Evolution

Before life could begin, the earth itself had to form. The generally accepted account starts with the Big Bang: the universe began with an enormous explosion, then expanded and cooled, and hydrogen and helium were the first elements formed. Galaxies, stars and planets condensed later, and our earth took shape roughly 4.5 billion years ago. The young earth had no atmosphere of the sort you are breathing now. Gases escaping from the molten interior gave it water vapour, methane, carbon dioxide, ammonia and hydrogen, and there was no free oxygen at all. That is what people mean when they call the early atmosphere a reducing atmosphere. Ultraviolet radiation reached the surface freely because there was no ozone shield, lightning was frequent, and volcanic heat was everywhere. Water vapour cooled and condensed to form the first oceans.

Several explanations for the appearance of life were offered before the modern one, and they are worth separating carefully. The theory of special creation holds that every kind of organism was made as it is by a supernatural power, that the variety of living things has never changed and never will, and that the earth is only a few thousand years old. Spontaneous generation, also called abiogenesis in its old sense, claimed that living organisms arose ready-made out of non-living matter such as decaying straw, mud or rotting meat. Louis Pasteur ended that debate by careful experiment. He boiled and sterilised sugar-yeast solution in flasks; where the flask was kept sealed against airborne contamination nothing grew, but where a similar flask was left open to air, living organisms appeared. His conclusion was that life comes only from pre-existing life. A third idea, panspermia, says that spores or units of life travelled to the earth from elsewhere in space. Notice that panspermia does not really answer the question; it only moves the problem to another planet.

The modern view is chemical evolution, proposed independently by Oparin in Russia and Haldane in Britain. Their hypothesis is that the first form of life arose slowly from non-living organic molecules such as RNA and proteins, and that the formation of life was preceded by chemical evolution, that is, the building of complex organic molecules from simpler inorganic ones. The energy for this came from ultraviolet light, lightning discharges and heat, and the reactions took place in the conditions then existing on earth, which cannot be repeated on the earth today because free oxygen would destroy such molecules as fast as they formed.

In 1953 Stanley Miller, an American scientist working with Harold Urey, tested this in the laboratory. He built a closed glass apparatus and filled it with the gases of the presumed early atmosphere: methane (CH₄), ammonia (NH₃), hydrogen (H₂) and water vapour. There was no oxygen in the flask. He kept the mixture at a high temperature, around 800 degrees Celsius, and passed electric discharge through it from a pair of electrodes to imitate lightning. The vapour was condensed back to liquid and circulated. After about a week he analysed the liquid and found amino acids had formed. Similar experiments by other workers, using slightly different mixtures, produced sugars, nitrogen bases, pigments and fats. Later, analysis of the organic material inside meteorites showed the same kinds of compounds, which supports the idea that this sort of chemistry is not unique to the earth.

Be very clear about what this experiment proves. Miller showed that the building blocks of living matter can form on their own under early-earth conditions from simple gases and an energy source. He did not make a cell and he did not make life. The step from a mixture of organic molecules to a membrane-bound, self-replicating cell is still not fully explained. On the present understanding, the first non-cellular forms of life could have appeared roughly 3 billion years ago as giant molecules such as RNA, proteins and polysaccharides that were able to reproduce their own molecules, while the first true cellular forms of life came much later, somewhere around 2 billion years ago. Those first cells lived in water, and the oxygen we now depend on was itself a later product of photosynthetic life, not a starting condition.

Palaeontological evidence. Fossils are remains or impressions of past organisms found in sedimentary rocks. Sediment settles in layers, so the deeper a layer, the older it is, and the age of a fossil can be worked out from the layer it sits in. Three things stand out. First, different-aged rock layers contain fossils of different life forms, so the kinds of organisms on earth have changed with time. Second, some forms appear only in certain layers and never again above them, which means they arose, lived for a period and then became extinct. Third, the sequence of forms in the rocks is orderly rather than random: simple forms appear in older rocks and more complex ones in younger rocks. Together these are direct historical proof that life has changed over geological time.

