brinnel H. Beards, baldness, and sweat secretion. Eu J 52. Garland T Jr. The relation between maximal running speed and body-
Appl Physiol Occup Physiol 58: 39-46, 1988. mass in terrestrial mammals. J Zool 199, 157-170, 1983.
22. Cabanac M, Caputa M. Natural selective cooling of the human brain: 53. Gisolfi CV, Sato K, Wall PT, Sato F. In vivo and in vitro characteristics
Evidence of its occurrence and magnitude. J Physiol 286: 255-264, of eccrine sweating in patas and rhesus monkeys. J Appl Physiol Respir
1979. Environ Exerc Physiol 53: 425-431, 1982.
23. Carey JW, Steegman, AT Jr. Human nasal protrusion, latitude and 54. Groves CP. A Theory of Human and Primate Evolution. Oxford: Oxford
climate. Am J Phys Anthropol 56: 313-319, 1981. University Press, 1989.
24. Carrier D. Energetic paradox of human running and hominid evolution. 55. Guy F, Lieberman DE, Pilbeam D, Ponce de Leon M, Likius A, Mack-
Curr Anthropol 25: 483-495, 1984. aye HT, Vignaud P, Zollikofer CP, Brunet M. Morphological affinities
25. Cerling TE, Levin NE, Quade J, Wynn JG, Fox DL, Kingston JD, Klein of the Sahelanthropus tchadensis (Late Miocene hominid from Chad)
RG, Brown FH. Comment on the paleoenvironment of Ardipithecus cranium. Proc Natl Acad Sci USA 102: 18836-18841, 2005.
ramidus. Science 328: 1105-1106, 2010. 56. Hahn I, Scherer PW, Mozell MM. Velocity profiles measured for airflow
26. Cerling TE, Wynn JG, Andanje SA, Bird MI, Korir DK, Levin NE, through a large-scale model of the human nasal cavity. J Appl Physiol
Mace W, Macharia AN, Quade J, Remien CH. Woody cover and 75: 2273-2287, 1993.
hominin environments in the past 6 million years. Nature 476: 51-56, 57. Haile-Selassie Y, Saylor BZ, Deino A, Levin NE, Alene M, Latimer
2011. BM. A new hominin foot from Ethiopia shows multiple Pliocene
27. Churchill SE. Weapon technology, prey size selection and hunting bipedal adaptations. Nature 483: 565-569, 2012.
methods in modern hunter-gatherers: Implications for hunting in the 58. Haile-Selassie Y, Suwa G, White TD. Hominidae. In: Haile-Selassie
Palaeolithic and Mesolithic. Arch Papers Am Anthropol Assoc 4: 11- Y, WoldeGabriel G, editors. Ardipithecus kadabba: Late Miocene Evi-
24, 1993. dence from the Middle Awash, Ethiopia. Berkeley: University of Cali-
28. Churchill SE, Holliday TW, Carlson KJ, Jashashvili T, Macias ME, fornia Press, 2009, pp. 159-236.
Mathews S, Sparling TL, Schmid P, de Ruiter DJ, Berger LR. The 59. Harcourt-Smith W, Aiello LC. Fossils, feet and the evolution of human
upper limb of Australopithecus sediba. Science 340: 1233477, 2013. bipedal locomotion. J Anat 204: 403-416, 2004.
29. Churchill SE, Shackelford LL, Georgi JN, Black MT. Morphological 60. Harcourt-Smith W. The origins of bipedal locomotion. In: Henke W,
variation and airflow dynamics in the human nose. Am J Hum Biol 16: Tattersall I, editors. Handbook of Paleoanthropology. Berlin: Springer,
625-638, 2004. 2007, pp. 1483-1518.
