11.1 Humans are apes #
11.1.1 Names, and the split #
Humans are apes. Hominoidea includes gibbons, orangutans, gorillas, chimpanzees, bonobos, and humans. โApesโ excluding humans is paraphyletic: it leaves out one descendant of the common ancestor. The closest living relatives are the two species of Pan, chimpanzee and bonobo, equally related to us. Gorillas are the next sister group, then orangutans, then gibbons. Humans are not descended from chimpanzees. The shared ancestor was neither.
Hominins, in current palaeoanthropological use, are the human clade after the split from the chimpanzee lineage. Hominids, in that same modern use, are the great apes including humans. Mid-twentieth-century texts use โhominidโ for what is here called hominin. These notes use hominin for the human-side clade.
The split is about 6โ7 million years ago. Published estimates spread a little wider, depending on mutation rate and calibration. It is not a date read off one fossil. Sahelanthropus, at about 7 million years, is proposed as an early hominin, and its locomotion is still debated. It is not a settled first human. Ardipithecus ramidus, at about 4.4 million years, is a mosaic: a pelvis reorganised towards bipedalism, and a grasping foot still suited to climbing. A fossil halfway human in every character would be the ladderโs fiction, not a prediction of phylogeny.
Human chromosome 2 corresponds to two separate chromosomes in the other great apes. It carries telomere-like sequence in the middle and a second, inactive centromere: the signature of an end-to-end fusion on the human side after the split. It is a fixed mutation, not a cause of the split.
11.1.2 A bush, not a file #
Several hominin species coexisted for long intervals. Later species are not, in general, the descendants of every earlier species that looks simpler. โApe, then australopith, then Homo erectus, then Neanderthal, then modern Europeanโ is false phylogenetically and geographically.
Australopiths, including Australopithecus afarensis, lived in Africa from about 4 to about 2 million years ago. Robust forms, often placed in Paranthropus, are a side branch that lasts later. Lucy is about 3.2 million years old, and the Laetoli trackway, about 3.7 million years old, records bipedal footprints. Brain volume in afarensis is about 400โ500 cmยณ, against a modern human mean of about 1350 cmยณ, so committed bipedalism precedes the large brain. The oldest widely discussed stone tools are about 3.3 million years old, before a securely identified Homo; fossils assigned to early Homo are known by about 2.4 million years ago. Homo emerged from within this grade. Several branches ended.
Homo erectus, in the broad sense, is in Africa by about 1.9 million years ago: more human-like body proportions, and a brain often about 600โ1100 cmยณ, larger in later fossils. It is the first hominin with an unmistakable range outside Africa. Fossils at Dmanisi, in Georgia, are about 1.8 million years old, and the species persists in eastern Asia well into the Pleistocene. It overlaps later Homo in time. It is not the single middle rung between Lucy and us.
Neanderthals lived in Europe and western Asia from roughly 400,000 years ago until about 40,000 years ago. Their brains were as large as ours or larger. They made sophisticated stone tools, used pigment, and at least sometimes buried their dead. They are a sister population of the modern-human lineage, not a failed draft. Denisovans are known from very few bones, first from Denisova Cave, and from genomes. They are a sister group of Neanderthals. A Tibetan mandible is attributed to them on protein evidence. The anatomical sample is thin. Homo floresiensis, small-bodied, persisted on Flores into the Late Pleistocene and is further evidence that the genus was a bush of contemporaries.
Homo sapiens is African in origin. Fossils assigned to early members, or to populations immediately ancestral, are known from Africa from about 300,000 years ago. The dispersal that gave rise to most living non-African populations was underway by about 60,000 years ago. Earlier exits occurred and did not simply become those later populations. Strict replacement without interbreeding is too clean. Independent evolution from local Homo erectus in Africa, Europe, and Asia is not supported. The evidence favours a recent African origin with limited archaic admixture.
11.2 Admixture and recent selection #
11.2.1 A few percent #
Populations that left Africa interbred with Neanderthals. Living non-African genomes carry Neanderthal ancestry amounting to a few percent. Some African populations carry a smaller amount from later gene flow back into Africa, not from a Neanderthal population spread across the continent. The fragments are short, broken by recombination since the contact. Some genomic regions are poor in Neanderthal ancestry, consistent with selection against particular hybrid combinations, and a few introgressed alleles were later favoured.
Denisovan ancestry is also a few percent where it is strongest, in many Oceanian populations, including Papuan and Indigenous Australian genomes, and lower in mainland East Asia. The contact was with a related archaic population in Asia, not a second copy of the Neanderthal event. Finer percentages move as methods change. These notes do not fix a figure beyond a few percent. Most ancestry of living people is the African-origin Homo sapiens lineage. The archaic contribution is real, larger than a pure replacement model allowed, and not large enough to make Neanderthals or Denisovans the main ancestors of any living population.
