Bones, Stones and Molecules
“Out of Africa” and Human Origins
Reconstruction of Paranthropus
(Adapted from Matternes [Isaac & McCown, 1976])
Bones, Stones and Molecules
“Out of Africa” and Human Origins
David W. Cameron
Department of Anatomy and Histology
The University of Sydney
and
Colin P. Groves
School of Archaeology and Anthropology (Faculties)
Australian National University
Acquisition Editor: David Cella
Project Manager: Troy Lilly
Editorial Coordinator: Kelly Sonnack
Marketing Manager: Linda Beattie
Marketing Manager: Clare Fleming
Cover Design: Cate Rickard Barr
Interior Design: Julio Esperas
Composition: Newgen Imaging Systems
Interior Printer: The Maple-Vail Book Manufacturing Group
Cover Printer: Phoenix Color Corporation
Elsevier Academic Press
200 Wheeler Road, Burlington, MA 01803, USA
525 B Street, Suite 1900, San Diego, California 92101-4495, USA
84 Theobald’s Road, London WC1X 8RR, UK
This book is printed on acid-free paper.
Copyright © 2004, Elsevier Inc.
All rights reserved.

v
This Page Intentionally Left Blank
vii
CONTENTS
Acknowledgments ix
Preface xi
1. Introduction 1
Interlude 1: Creationism and Other Brainstorms 29
2. Evolution of the Miocene Great Apes 35
3. The Later Miocene and Early Pliocene Hominids 59
Interlude 2: The Importance of Being an Ape 79
4. Our Kind of Hominins 83
5. A Systematic Scheme for the Pliocene and Early
Pleistocene Hominids 105
Interlude 3: Of Men’s Beards and Peacock’s Tails 151
6. The First African Exodus: The Emergence of Early Homo
in Europe and Asia 157
7. Human Evolution in the Middle Pleistocene 181
Interlude 4: The Geography of Humanity 201
8. “The Grisly Folk”: The Emergence of the Neanderthals 207
9. The Second African Exodus: The Emergence of
Modern Humans 233
10. The Emergence of Modern Humans in Asia and Australia 251
Interlude 5: Milford Wolpoff in the Garden of Eden 275
11. Epilogue 279
Appendix: Detailed Description of Characters (DWC) 287
References 345
Index 395
viii Contents
ix

logy has enabled us to identify the likely period when proto-chimpanzees
and proto-humans last shared a common ancestor (around 6 million years
ago), and the most recent contribution from this field to the study of human
evolution has been the extraction and analysis of Neanderthal mtDNA. All
of this evidence supports the idea that human evolution over the last few mil-
lion years is a complex story, defined by considerable species diversity.
It is becoming increasingly clear to both authors that the “Out of Africa”
model for recent human origins is supported by the available fossil, archae-
ological and molecular evidence, though, as we will also argue, there was
more than one “Out of Africa,” and in some cases there were dispersals into
Africa during the early Pleistocene by some human species. That is not
to say that we both agree on the details of human evolution over the last
5 million years or so. As the reader will see, we agree to disagree, which is
shown most markedly in our differing taxonomies of the hominids, both of
which suggest distinct relationships within the more recent members of our
own family, the Hominidae.
This Page Intentionally Left Blank
CHAPTER 1
Introduction
I
s the evolution of modern humans an African genesis followed by prehis-
toric worldwide genocide of earlier pre-sapiens, or is it a slow progression
from pre-sapiens to modern humans? Theories concerned with modern
human evolution have been polarized by these extreme views. These two
basic positions have been referred to, respectively, as the “Out of Africa” and
the “Multiregional” hypotheses. Does the paleontological, archaeological,
and molecular evidence support the mass extinction of earlier humans, the
last of all being the Neanderthals, or did these diverse pre-sapiens interbreed
with the more “successful,” modern H. sapiens, thus being swamped gene-
tically and physically? Indeed, are Neanderthals just an extreme version of

way. This tempo and mode of evolution best fits the “Out of Africa” hypothesis for
modern human origins.
From Eldredge and Gould (1972), p. 113.
Stanley, 1978, 1979; Tattersall, 1986; Eldredge, 1989). Under this model,
the many gaps in the fossil record are not merely annoying hiatuses, they
are actually data: they are informing us about the tempo of evolution, that
in many cases these gaps are the result of rapid speciation (rapid in geo-
logical time, that is, about 100,000 years!). Given this rapid turnover, then,
the transitional forms were unlikely to be fossilized; or if they were fos-
silized, they are unlikely ever to be discovered, given their small popula-
tion size and occupation of a restricted geographical region. While there
certainly are many, many gaps in the hominid fossil record, it is perhaps
the Miocene hominid record from 23 to 6 million years ago that has been
most clearly shown to be characterized by a tempo and mode of evolution
that best fits a punctuationist model (Cameron, in press a). This will surely
prove to be the case for the hominids and hominins of the Old World, for
they are marked by a sudden explosion of contemporary species, many of
which appear to have left no direct descendants. This is further empha-
sized because the fossil record will always underestimate the number of
species, and we will never have fossils representing all of the species that
have ever existed.
The theory of punctuated equilibrium argues that the mode of speciation
is the result of reproductive isolation at the periphery of a species’ range,
the emphasis being on cladogenesis as opposed to anagenesis (see
Eldredge & Gould, 1972; Gould & Eldredge, 1977; Stanley, 1978, 1979,
1996; Eldredge, 1989; Gould, 2002). Cladogenesis is the splitting of a sin-
gle species into two reproductively isolated or genetically distinct lineages
so that species remain relatively unchanged for long periods of time, occa-
sionally interrupted by rapid or short bursts of evolutionary change result-
ing in speciation. The isolation of a marginalized population results in a