30. Cole P. Modification of inspired air. In: Proctor DF, Andersen HP, edi- 61. Harrison RJ, Montagna W. Man. New York: Appleton-Century-Crofts,
tors. The Nose: Upper Airway Physiology and the Atmospheric Envi- 1969.
ronment. Amsterdam: Eslevier Biomedical Press, 1982, pp. 351-375. 62. Hawks J, Wang ET, Cochran GM, Harpending HC, Moyzis RK. Recent
31. Courtiss EH, Gargan TJ, Courtiss GB. Nasal physiology. Ann Plast acceleration of human adaptive evolution. Proc Natl Acad Sci U S A
Surg 13: 214-223, 1984. 104: 20753-20758, 2007.
32. Crompton RH, Li Y, Thorpe SK, Wang WJ, Savage R, Payne R, Carey 63. Heglund NC, Taylor CR. Speed, stride frequency and energy cost per
TC, Aerts P, Van Elsacker L, Hofstetter A, Gunther MM, D’Aout K, De stride. How do they change with body size and gait? J Exp Biol 138,
Clerq D. The biomechanical evolution of erect bipedality. Cour Forsch 301-318, 1988.
Inst Senckenberg 243: 115-126, 2003. 64. Hewes G. Food transport and the origin of Hominid bipedalism. Am
33. Crompton RH, Vereecke EE, Thorpe SK. Locomotion and posture from Anthropol 63: 687-710, 1961.
the common hominoid ancestor to fully modern hominins, with special 65. Hiley PG. The thermoregulatory responses of the galago (Galago cras-
reference to the last common panin/hominin ancestor. J Anat 212: 501- sicaudatus), the baboon (Papio cynocephalus) and the chimpanzee (Pan
543, 2008. stayrus) to heat stress. J Physiol 254: 657-671, 1976.
34. Daniel PM, Dawes JDK, Pritchard MML. Studies of the carotid rete 66. Holliday TW. Body proportions in Late Pleistocene Europe and modern
and its associated arteries. Phil Trans R Soc Lond B 237: 173-208, 1953. human origins. J Hum Evol 32: 423-448, 1997.
35. Darwin CR. The Descent of Man, and Selection in Relation to Sex. 67. Holton NE, Yokley TR, Butaric LN. The morphological interaction
London: John Murray, 1871. between the nasal cavity and maxillary sinuses in living humans. Anat
36. Dawson TJ, Blaney CE, Munn AJ, Krockenberger A, Maloney SK. Rec 296: 414-426, 2013.
Thermoregulation by kangaroos from mesic and arid habitats: Influ- 68. Hunt KD. The evolution of human bipedality: Ecology and functional
ence of temperature on routes of heat loss in eastern grey kangaroos morphology. J Hum Evol 26: 183-203, 1994.
(Macropus giganteus) and red kangaroos (Macropus rufus). Physiol 69. Jablonski NG, Chaplin G. Epidermal pigmentation in the human lineage
Biochem Zool. 73: 374-381, 2000. is an adaptation to ultraviolet radiation. J Hum Evol 65: 671-675, 2013.
37. Dawson TJ, Robertshaw D, Taylor CR. Sweating in the kangaroo: A 70. Jenkinson DM, Robertshaw D. (Studies on the nature of sweat gland
cooling mechanism during exercise, but not in the heat. Am J Physiol ‘fatigue’ in the goat. J Physiol 212: 455-465, 1971.
227: 494-498, 1974. 71. Johnson GS, Elizondo RS. Thermoregulation in Macaca mulatta: A
38. deMenocal PB. Climate and human evolution. Science 331: 540-542, thermal balance study. J Appl Physiol 46: 268-277, 1979.
2011. 72. Jungers WL. Relative joint size and hominid locomotor adaptations
39. DeSilva JM, Holt KG, Churchill SE, Carlson KJ, Walker CS, Zipfel B, with implications for the evolution of hominid bipedalism. J Hum Evol
Berger LR. The lower limb and mechanics of walking in Australopithe- 17: 247-265, 1988.