Ranking Neanderthals as Homo neanderthalensis or as a subspecies of Homo sapiens is a ranking decision. Limited interbreeding does not settle it, because the biological species concept does not require absolute isolation (Chapter 7). Denisovans are not given a formal species name here. They are diagnosed primarily by genome. Deeper archaic structure inside Africa is an active question and is not given a fossil name in these notes.
11.2.2 Pigmentation, altitude, and lactase #
Recent human selection is ordinary selection, recent enough that sweep-like haplotypes are still visible. Three cases are well enough supported to teach directly.
Dark skin is ancestral for Homo sapiens in the operational sense that identified light-skin alleles are the derived ones, and the species originated in equatorial Africa. Melanin reduces ultraviolet penetration. The leading selective account is folate protection where ultraviolet is strong, and sufficient ultraviolet for vitamin D where it is weak. The genetic routes to lighter skin differ. A derived variant of SLC24A5 has large effect in European populations and is not the cause of light skin in East Asia. Light skin evolved more than once. Skin colour tracks ultraviolet regime. It is a poor proxy for ancestry as a whole.
Tibetan populations carry, at high frequency, a variant of EPAS1, a subunit of the hypoxia-inducible factor. It is associated with a blunted rise in haemoglobin at altitude, avoiding some of the cost of thick blood. The haplotype matches Denisovan sequence far more closely than it matches other human sequences. The favoured reading is introgression from a Denisovan-related population, then selection at altitude: adaptive introgression, as Chapter 5 allowed. Andean and Ethiopian highlanders have adapted as well, with genetic changes that are not this haplotype. The environment is shared. The mutations are local.
Lactase persistence is the subsistence case. Most mammals downregulate the enzyme that digests lactose after weaning. Several dairying populations, in Europe, East Africa, and the Middle East, carry different regulatory mutations that keep lactase expressed in adults. The European variant lies upstream of the lactase gene. One ecological pressure, more than one mutation, each risen where it was useful: the same pattern as pigmentation. None of the three cases ranks populations. Each is local, trait-specific, and, in the Tibetan case, dependent on an allele that gene flow supplied.
11.3 Variation inside the species #
11.3.1 Clines, not types #
Human genetic variation is largely clinal. Allele frequencies change gradually where people could move, and more sharply at a barrier or after a founder event. The dispersal out of Africa was a series of founder events, so heterozygosity declines with distance from Africa, and African populations hold more of the speciesโ diversity than populations elsewhere. That fact already forbids a typology of three uniform continental kinds.
Most variation lies within local populations rather than among continental groups. In Lewontinโs classical apportionment, and in many later marker studies, the within-population share is on the order of four-fifths or more. The exact fraction depends on the loci. A smaller among-population fraction can still be correlated across loci, so a large panel can often assign a person to a broad region. Ancestry assignment is a summary of frequencies. It is not a type. A racial type, in the sense rejected here, would be a discrete kind, internally uniform, with private characters, matching the traditional continental races. Human variation is not shaped that way. Neighbours differ. Boundaries on a cline are choices. Traits under local selection, such as pigmentation or lactase, cut across the clusters suggested by neutral markers, which is why a classification by skin colour disagrees with one built from the rest of the genome.
There are no human subspecies corresponding to races. Homo sapiens is one species with population structure, as expected of a young, mobile, sexual species with incomplete barriers. Social categories called races have social histories and, in some settings, medical relevance as crude correlates of ancestry and environment. A correlate is not a natural kind. Living human populations are contemporary tips of one tree, with the same time back to the African origin of the species. Differences in technology are not a ranking of genomes, and the hominin record โ coexisting species, a late large brain, admixture rather than a single file โ does not support that ranking either.
11.4 Key terms, summary, and self-check #
Key terms. Hominin; hominid; australopith; Homo erectus; Neanderthal; Denisovan; recent African origin; archaic admixture; EPAS1; lactase persistence; cline.
Summary. Humans sit inside the ape clade. Hominins are the branch after the split from chimpanzees and bonobos, about 6โ7 million years ago. Australopiths were bipedal with ape-sized brains. Homo erectus extended the range out of Africa. Neanderthals, Denisovans, and Homo sapiens overlapped in time and were not a single file. Non-African genomes carry a few percent Neanderthal ancestry, and many Oceanian genomes a few percent Denisovan ancestry where that contribution is strongest. Light skin, Tibetan EPAS1, and lactase persistence are well-supported cases of recent local selection, the Tibetan haplotype by introgression. Most human genetic variation is within populations, geography is clinal, and continental ancestry assignment does not make races into biological types.
Self-check.
1. Why is โapesโ excluding humans paraphyletic, and which living genus is the sister group of Homo?
2. What does Australopithecus afarensis do to the claim that a large brain evolved before bipedalism?
3. What archaic admixture is expected in a Papuan genome and in a European genome, at the resolution used here? Why is a figure to one decimal place not given?