Time
Species 1
Species 2
Anatomical chan
g
e
Species 3
Arbitrary
boundaries
Figure 1.2 ᭤ Evolution via anagensis, in which there is limited or no cladogenesis. One
species is considered to have evolved into another through gradual evolution, resulting in
“chronospecies.”
the three living species, bulldozes whole stands of trees and turns bush and
forest into savannah or even desert, affecting the livelihood and abundance
of the other mammals that live in the same habitat; so every year hundreds
of elephants are shot in southern African game reserves and national parks,
based on the premise that uncontrolled populations of elephants will
destroy the whole ecosystem. Yet what sounds like a clear-cut Red Queen
scenario has been challenged. On a large geographic scale, the effect may
be cyclical (a stable limit cycle): The elephants eat themselves out of
house and home, their populations plummet and the survivors emigrate,
the vegetation recovers, the elephants increase again, the circle is closed.
If there is no sustained Red Queen effect, there is no anagenesis, at least
in its traditional (gradualistic) form; or else it must depend solely on grad-
ual, continuous nonbiological changes such as long-term, unidirectional cli-
mate or sea-level change. But these seem to have been episodic, not
sustained uninterruptedly. At most there is the possibility, even likelihood,
that a local environment is somewhat altered after each cycle so that the
cumulative effect of a long chain of cycles is really noticeable. But this
begins to stretch the concept of anagenesis as gradualism. The Red Queen,

The importance of speciation has been promoted many times in the fossil
record. Groves (1989a) argued that, if it is true that evolutionary change is
concentrated at the point of speciation, we can predict that, of two sister
Chapter 1 Introduction 5
species, the one that is more changed (highly autapomorphic) from the
common ancestor will have undergone more cladogenesis (its lineage has
gone through more speciation events) than the one that is less changed.
Unfortunately, the record of human evolution offers only a partial test of
this. The human species is much more different than is the chimpanzee from
our common ancestor, and the human fossil record is certainly enormously
speciose, but the chimpanzee fossil record is empty. All we can do is predict
that, when paleontologists start prospecting in the right place to find proto-
chimpanzees, they will not be very speciose. Chimpanzee evolution will
prove to be, let us say, as nearly unlinear in reality as human evolution was
held to be up until the 1970s, when the single-species model finally became
untenable. But, as we will see presently, the single-species hypothesis has
reared its head again, though not through an analysis of fossil material but,
rather, by an abstract discussion of the molecular evidence.
If any statement regarding our own origins is correct, it is that humans origi-
nally evolved in Africa. We can all trace our prehistoric roots back to the
African continent around 6 million years ago. It was at this time that popula-
tions of proto-chimpanzees and proto-humans split from a common ancestor
and each started its own evolutionary journey. The recently described fossils
allocated to Sahelanthropus from Chad, dating to between 6–7 million
years ago, and Orrorin from Kenya, dating to around 6.1–5.8 million years
ago, are close to the point of separation (Brunet et al., 2002; Senut et al.,
2001; Pickford et al., 2002), as is the earlier hominid discovery from
Lothagam, dated to between 5.0–5.2 million years ago (see M.G. Leakey &
Walker, 2003).
Following on from these late Miocene genera comes Ardipithecus,

Ardipithecus
“A.” anamensis
Praeanthropus
“A.” garhi
“A.” bahrelghazali
K. platyops
K. rudolfensis
A. africanus
P. walkeri
P. robustus
P. boisei
H. habilis
H. ergaster
?
?
?
?
?
Later
Homo
H. erectus
Figure 1.3 ᭤ Proposed evolutionary scheme for the Plio-/Pleistocene hominids and
hominins. LCA ϭ last common ancestor (Miocene, e.g., Sahelanthropus and/or Orrorion?).
and Grine (1998, 2001) have reallocated “A.” afarensis (which contains
the famous “Lucy” skeleton) to the genus Praeanthropus. This genus was
first described in the 1950s (see also Harrison, 1993). Thus they and
Cameron would argue that only one species, A. africanus (the type
species), exists within the genus Australopithecus.
The evolution of the later, more derived hominins, Paranthropus, the
“rudolfensis group” (represented by the famous 1470 skull), and early Homo,