cus sediba. Science 340: 1232999, 2013. 73. Kamberov YG, Wang S, Tan J, Gerbault P, Wark A, Tan L, Yang Y, Li
40. Dill DB, Edwards HT, Talbott JH. Studies in muscular activity: VII. S, Tang K, Chen H, Powell A, Itan Y, Fuller D, Lohmueller J, Mao J,
Factors limiting the capacity for work. J Physiol 77: 49-62, 1932. Schachar A, Paymer M, Hostetter E, Byrne E, Burnett M, McMahon
41. Dingwall HL, Hatala KG, Wunderlich RE, Richmond BG. Hominin AP, Thomas MG, Lieberman DE, Jin L, Tabin CJ, Morgan BA, Sabeti
stature, body mass, and walking speed estimates based on 1.5 million- PC. Modeling Recent Human Evolution in Mice by Expression of a
year-old fossil footprints at Ileret, Kenya. J Hum Evol 64: 556-568, Selected EDAR Variant. Cell 152: 691-702, 2013.
2013. 74. Kelly RL. The Foraging Spectrum: Diversity in Hunter-Gatherer Life-
42. Dominguez-Rodrigo M. Hunting and scavenging by early humans: The ways. Clinton Corners, New York: Percheron Press, 2007.
state of the debate. J World Prehist 16: 1-54, 2002. 75. Kimbel WH, White TD, Johanson DC. Cranial morphology of Aus-
43. Ellis RA, Montagna W. The skin of primates VI. The skin of the gorilla tralopithecus afarensis: A comparative study based on a composite
(Gorilla gorilla). Am J Phys Anthropol 20: 79-85, 1962. reconstruction. Am J Phys Anthropol 64: 337-388, 1984.
44. Falk D. Braindance (2nd ed). Gainesville: University Press of Florida, 76. Kingston JD. Shifting Adaptive Landscapes: Progress and challenges
2004. in reconstructing early hominid environments. Yrbk Phys Anthropol 50:
45. Falk D. Constraints on brain size: The radiator hypothesis. In: Kaas 20-58, 2007.
JH, Preuss TM, editors. The Evolution of Primate Nervous Systems. 77. Kivell TL, Schmitt D. Independent evolution of knuckle-walking in
Oxford: Elsevier Press, 2007, pp. 347-354. African apes shows that humans did not evolve from a knuckle-walking
46. Fleagle, JG. Primate Adaptation and Evolution (3rd ed). San Diego: ancestor. Proc Natl Acad Sci U S A 106: 14241-14246, 2009.
Academic Press, 2013. 78. Klein RG. The Human Career: Human Biological and Cultural Origins
47. Folk GE Jr. Textbook of Environmental Physiology (2nd ed). Philadel- (3rd ed). Chicago: University of Chicago Press, 2009.
phia: Lea and Febiger, 1974. 79. Knip AS. Ethnic studies on sweat gland counts. In: Weiner, JS, editor.
48. Folk GE Jr, Semken HA Jr. The evolution of sweat glands. Int J Biome- Physiological Variation and Its Genetic Basis. London: Taylor and
tereology 35: 181-186, 1991. Francis, 1977, pp. 113-123.
49. Foster F, Collard M. A reassessment of Bergmann’s rule in modern 80. Kuno Y. Human Perspiration. Springfield, IL: Charles C Thomas, 1956.
humans. PLoS One 8: e72269, 2013. 81. Langdon DH. Umbrella hypotheses and parsimony in human evolution:
50. Franciscus RG, Long JC. Variation in human nasal height and breadth. A critique of the Aquatic Ape Hypothesis. J Hum Evol 33: 479-494,
Am J Phys Anthropol 85: 419-427, 1991. 1997.
51. Franciscus RG, Trinkaus E. Nasal morphology and the emergence of 82. Lebatard AE, Bourlès DL, Duringer P, Jolivet M, Braucher R, Carcail-
Homo erectus. Am J Phys Anthropol 75: 517-527, 1988. let J, Schuster M, Arnaud N, Monié P, Lihoreau F, Likius A, Mackaye

Volume 5, January 2015 115

Evolution of Human Locomotion and Heat Loss Comprehensive Physiology

HT, Vignaud P, Brunet M. Cosmogenic nuclide dating of Sahelanthro- 117. Pontzer H, Raichlen DA, Rodman PS. Bipedal and quadrupedal loco-
pus tchadensis and Australopithecus bahrelghazali Mio-Pliocene early motion in chimpanzees. J Hum Evol 66: 64-82, 2014.
