Under the current taxonomy (based on genetic rather than behavioral criteria), the term "hominid" refers to members of the biological human family Hominidae: living humans, all human ancestors, the many extinct members of Australopithecus, and our closest primate relatives, the chimpanzee and gorilla. According to The Tree of Life by Guillaume Lecointre and Hervé Le Guyader (Harvard University Press: 2006), the similarly named and easily confused categories of humans and near human apes, in order of increasing inclusiveness, are:

• Hominini - modern humans and all previous human, australopithecine, paranthropine and ardipithecine ancestors

• Homininae - all of the above, plus chimpanzees (Panini), our closest living biological kin (a genetic kinship so close that some scientists have suggested their genus name should be changed from Pan to Homo).

• Hominidae - all of the above, plus gorillas (Gorillinae)

• Hominoidae - all of the above, plus orangutans (Pongidae)

• Hominoidea - all of the above, plus gibbons (Hylobatoidae).

The chart (at right) shows the evolutionary chronology inputed to these biological branches. Ardipithecus, the common primate ancestor to paranthropines, australopithecines and humans, went extinct about 4 million years ago.

Human evolution is a puzzle made up of thousands of fossil pieces. The Chart of Human Evolution (below) shows the major pieces of that puzzle arranged in a likely solution.

• Each colored bar represents the time interval spanned by recovered fossils associated with that species. Dotted lines indicate the conjectural evolutionary lines of descent. (Different paleoanthropologists will connect these in different ways, while preserving the chronological sequence.)

• Under each species name is a list of the national or geographical areas where all or most of its fossil remains have been found.

• White numbers inside the species bars indicate the approximate count of distinct individuals in each species from whom fossil remains survived as of c.1995. (Subsequent discoveries and reclassifications will have changed these numbers.) The number of individuals is considerably smaller than the number of fossil specimens, because a specimen can be a single tooth, bone or bone fragment. These numbers suggest the relative reliability of the species classifications they support.

• The human fossil record from about 2.5 to 1.0 million years ago is especially sparse — only about 50 individuals are known, some of them represented by only a single tooth or bone fragment — and the evolutionary connections from Australopithecus to Homo erectus, including the evolutionary relationships between habilis, ergaster and erectus, are in dire need of clarification.

• Time spans for modern humans, Neanderthals and archaic Homo sapiens (H. heidelbergensis) have been extended back beyond accepted fossil limits to accommodate recent genetic evidence that the divergence between the Neanderthal and human lines occurred around 500,000 years ago.

• As environmental or climate context, the major Ice Age epochs in recent human experience were [1] the Wisconsin, 11,000-35,000 years ago (the most extreme of recent coolings), and [2] the Illinoian, 130,000-190,000 years ago, with an intermediate ice era around 60,000-70,000 years ago.

• Four human species proposed in the literature — H. floresiensis, H. pekinensis, H. georgicus and H. rhodesiensis — have been omitted as conjectural or controversial. Homo rudolfensis is now assigned to the "1470 group" variant of Homo habilis, designated by the "1813 group" label. (The numbers refer to the taxonomic type specimens used to anchor the species.)

There are at least six independent factors contributing to this remarkable evolutionary emergence: genetic variability, climate change, migration, dietary flexibility, alloparenting of extended early development, and technology.

Natural selection can only produce new species out of the genetic variation already within existing species. The origins of human variability are genetic — in the human genome — and this variability is a characteristic fact of human nature. The recent discovery of Australopithecus sediba shows clearly the combinatorial churning of modular components in the early hominid lineage.

Evolution was helped along by major cycles of climate change (including changes in African ecological mosaics, sea level, land bridges and temperature), which likely propelled the human prehistory of migration and the resulting geographic isolation of different hominid groups — as documented in Hominid Fossil Sites and Patterns of Hominid Dispersal. These migrations were facilitated by an opportunistic, omnivorous dietary range adapted to local food sources.

In all hominids, males are larger than females, and adults are larger than juveniles, although these sex and age differences vary across species. These distinctive, within species variations in age and sex complicate the problem of distinguishing one species from another on the basis of fragmentary fossil evidence.

Academic debates about how to interpret the evidence are sometimes driven by career, partisan or political considerations: researchers have been known to hoard fossils they have discovered to extract the maximum career advantage or ideological leverage. In my view, the reliability of the overall picture of human descent and geographic dispersal should console us for the remaining uncertainties. Indeed, the variety discovered in the fossils, and the diversity of expert judgments as to how the variety in the fossils should be interpreted, together illustrate the physical and cultural divergences that seem characteristic of all human existence, then as now.