Comparative anatomy and morphology. The two definitions below differ in only two words each, so read them slowly. Homologous organs have the same basic structural plan and the same embryonic origin but perform different functions. The classic example is the forelimb of a whale, a bat, a cheetah and a human being. In every one of them you find the same bones arranged in the same order: humerus, then radius and ulna, then carpals, metacarpals and phalanges. Yet the whale uses it to swim, the bat to fly, the cheetah to run and the human being to grip. The only sensible explanation is that all four inherited that limb pattern from a common ancestor and then modified it for different jobs. This pattern is called divergent evolution, and homologous organs are evidence of common ancestry. Plants show it too: the thorn of Bougainvillea and the tendril of Cucurbita both arise in the axil of a leaf, so both are modified stems, though one defends the plant and the other helps it climb. The hearts and the brains of different vertebrates are homologous in the same sense.

Analogous organs are the opposite case: different structure and different embryonic origin, but the same function, because unrelated organisms met the same problem in the same environment. The wing of a butterfly and the wing of a bird both fly, but the butterfly wing is a fold of the body wall with no bones while the bird wing is a bony forelimb covered with feathers. The eye of an octopus and the eye of a mammal both form images but develop very differently. The flipper of a penguin and the flipper of a dolphin both push water. In plants, the sweet potato is a modified root while the potato is a modified stem, yet both store food underground. This pattern is convergent evolution, and analogous organs are evidence of similar habits or habitat, not of common ancestry. The one-line test: same origin and different function means homology and common ancestry; different origin and same function means analogy and convergent evolution.

Embryological evidence. Ernst Haeckel pointed out that the embryos of all vertebrates, at an early stage, show features such as rudimentary gill slits behind the head, which are present and functional only in fish and disappear in the others as development proceeds. He read this as evidence that an embryo repeats the adult stages of its ancestors. This particular claim was rejected by Karl Ernst von Baer, who showed on careful observation that embryos never pass through the adult stages of other animals; they only resemble each other's early embryonic stages. So the honest statement for your answer sheet is that the embryonic resemblance is real and does support common ancestry, but Haeckel's stronger interpretation of it was disproved.

Molecular evidence. The strongest modern evidence is inside the cell. Nearly all organisms use the same genetic code, the same twenty amino acids, DNA and RNA as information molecules, ATP as the energy currency, and closely similar core metabolic pathways such as glycolysis. Proteins that do the same job in different species, for example cytochrome c, have very similar amino acid sequences, and the number of differences between two species is small when they are closely related and large when they are distantly related. This gives a quantitative measure of relatedness that agrees remarkably well with the tree built from fossils and anatomy. Independent lines of evidence agreeing with each other is what makes the case strong.

Evolution you can watch. Some evidence comes from changes seen within a human lifetime. In England before industrialisation, collections of the peppered moth Biston betularia contained mostly white-winged moths, which were camouflaged on lichen-covered tree trunks, and few dark ones. After industrialisation, soot killed the lichens and blackened the trunks; now the dark, melanised moths were hidden and the white ones were picked off by birds, so dark moths came to dominate in industrial areas. No moth changed colour during its life; the proportion of the two kinds in the population changed. In the same way, excessive use of antibiotics, pesticides and herbicides leaves behind the few resistant individuals that were already present, and resistant populations build up in a few years. Artificial selection makes the same point: from the wild cabbage plant, Brassica oleracea, human beings selected different parts over generations to produce cabbage, broccoli, kohlrabi and kale.

Adaptive radiation is the process by which several different species evolve from a single ancestral form within a geographical area, each one fitted to a different habit or habitat. The key conditions are an ancestral population that reaches a new or newly emptied area, plenty of unused ways of making a living, and isolation that lets the descendant groups diverge without constantly interbreeding.