hominids from Chad. Proc Natl Acad Sci U S A 105: 3226-3231, 2008. 118. Pontzer H, Rolian C, Rightmire GP, Jashashvili T, Ponce de León MS,
83. Lee SS, Piazza SJ. Built for speed: Musculoskeletal structure and sprint- Lordkipanidze D, Zollikofer CP. Locomotor anatomy and biomechan-
ing ability. J Exp Biol 212: 3700-3707, 2009. ics of the Dmanisi hominins. J Hum Evol 58: 492-504, 2010.
84. Lewis OJ. Functional Morphology of the Evolving Hand and Foot. 119. Pontzer HD, Wrangham RW. The ontogeny of ranging in wild chim-
Oxford: Oxford Univ. Press, 1989. panzees. Int J Primatol 27: 295-309, 2006.
85. Lewontin RC. The Apportionment of human diversity. Evol Bio 6: 120. Raichlen DA, Armstrong H, Lieberman DE. Calcaneus length deter-
391-398, 1972. mines running economy: Implications for endurance running perfor-
86. Liebenberg L. The Art of Tracking: The Origin of Science. Claremont, mance in modern humans and Neandertals. J Hum Evol 60: 299-308,
South Africa: David Philip Publishers, 2001. 2011.
87. Liebenberg L. Persistence hunting by modern hunter-gatherers. Curr 121. Raichlen DA, Gordon AD, Harcourt-Smith WE, Foster AD, Haas
Anthropol 47: 1017-1026, 2006. WR. Laetoli footprints preserve earliest direct evidence of human-like
88. Lieberman DE. The Evolution of the Human Head. Cambridge MA: bipedal biomechanics. PLoS One 5: e9769, 2010.
Harvard University Press, 2011. 122. Rantala M. Human nakedness: Adaptation against ectoparasites? Int J
89. Lieberman DE. The Story of the Human Body: Evolution, Health and Parasitol 29: 1987-1989, 1999.
Disease. New York: Pantheon, 2013. 123. Rantala M. Evolution of nakedness in Homo sapiens. J Zool 273: 1-7,
90. Lieberman DE, Bramble DM. The evolution of marathon running: 2007.
Capabilities in humans. Sports Med 37: 288-290, 2007. 124. Reich D, Patterson N, Kircher M, Delfin F, Nandineni MR, Pugach I,
91. Lieberman DE, Raichlen DA, Pontzer H, Bramble DM, Cutright-Smith Ko AM, Ko YC, Jinam TA, Phipps ME, Saitou N, Wollstein A, Kayser
E The human gluteus maximus and its role in running. J Exp Biol 209: M, Pääbo S, Stoneking M. Denisova admixture and the first modern
2143-2155, 2006. human dispersals into Southeast Asia and Oceania. Am J Hum Genet
92. López Herráez D, Bauchet M, Tang K, Theunert C, Pugach I, Li J, 89: 516-528, 2011.
Nandineni MR, Gross A, Scholz M, Stoneking M. Genetic variation and 125. Richmond BG, Begun DR, Strait DS. Origin of human bipedalism: The
recent positive selection in worldwide human populations: Evidence knuckle-walking hypothesis revisited. Yrbk Phys Anthropol 44: 71-105,
from nearly 1 million SNPs. PLoS One 4: e7888, 2009. 2001.
93. Lovejoy CO. The origin of man. Science 211: 341-350, 1981. 126. Richmond BG, Jungers WL. Orrorin tugenensis femoral morphology
94. Lovejoy CO, Heiple KG, Burstein AH. The gait of Australopithecus. and the evolution of hominin bipedalism. Science 319: 1662-1665,
Am J Phys Anthropol 38: 757-779, 1973. 2008.