Fossils document the coexistence of clearly different hominid species over the last 2 million years — sometimes in adjacent or overlapping geographic regions. Homo erectus and Homo habilis coexisted in Africa, probably in different ecological niches, for almost 500,000 years. How these different species may have interacted, interbred or contested for resources is unclear.

More recently, the disappearance within about 20,000 years of Homo neanderthalensis, as Homo sapiens migrated from the Middle East into Europe, may have been only the uncontested and opportunistic replacement of one species by another, as some climate evidence suggests. Or it may have been the outcome of a protracted racial struggle, as the many historical instances of predatory wars, genocide and slavery make all too plausible. The genetic evidence of interbreeding between the species is consistent with the common practice in ancient Eurasian cultures of the sexual and labor exploitation of the defeated.

Biologists use three types of evidence to deduce evolutionary connections: genetics, morphology, and geologic dating. (Behavior, normally a key part of evolutionary studies, can only be inferred in extinct species — for example, by examining the ecology in which the species flourished and the species adaptations for eating and locomotion.)

Analyses of primate fossils and the genetic relatedness of living primates converge to the conclusion that humans and chimpanzees branched from a common ancestor about 7 million years ago. DNA recovered from several uncontaminated Neanderthalensis fossils indicated that modern humans and extinct neanderthals diverged about 400,000 years ago; but more recent studies show that they must have interbred within Europe or the Middle East since then. Genetic samples collected from indigenous populations around the world indicate that the ancestors of the world human population diverged from the indigenous African population about 200,000 years ago. These studies also provide remarkably detailed evidence for subsequent waves of human migration, including a final migration out of Africa around 90,000 years ago by the first humans similar to ourselves.

The rest of the puzzle must be deduced from morphology (physical form, as reconstructed from the bones) and geologic dating. "Absolute" fossil dating can be quite reliable for fossils buried within intact rocks, but for fossils found exposed on the surface, or buried within alluvial or eroded deposits, dates can be grossly conjectural. And morphology becomes a subtle interpretation when the available fossils are crushed and incomplete, or collated from different fragmentary fossils found in different geographic locations.

The cladogram for human evolution shown above currently lacks key pieces of evidence. For that reason I have omitted several descent connections in the diagram, although the species distinctions and time spans shown in the diagram appear to be adequately documented. Note that there is no dispute whatsoever among biologists that all these different species should be interconnected as a single branching history or "family tree" of hominid descent. The diagram simply indicates that we lack sufficient evidence, at present, to state with confidence where and when those branchings occurred.

These are some of the most interesting but currently unanswered questions about human descent:

(1) Does Homo ergaster or Homo habilis have chronological priority as the earlier hominid in a single line of descent to modern humans, or (more likely) were these separate and coexisting hominid lines by about 2 million years ago?

(2) Is Homo ergaster a distinct species, or an African variant of early Homo erectus?

(3) Are the (mostly African) examples of Homo erectus descended from Homo ergaster; or are they either (a) descended from Homo habilis or (b) descended from the same ancestor as Homo habilis?

(4) Do the (mostly Asian) fossils of Homo erectus represent a single species that lived for 2 million years or a sequence of species flourishing around 1.6 million, 1.2 million and 200,000 years ago?

(5) Is archaic Homo sapiens (Homo antecessor or Homo heidelbergensis) descended directly from Homo ergaster and coexisting with Homo erectus, or is it a branching out of Homo erectus?

Thigh bones are next most often retrieved, while remains of the feet, hands, pelvis or spine are extremely rare.

Specific behavioral conclusions require specific parts of the skeleton. For example, adaptation for a crouching or upright posture can be inferred from the connection of the spine to the skull, but bipedalism (habitual walking on two legs) requires evidence from bones involved in the thigh, knee, or foot joints. An opposable thumb requires evidence from wrist or hand bones.

Skulls are used as evidence for the evolution of The Hominid Brain. Endocasts (models of the inside of a skull) offer good evidence for the size and shape of the brain that was in the skull, and brain anatomy is sometimes (tenuously) used to infer the cognitive capabilities of the different species; cognitive abilities can also be inferred from the skills required to make fire and manufacture Hominid Tools.