Darwin's finches. When Charles Darwin visited the Galapagos Islands he found an extraordinary variety of small black birds. He concluded that they had all descended from the original seed-eating ground finches that reached the islands from the mainland. Once there, different populations took to different foods, and the beak is the tool that changed most: some kept stout crushing beaks for seeds, some developed slender beaks for insects, and some became vegetarian, feeding on buds and fruit. Same island group, one ancestral stock, many beak types and many species. That is adaptive radiation in one sentence.

Australian marsupials. Australia was isolated as an island continent, and a large number of marsupials, each different from the others, evolved there from a single ancestral marsupial stock. There are marsupial moles that burrow, marsupial mice, the numbat that feeds on termites, the flying phalanger that glides between trees, the cuscus that lives in trees, the marsupial wolf and the marsupial cat. Once again: one ancestral stock, one land mass, many descendant species suited to many ways of life.

Adaptive radiation versus convergent evolution. Here is the subtle part. In another part of the world, the placental mammals underwent their own adaptive radiation from their own ancestral stock. When you place the two radiations side by side, pairs of animals look strikingly similar even though they are only distantly related: the placental wolf and the Tasmanian wolf, the placental mole and the marsupial mole, the anteater and the numbat, the flying squirrel and the flying phalanger, the mouse and the marsupial mouse, the lemur and the spotted cuscus, the bobcat and the Tasmanian tiger cat. Each pair looks alike because a similar habitat selected for a similar body plan. So when more than one adaptive radiation has happened in isolation and the results resemble each other, that resemblance is convergent evolution. Adaptive radiation describes the spreading out from one ancestor; convergent evolution describes the coming together in appearance of two lineages that were never close relatives. Note also that adaptive radiation is a case of divergent evolution at the level of a whole group of species, because all the descendants diverge from one starting form.

Lamarck. Jean Baptiste Lamarck proposed that organisms change during their own lifetime through the use and disuse of organs, and that these acquired changes are passed on to the offspring. His famous illustration was the giraffe: ancestors stretched their necks to reach leaves on tall trees, the neck lengthened a little during each animal's life, and the lengthened neck was inherited, so over generations giraffes came to have long necks. The theory is rejected today for one clear reason: characters acquired by the body during life are not written into the reproductive cells, so they are not inherited. A blacksmith's strong arms, a person's tan, or a mouse whose tail has been cut off do not produce children with those features. Only variations present in the genetic material can be passed on.

Darwin. Charles Darwin sailed around the world on the ship H.M.S. Beagle, observing plants, animals and fossils, and after many years published his conclusions with the parallel support of Alfred Russel Wallace, a naturalist who had been working in the Malay Archipelago and had independently reached the same idea. Darwin's argument runs as a chain. Every population produces far more offspring than can possibly survive on the available resources, an idea he took from Malthus's writing on populations. This creates a struggle for existence. The individuals in a population are not identical; they vary, and some of that variation is heritable. Individuals whose variations happen to suit the environment better leave behind more surviving offspring than the rest. Over many generations the useful variations accumulate and the population changes. This is natural selection, sometimes summed up in Herbert Spencer's phrase about the survival of the fittest.

Two clarifications are needed here. First, fitness in this context does not mean strength or health; it means reproductive fitness, that is, the number of offspring an individual leaves that themselves survive to reproduce. An animal that lives long but leaves no descendants has zero fitness. Second, nature does not create variation to order. Variation already exists, and nature simply selects among it. Darwin's two central concepts are branching descent, which explains why organisms fall into a tree of relationships, and natural selection, which explains adaptation. The one thing Darwin could not explain was where the variation came from and how it was inherited, because Mendel's work was not known to him.