95. Lovejoy CO, Latimer B, Suwa G, Asfaw B, White TD. Combining 127. Roach NT, Venkadesan M, Rainbow MJ, Lieberman DE. Elastic energy
prehension and propulsion: The foot of Ardipithecus ramidus. Science storage in the shoulder and the evolution of high-speed throwing in
326: 72e1-72e8, 2009b. Homo. Nature 498: 483-486, 2013.
96. Lovejoy CO, Suwa G, Spurlock L, Asfaw B, White TD. The pelvis and 128. Robertshaw D. Sweat and heat exchange in man and other mammals.
femur of Ardipithecus ramidus: The emergence of upright walking. J Hum Evol 14: 63-73, 1985.
Science 326: 71e1-71e6, 2009a. 129. Robertshaw D. Mechanisms for the control of respiratory evaporative
97. MacLatchy L, Gebo D, Kityo R, Pilbeam D. Postcranial functional mor- heat loss in panting animals. J Appl Physiol 101: 664-668, 2006.
phology of Morotopithecus bishopi, with implications for the evolution 130. Robertshaw D, Taylor CR. A comparison of sweat gland activity in
of modern ape locomotion. J Hum Evol 39: 159-183, 2000. eight species of East African bovids. J Phsyiol 203: 135-143, 1969.
98. Mahoney SA. Cost of locomotion and heat balance during rest and 131. Rodman PS, McHenry HM. Bioenergetics and the origin of hominid
running from 0 to 55◦ in a patas monkey. J Appl Physiol 49: 789-800, bipedalism. Am J Phys Anthropol 52: 103-106, 1980.
1980. 132. Rolian C, Lieberman DE, Hamill J, Scott JW, Werbel W. Walking,
99. Marks JG, Miller J. Lookingbill and Marks’ Principles of Dermatology running and the evolution of short toes in humans. J Exp Biol 212:
(4th ed). Philadephia: Saunders, 2006. 713-721, 2009.
100. Marlowe FW. The Hadza: Hunter-Gatherers of Tanzania. Berkeley: 133. Rubenson J, Heliams DB, Maloney SK, Withers PC, Lloyd DG,
Univ. California Press, 2010. Fournier PA. Reappraisal of the comparative cost of human locomo-
101. Martin RD. Relative brain size and basal metabolic rate in terrestrial tion using gait-specific allometric analyses. J Exp Biol 210: 3513-3524,
vertebrates. Nature 293: 57-60, 1981. 2007.
102. McGrew WC. Chimpanzee Material Culture: Implications for Human 134. Ruff CB. Climatic Adaptation and Hominid Evolution: The Thermoreg-
Evolution. Cambridge: Cambridge University Press. 1992. ulatory Imperative. Evol Anthropol 2: 53-60, 1993.
103. Meiri S, Dayan,T. On the validity of Bergmann’s rule. J Biogeogr 30: 135. Ruff CB, Trinkaus E, Walker A, Larsen CS. Postcranial robusticity in
331-351, 2003. Homo, I: Temporal trends and mechanical interpretation. Am J Phys
104. Montagna W. The skin of nonhuman primates. Am Zool 12: 109-124, Anthropol 91: 21-53, 1983.
1972. 136. Ruxton GD, Wilkinson DM. 2011. Avoidance of overheating and selec-
105. Montagna W, Yun JS. The skin of primates XV. The skin of the chim- tion for both hair loss and bipedality in hominins. Proc Natl Acad Sci
panzee (Pan satyrus). Am J Phys Anthropol 21: 189-203, 1963. U S A 108: 20965-20969, 2011.
106. Morgan E. The Aquatic Ape: A Theory of Human Evolution. New York: 137. Sabeti PC, Schaffner SF, Fry B, Lohmueller J, Varilly P, Shamovsky
Stein and Day, 1982. O, Palma A, Mikkelsen TS, Altshuler D, Lander ES. Positive natural
107. Niinimaa V. Effect of nasal or oral breathing route on upper airway selection in the human lineage. Science 312: 1614-1620, 2006.
resistance. Acta Oto-Laryngologica 95: 161-166, 1983. 138. Saketkhoo K, Kaplan I, Sackner MA. Effect of exercise on nasal mucous
108. O’Connell JF, Hawkes K, Blurton Jones NG. Hadza scavenging: Impli- velocity and nasal airflow resistance in normal subjects. J Appl Physiol
cations for Plio-Pleistocene hominid subsistence. Curr Anthropol 29: 46: 369-371, 1979.