and Europe). It is true that a general likeness of skull shape is maintained
over vast eons of time — hundreds of thousands of years — within each of
8 Chapter 1 Introduction
these regions, though this is to be expected given the similar rate of
encephalization. Only in Europe, however, was there a measurable change
within one of these species: After about 400,000 years ago, Homo heidel-
bergensis, which had entered Europe from Africa a few hundred thousand
years before, had by 120,000 years ago become Homo neanderthalensis,
the famous Neanderthal people (Stringer, 1989, 1994; Stringer & McKie,
1996), whereas the deme that remained in Africa had by 160,000 years ago
emerged into near modern H. sapiens, as defined by the recent significant
discoveries of the Herto specimens from Ethiopia (T.D. White et al., 2003;
Clark et al., 2003; see also Stringer, 2003).
It has also been suggested by some, however, that the lineage leading to
H. neanderthalensis had already been established as early as 780,000 years
ago, as represented by the hominins from Atapuerca (Gran Dolina), Spain,
sometimes referred to as H. antecessor (Bermúdez et al., 1997). They sug-
gest that H. heidelbergensis was already a part of the Neanderthal lineage,
and as such the African hominins usually allocated to the same species
must be a different species because they are not part of the Neanderthal line-
age. Thus a separate and parallel line in Africa (H. rhodesiensis?) may have
led to the evolution of H. sapiens via African populations, as represented
by the Herto, Elandsfontein, and Kabwe specimens (see Stringer, 1998,
2003; Clark et al., 2003; T.D. White et al., 2003), so having nothing to do
with the emergence of the Neanderthals.
Other authorities (multiregionalists) disagree with these interpretations.
These are not different species, they say, but races of early Homo sapiens;
just as modern Homo sapiens has somewhat different geographic varieties,
which we sometimes refer to as “races,” so did ancient Homo sapiens
(Wolpoff, 1989, 1999; Wolpoff & Caspari, 1997; Wolpoff et al., 1984,

recently in that group. For example, quadrupedal locomotion is a primitive
character of the primates (we know this because almost all other mammals
are quadrupedal), which tells us little about the evolutionary relationships
within this large group. Habitual bipedal locomotion, however, is a derived
feature linking humans and the proto-australopithecines and their immedi-
ate ancestors, to the exclusion of most other primates (see next chapter). In
summary, fossils enable us to identify evolutionary relationships among
species and likely physical adaptive trends through time and space.
Stone tools, and an interpretation of their immediate context, are an
important source of information when trying to reconstruct past human
behavior and cultural evolution. While early humans undoubtedly used
other materials (such as wood and animal skins), these are not usually pre-
served in the archaeological record. The development of ever more sophis-
ticated stone “tool kits” by early humans enabled them to adapt more
readily to and extract new food resources from their ever-changing envi-
ronments and habitats. It also allowed them to defend themselves from
much larger and more ferocious animals, and it enabled them to hunt and
thus to develop an increased sense of community. In developing this tech-
nology, early humans started their long journey on the road to reshaping
their environment, rather than simply being shaped by it. Through time,
10 Chapter 1 Introduction
a number of different tool traditions were developed. Archaeologists have
been able to associate some of these tool traditions with particular human
groups (Bordes, 1950, 1961, 1969; Bordes & Sonneville-Bordes, 1970;
Foley & Lar, 1997), while other tool kits are clearly designed for specific
functions and not related to differing “cultural” traditions (Binford &
Binford, 1966; Binford, 1983). Interpreting how these tools were used has
enabled archaeologists to help reconstruct aspects of past human behavior.
The recent application of molecular biology to human evolutionary
studies has greatly influenced current interpretations of human origins.

present in the nucleus of every cell; the DNA they contain is called nuclear
DNA (nDNA). It is important to realize that genes actually make up only
a very small part of nDNA; the rest does not code for anything and is
(rightly or wrongly) often referred to as “junk DNA.” There are pseudo-
genes (segments of DNA that used to be genes in the distant, evolutionary
past but that have been “switched off” over time); introns (meaningless
segments inserted in the middle of genes); and repetitive DNA (varying
from long sequences repeated thousands of times to short sequences
repeated hundreds of thousands of times, called microsatellites). Between
them, these “junk” bits make up 90% or more of the complement of nDNA
(Pilbeam, 1996; Dover, 1999; Relethford, 2001).
Outside the cell nucleus, in the body of the cell itself (the cytoplasm), are
thousands of tiny bodies called mitochondria, which provide the energy on
which the body’s metabolism runs. The mitochondria have their own DNA,
mitochondrial DNA (mtDNA). Because mtDNA mutates without any of
the “correction” mechanisms operating in nDNA, it changes much faster,
and so its variation is an important source of information with regard to the
timing of a speciation event among species, as well as identifying likely
evolutionary relationships within and between groups. Importantly, mtDNA
is inherited, to all intents and purposes, solely from our mothers, for the
contribution from the sperm is minute compared to that from the ovum; so
mtDNA traces the path of genetic development for our female ancestors in
the evolutionary past. If we want to trace where male ancestors went, we
have to look at the nDNA of the chromosome that is unique to males: the
Y chromosome (Sykes, 2001; Relethford, 2001).
For mtDNA, as for much of DNA, a constant rate of mutation has been
assumed. Whether this assumption is always justified is another matter.
Certainly mtDNA includes some genes that provide energy for the cell. But
because of the way in which the genetic code operates, most mutations do
not seem to affect the functioning of the organism, so the assumption of