Hugo de Vries. Working on the evening primrose, Oenothera lamarckiana, de Vries observed offspring that differed sharply and suddenly from the parent plants. He called such large, sudden, heritable changes mutations and argued that mutation, not the slow accumulation of small differences, is what brings about the origin of species. Because a new species could in his view appear in a single step, he called this saltation, meaning a jump. Compare the three ideas carefully. Darwinian variations are small, numerous and directional in their effect, and evolution by them is gradual; de Vries' mutations are large, random and undirected, and produce change in a single step. Lamarck's changes are acquired during life and are not heritable at all, which is why his theory fails. Modern biology keeps mutation as the ultimate source of new variation but keeps natural selection as the process that decides which variants spread, so the present picture combines Darwin and de Vries rather than choosing between them. That combined account is the modern synthetic theory of evolution. It brings together Darwin's natural selection with the genetics that Darwin lacked, and treats evolution as a change in the allele frequencies of a population brought about by five interacting agents: mutation, genetic recombination, gene flow, genetic drift and natural selection. Mutation and recombination supply the raw variation, gene flow and drift move allele frequencies about, and natural selection is the one agent that shapes that variation into adaptation. Speciation follows when populations become reproductively isolated and their gene pools drift apart far enough that they no longer interbreed.

A population is a group of individuals of the same species living in one area and interbreeding. All the alleles of all the genes carried by that population together form its gene pool. The frequency of an allele is simply the fraction of all copies of that gene in the pool which are of that type. The Hardy-Weinberg principle states that in a large, randomly mating population the allele frequencies remain constant from generation to generation, and the sum of all allele frequencies for a gene equals 1. A population in this state is said to be in genetic equilibrium, and by definition it is not evolving.

For a gene with two alleles, let p be the frequency of the dominant allele A and q the frequency of the recessive allele a. Since these are the only two alleles, p + q = 1. Under random mating, the chance of an AA individual is p multiplied by p, the chance of aa is q multiplied by q, and an Aa individual can be formed in two ways, A from the father with a from the mother or the other way round, which gives 2pq. Adding the three genotypes gives the binomial expansion of (p + q)², that is, p² + 2pq + q² = 1. Here p² is the frequency of homozygous dominant individuals, 2pq is the frequency of heterozygotes and q² is the frequency of homozygous recessive individuals.

Worked example. Suppose a recessive condition is shown by 4 individuals in every 100 in a population that is in equilibrium. Then q² = 0.04, so q = 0.2 and p = 1 - 0.2 = 0.8. The frequency of carriers is 2pq = 2 x 0.8 x 0.2 = 0.32, that is 32 in every 100. The frequency of homozygous dominants is p² = 0.64. Check: 0.64 + 0.32 + 0.04 = 1. Two traps to avoid. First, the people who show the recessive trait are q², not q, so always take the square root before you do anything else. Second, the carriers are 2pq, not pq, because the heterozygote can be formed in two ways.

The principle only holds if a set of ideal conditions is met, and these are the five conditions you should be able to list. One, the population must be very large, so that chance sampling of gametes has no measurable effect. Two, mating must be random with respect to the gene concerned, with no preference for particular mates. Three, there must be no mutation, so no new alleles are created or destroyed. Four, there must be no gene flow, meaning no migration of individuals into or out of the population. Five, there must be no natural selection, so all genotypes must survive and reproduce equally well. If all five hold, the allele frequencies cannot change and the population stays at equilibrium.

Real populations rarely meet all five, and the departures are exactly the factors that disturb equilibrium, which is another way of saying they are the agents of evolution. Gene migration or gene flow occurs when individuals move between populations and breed; the gene pool losing them and the gene pool receiving them both change, and if the migration is repeated the two populations become more alike. Genetic drift is a change in allele frequency caused purely by chance, and it matters most in small populations, where a few accidental deaths can remove an allele entirely. When drift happens to a small group that has just colonised a new place, so that the new population starts with a set of allele frequencies quite different from the original one, the founding individuals are called the founders and the effect is called the founder effect; the drifted population may in time become a separate species. Mutation introduces genuinely new alleles into the gene pool. Genetic recombination during sexual reproduction reshuffles alleles into new combinations, creating variation on which selection can act. Natural selection changes frequencies systematically by allowing some genotypes to leave more offspring than others. Selection is the only one of these that reliably produces adaptation; drift, by contrast, is random and can just as easily remove a useful allele as a harmful one.