356-363, 1988. 139. Sanders WJ. Comparative morphometric study of the australopithecine
109. Olson LG, Strohl KP. The response of the nasal airway to exercise. Am vertebral series Stw-H8/H41. J Hum Evol 34: 249-302, 1998.
Rev Respir Dis 135: 356-359, 1987. 140. Sankararaman S, Mallick S, Dannemann M, Prüfer K, Kelso J, Pääbo S,
110. Pagel M, Bodmer W. A naked ape would have fewer parasites. Proc Patterson N, Reich D. The genomic landscape of Neanderthal ancestry
Roy Soc B: Biol Sci 270: S117-S119, 2003. in present-day humans. Nature. doi: 10.1038/nature12961, 2014.
111. Passey BH, Levin NE, Cerling TE, Brown FH, Eiler JM. High- 141. Sato F, Owen M, Matthes R, Sato K, Gisolfi CV. Functional and mor-
temperature environments of human evolution in East Africa based phological changes in the eccrine sweat gland with heat acclimation.
on bond ordering in paleosol carbonates. Proc Natl Acad Sci U S A J Appl Physiol 69: 232-236, 1990.
107: 11245-11249, 2010. 142. Sato K, Leidal R, Sato F. Morphology and development of an apoec-
112. Pickering TR, Bunn HT. The endurance running hypothesis and hunt- crine sweat gland in human axillae. Am J Physiol 252: R166-R180,
ing and scavenging in savanna-woodlands. J Hum Evol 53: 434-438, 1987.
2007. 143. Scherer PW, Hahn II, Mozell MN. The biophysics of nasal airflow.
113. Pickford M, Senut B. Millennium ancestor, a 6-million-year-old bipedal Otolaryngologic Clin N Am 22: 265-278, 1989.
hominid from Kenya. Comptes rendus de l’Académie des Sciences de 144. Schmidt-Nielsen K. The physiology of the camel. Scient Am 201: 140-
Paris, série 2a, 332: 134-144, 2001. 151, 1959.
114. Pilbeam D, Young N. Hominoid Evolution: Synthesizing Disparate 145. Schmidt-Nielsen K. Desert Animals: Physiological Problems of Heat
Data. Comptes Rendus Palevol 3: 303-319, 2004. and Water. Oxford: Clarendon Press, 1964.
115. Pontzer H. Predicting the energy cost of terrestrial locomotion: A test of 146. Schmidt-Nielsen K, Bretz WL, Taylor CR. Panting in dogs. Science
the LiMb model in humans and quadrupeds. J Exp Biol 210: 484-494, 169: 1102-1104, 1970.
2007. 147. Schmitt D. Insights into the evolution of human bipedalism from exper-
116. Pontzer H. Relating ranging ecology, limb length, and locomotor econ- imental studies of humans and other primates. J Exp Biol 206: 1437-
omy in terrestrial animals. J Theor Biol 296: 6-12, 2012. 1448, 2003.

116 Volume 5, January 2015

Comprehensive Physiology Evolution of Human Locomotion and Heat Loss

148. Schultz AA. The density of hair in primates. Hum Biol 3: 303-317, 169. Vogel S. Comparative Biomechanics: Life’s Physical Word (2nd ed).
1931. Princeton: Princeton University Press, 2013.
149. Schwartz GG, Rosenblum LA. Allometry of primate hair density and 170. Vogt A, McElwee K, Blume-Peytavi U. Biology of the hair follicle. In:
the evolution of human hairlessness. Am J Phys Anthropol 55: 9-12, Blume-Peytavi U, Tosti A, Whiting D, Trüeb R, editors. Hair Growth
1981. and Disorders. Berlin: Springer, 2008, pp. 1-22.