4. Why is Tibetan EPAS1 evidence of selection and of introgression together? Why do Andean highlanders not simply confirm the same mutation?
5. A panel of neutral markers assigns a person to a continent. Why does that fail to revive racial typology?
# Closing: how the chapters fit together
Chapter 1 separates a fact from its mechanisms. Populations change, and they are related by descent. Inheritance of acquired use is not why. Darwin and Wallaceโs three conditions are the start of why, and the modern synthesis put Mendelian factors under those conditions and left room for drift. Chapter 2 is the evidence: nested homology, vestiges, intermediate fossils that are not a ladder, barriers, selection watched in real time, and molecular accidents shared by descent. Chapter 3 limits what selection can be asked to explain. Fitness is reproductive success relative to rivals. Modes differ in what happens to the mean and the variance. Sexual selection tracks mating success. Adaptation is a tested hypothesis. Kin selection is rB > C, not a slogan about 0.75.
Chapters 4 to 6 are the engine. HardyโWeinberg, imported from Genetics as a result, is the idle state. Mutation supplies alleles, and almost every new allele is lost. Drift samples 2N copies and is faster at small Ne. Migration homogenises neutral loci and can swamp or feed local selection. Inbreeding exposes homozygotes without changing p. Selection sees a rare dominant and misses a rare recessive, which is why mutationโselection balance sits near โ(ฮผ/s) for a complete recessive and far lower when the heterozygote is even slightly unfit. Sweeps and frequency dependence are what linkage and a non-constant fitness do to that picture.
Chapters 7 to 10 scale the same engine. Species are lineages that have stopped exchanging genes freely. Geography is the usual reason. Polyploidy is the clean exception. DobzhanskyโMuller incompatibilities make hybrids unfit without either parent having been unfit. Trees read the hierarchy and are not ladders. Reptiles without birds, fish without tetrapods, and apes without humans are paraphyletic grades. Gene trees scatter around the species tree when speciation is rapid. Deep time puts numbers on the sequence: 4.54 billion years, cellular life more than about 3.4 billion, endosymbiosis once for mitochondria and once for the primary plastid, a Cambrian expansion of animal bodies that is real and is not the origin of animals from nothing, land colonised by plants then arthropods then vertebrates, and five mass extinctions of which only two have a cause one should actually state. Macroevolution is the fossil expression of rate, of differential extinction, of co-option, and of changes in timing and copy number of a shared developmental toolkit.
Chapter 11 fills in the names for one clade: the split, the bush of hominins, a few percent of archaic ancestry, local sweeps, and clinal variation. It uses no process the earlier chapters did not define.
Revision checklist #
1. State the three conditions of Darwin and Wallace, and say what was wrong with blending inheritance as an objection.
2. Give one homologous structure, one analogous function, one transitional grade of fossils, and one shared molecular accident. Say what each does not show.
3. Write R = hยฒS and rB > C, and define every symbol.
4. Give the HardyโWeinberg frequencies and the equilibrium frequency of a deleterious recessive under mutationโselection balance, with the assumptions of each.
5. Explain why Ne is usually below N, using the sex-ratio formula or the harmonic mean.
6. Explain why a rare recessive is nearly invisible to selection, and why a new beneficial additive allele is still usually lost.
7. Contrast allopatric speciation with the conditions for sympatric speciation, and place polyploidy on the correct side.
8. Define synapomorphy, say why birds make traditional reptiles paraphyletic, and give one reason a gene tree disagrees with the species tree.
9. List the five mass extinctions and attach a cause only where these notes allow one. Order plants, arthropods, and tetrapods on land.
10. State the humanโchimpanzee split, one australopith fact that breaks a brain-first ladder, the resolution used for archaic admixture, and why within-population variation rejects racial types.
Selected sources #
These notes are an original teaching summary, not a transcript of any textbook.
- Darwin C. On the Origin of Species. John Murray, 1859.
- Futuyma DJ, Kirkpatrick M. Evolution. Sinauer / Oxford University Press, 4th and later editions.
- Herron JC, Freeman S. Evolutionary Analysis. Pearson.
- Mayr E. Systematics and the Origin of Species. Columbia University Press, 1942.
- Mayr E. The Growth of Biological Thought: Diversity, Evolution, and Inheritance. Harvard University Press, 1982.
- Kimura M. Evolutionary rate at the molecular level. Nature 217:624โ626 (1968).
- Gould SJ, Lewontin RC. The spandrels of San Marco and the Panglossian paradigm: a critique of the adaptationist programme. Proceedings of the Royal Society of London B 205:581โ598 (1979).
# Part C. Visual Figures & Scientific Diagrams Gallery
Comprehensive high-resolution visual reference repository for Evolutionary Biology from the B10logy Visual Atlas.
Allopatric Speciation Caused By Topography
# Part C. Visual Figures & Scientific Diagrams Gallery
Essential high-resolution reference diagrams for Evolutionary Biology from the B10logy Visual Atlas.