A last point on how the principle is actually used. Because the equation predicts what the genotype frequencies should be if nothing is happening, you can compare the observed frequencies with the predicted ones. If they match, the population is at equilibrium for that gene. If they differ, something is changing the gene pool, and that measured difference is a measure of the extent of evolution going on. So the Hardy-Weinberg principle is not a claim that populations do not evolve; it is the yardstick against which evolution is detected.

Natural selection acts on a character that varies in a population, and it can act in three different ways. To follow the three types, picture the population as a list of measurements of one character, say body size, with most individuals close to the average value and fewer and fewer individuals as you move towards either extreme.

Stabilising selection. Here the individuals with the average value of the character do best, and both extremes are selected against. The average value of the character does not shift, but the amount of variation in the population becomes smaller, since fewer and fewer individuals lie far from the mean. More individuals therefore end up close to the mean character value. Human birth weight is the standard illustration: very small babies and very large babies both faced greater difficulty in the past, so weights near the middle were favoured. Stabilising selection tends to keep a well-adapted population as it is, so it does not push evolution forward.

Directional selection. Here one extreme is favoured and the other is selected against, so the whole distribution shifts in one direction and more individuals acquire a value other than the mean character value. The peppered moth is the standard example: once the tree trunks turned dark with soot, selection consistently favoured darker and darker moths, so the population moved towards one end of the colour range. Development of resistance to an antibiotic or a pesticide follows the same pattern, with the population shifting steadily towards higher resistance. Directional selection is what produces visible adaptive change over time.

Disruptive selection. Here both extremes are favoured and the intermediate individuals are selected against, so more individuals acquire the peripheral character values at both ends of the range. Instead of one hump in the distribution there come to be two, so the spread of the character widens even though the average value may not move at all, and the population may eventually split into two distinct groups. This can be the beginning of the formation of two species if the two groups also stop interbreeding. An example is a bird population where very small beaks handle small soft seeds well and very large beaks crack large hard seeds well, while medium beaks do neither job efficiently.

A quick way to identify them in a question: if the variation shrinks and the average stays put, it is stabilising; if the average moves, it is directional; if the population develops two peaks, it is disruptive. Also do not confuse any of these with artificial selection, in which human beings, not the environment, do the choosing, as when wild cabbage was bred into cabbage, broccoli, kohlrabi and kale.

The story is reconstructed mainly from fossils, and the details keep being refined, but the outline you are expected to know is stable. About 15 million years ago primates called Dryopithecus and Ramapithecus existed. Both were hairy and walked something like gorillas and chimpanzees. Of the two, Dryopithecus was more ape-like and Ramapithecus was more man-like. The two names are easy to swap, so fix that pairing firmly.

Man-like fossil bones about 3 to 4 million years old have been discovered in Ethiopia and Tanzania in eastern Africa. These primates were hairy, walked upright and were short, roughly four feet tall. Around 2 million years ago the Australopithecines lived in the East African grasslands. Evidence suggests they hunted with stone weapons but ate mainly fruit.

The first member of our own genus was Homo habilis, with a brain capacity of about 650 to 800 cubic centimetres. These people probably did not eat meat. Around 1.5 million years ago came Homo erectus, with a larger brain of about 900 cubic centimetres, and this species probably did eat meat; the first fossils of this kind were found in Java. Neanderthal man, with a brain size of about 1400 cubic centimetres, lived in near east and central Asia between about 1,00,000 and 40,000 years ago; they used hides to protect the body and buried their dead.

Modern Homo sapiens arose in Africa and moved out across the continents during the ice age between about 75,000 and 10,000 years ago, gradually developing into the distinct populations we see today. Pre-historic cave art was being produced about 18,000 years ago, and agriculture began around 10,000 years ago, after which settled human communities appeared and the pace of cultural change far outran the pace of biological change. Two things to keep straight: human beings did not descend from any living monkey or ape, but share common ancestors with them, and the sequence of fossils is a branching bush rather than a single ladder, with several of these forms overlapping in time rather than following one another neatly.