150. Semaw S, Simpson SW, Quade J, Renne PR, Butler RF, McIntosh WC, 171. Walker A, Shipman P. The Ape in the Tree: An Intellectual and Natu-
Levin N, Dominguez-Rodrigo M, Rogers MJ. Early Pliocene hominids ral History of Proconsul. Cambridge, MA: Harvard University Press,
from Gona, Ethiopia. Nature 433: 301-305, 2005. 2005.
151. Shea BT. Ontogenetic allometry and scaling: A discussion based on the 172. Ward CV. Interpreting the posture and locomotion of Australopithe-
growth and form of the skull in African apes. In: Jungers WL, editor. cus afarensis: Where do we stand? Yrbk Phys Anthropol 35: 185-215,
Size and Scaling in Primate Morphology. New York: Plenum Press, 2002.
1985, pp. 175-206. 173. Washburn SL. Tools and human evolution. Sci Am 203: 63-75, 1960.
152. Shea JJ. The origins of lithic projectile point technology: Evidence 174. Weiner JS. Variation in Sweating. In: Weiner JS, editor. Physiological
from Africa, the Levant, and Europe. J Arch Sci 33: 823-846, 2006. Variation and Its Genetic Basis. London: Taylor and Francis, 1977,
153. Shostak M. Nisa, the Life and Words of a !Kung Woman. Cambridge, pp. 125-137.
MA: Harvard University Press, 1981. 175. Wheatley JR, Amis TC, Engel LA. Oronasal partitioning of ventilation
154. Sockol MD, Raichlen D, Pontzer HD. Chimpanzee locomotor energet- during exercise in humans. J Appl Physiol 71: 546-551, 1991.
ics and the origin of human bipedalism. Proc Natl Acad Sci U S A 104: 176. Wheeler PE. The evolution of bipedality and loss of functional body
12265-12269, 2007. hair in hominids. J Hum Evol 13: 91-98, 1984.
155. Spoor F, Wood BA, Zonneveld F. Implications of early hominid 177. Wheeler PE. The loss of functional body hair in man: The influence of
labyrinthine morphology for the evolution of human bipedal locomo- thermal environment, body form and bipedality. J Hum Evol 14: 23-28,
tion. Nature 369: 645-648, 1994. 1985.
156. Steudel-Numbers KL, Wall-Scheffler CM. Optimal running speed and 178. Wheeler PE. The thermoregulatory advantages of hominid bipedalism
the evolution of hominin hunting strategies. J Hum Evol 56: 355-360, in open equatorial environments: The contribution of increased con-
2009. vective heat loss and cutaneous evaporative cooling. J Hum Evol 21:
157. Stitt JT, Hardy JD. Thermoregulation in the squirrel monkey (Saimiri 107-115, 1991.
sciureus). J Appl Physiol 31: 48-54, 1971. 179. Wheeler PE. The influence of the loss of functional body hair on the
158. Susman RL, Brain TM. New first metatarsal (SKX 5017) from water budgets of early hominids. J Hum Evol 23: 379- 388, 1992.
Swartkrans and the gait of Paranthropus robustus. Am J Phys Anthropol 180. White TD, Ambrose SH, Suwa G, Su DF, DeGusta D, Bernor RL,
77: 7-15, 1988. Boisserie JR, Brunet M, Delson E, Frost S, Garcia N, Giaourtsakis
159. Susman RL, de Ruiter DJ. New hominin first metatarsal (SK 1813) IX, Haile-Selassie Y, Howell FC, Lehmann T, Likius A, Pehlevan
from Swartkrans. J Hum Evol 47: 171-181, 2004. C, Saegusa H, Semprebon G, Teaford M, Vrba E. Macrovertebrate
160. Susman RL, Stern JT Jr, Jungers WL. Arboreality and bipedality in the paleontology and the Pliocene habitat of Ardipithecus ramidus. Science
Hadar hominids. Folia Primatol. 43: 113-156, 1984. 326: 87-93, 2009.
161. Taylor CR, Heglund NC, Maloiy GM. Energetics and mechanics of 181. White TD, Asfaw B, Beyene Y, Haile-Selassie Y, Lovejoy CO, Suwa
terrestrial locomotion. I. Metabolic energy consumption as a function G, WoldeGabriel G. Ardipithecus ramidus and the paleobiology of early
of speed and body size in birds and mammals. J Exp Biol 97: 1-21, hominids. Science 326: 75-86, 2009.
1982. 182. White TD, Suwa G, Asfaw B. Australopithecus ramidus, a new species
162. Taylor CR, Rowntree VJ. Temperature regulation and heat balance in of early hominid from Aramis, Ethiopia. Nature 371: 306-312, 1994.
running cheetahs: A strategy for sprinters? Am J Physiol 224: 848-851, 183. Whittow GC. Comparative Physiology of Thermoregulation. II Mam-
1973. mals. New York: Academic Press, 1971.
163. Thorpe SKS, Holder RL, Crompton RH. Origin of human bipedalism 184. Wilkins J, Schoville BJ, Brown KS, Chazan M. Evidence for Early
as an adaptation for locomotion on flexible branches. Science 316: Hafted Hunting Technology. Science 338: 942-946, 2012.
1328-1331, 2007. 185. Wolpoff MH. Climatic influence on the nasal aperture. Am J Phys
164. Tilkens MJ, Wall-Scheffler C, Weaver TD, Steudel-Numbers K. The Anthropol 29: 405-423, 1968.
effects of body proportions on thermoregulation: An experimental 186. Wood B, Harrison T. The evolutionary context of the first hominins.
assessment of Allen’s rule. J Hum Evol 53: 286-291, 2007. Nature 470: 347-352, 2012.
165. Trinkaus E. Neanderthal limb proportions and cold adaptations. In: 187. Wrangham R, Cheney D, Seyfarth R, Sarmiento E. Shallow-water habi-
Stringer CB, editor. Aspects of Human Evolution. London: Taylor Fran- tats as sources of fallback foods for hominins. Am J Phys Anthropol
cis, 1981, pp. 187-224. 140: 630-642, 2009.
166. Turner A. The evolution of the guild of larger terrestrial carnivores 188. Yang N, MacArthur DG, Gulbin JP, Hahn AG, Beggs AH, Easteal
during the Plio-Pleistocene in Africa. Geobios 23: 349-368, 1990. S, North K. ACTN3 genotype is associated with human elite athletic
167. Van Valkenburgh B. The dog-eat-dog world of carnivores: A review performance. Am J Hum Genet 73: 627-631, 2003.
of past and present carnivore community dynamics. In: Stanford CB, 189. Yokley TR. Ecogeographic variation in human nasal passages. Am J
Bunn HT, editors. Meat-Eating and Human Evolution. Oxford: Oxford Phys Anthropol 138: 11-22, 2009.
University Press, 2001, pp. 101-121. 190. Yousef MK, Dill DB. Energy expenditure in desert walks: Man and
168. Vignaud P, Duringer P, Mackaye HT, Likius A, Blondel C, Bois- burro Equus asinus. J Appl Physiol 27: 681-683, 1969.
serie JR, De Bonis L, Eisenmann V, Etienne ME, Geraads D, Guy F, 191. Zenker W, Kubik S. Brain cooling in humans – anatomical considera-
Lehmann T, Lihoreau F, Lopez-Martinez N, Mourer-Chauviré C, Otero tions. Anat Embryol 193: 1-13, 1996.
O, Rage JC, Schuster M, Viriot L, Zazzo A, Brunet M. Geology and 192. Zollikofer CP, Ponce de León MS, Lieberman DE, Guy F, Pilbeam D,
palaeontology of the Upper Miocene Toros-Menalla hominid locality, Likius A, Mackaye HT, Vignaud P, Brunet M. Virtual cranial recon-
Chad. Nature 418: 152-155, 2002. struction of Sahelanthropus tchadensis. Nature 434: 755-759, 2005.

Volume 5, January 2015 117