Principles of Evolutionary Change: Darwin's Theory of Evolution PDF

2 Principles of Evolutionary Change Key Concepts natural selection heritable variation reproductive success fitness genes chromosomes Mendelian genetics genotype phenotype mutation DNA heritability of characteristics genome-wide association behavioural epigeneti...

2 Principles of Evolutionary Change Key Concepts natural selection heritable variation reproductive success fitness genes chromosomes Mendelian genetics genotype phenotype mutation DNA heritability of characteristics genome-wide association behavioural epigenetics group selection individual selection gene selection altruism the selfish gene Darwin’s ideas have had a major impact on our understanding of the relationship between evolution and behaviour. In this and the next chapter we consider in more detail the foundations that he laid for evolutionary psychology and the contributions made by subsequent evolutionists. Since much of the work on the relationship between evolution and behaviour has been conducted on non-human species we consider a number of examples from the literature on animal behaviour. No understand- ing of evolution would be complete without considering genetics. What exactly a gene is and what it does are introduced. Darwin’s Theory of Evolution Artificial Selection The Nobel laureate Herbert Simon once argued that a cow is a man-made object. What he meant was that domestic animals such as cows, chickens and dogs have been selectively bred for features that humans can make use of. In a similar way that humans have fashioned tools and other artefacts from natural materials, so they have fashioned living organisms to fulfil their needs. Animal breeders have known for centuries that if you mate individuals that have desirable characteristics, then their offspring are also likely to have the favoured characteristics. If, for ex- ample, you breed large-breasted turkeys together and constantly select and breed from the progeny with the largest breast muscle, then over a number of generations you can substantially increase the amount of breast muscle in your flock. Therefore, by choosing individuals with a particular feature you can, over a number of generations, create a change in the direction that you have selected. Selective breeding has also been responsible for more radical changes in an organism’s physical form. The wide variety of domestic dogs that we see today – from Chihuahuas to St Ber- nards – have been created, in only a few hundred years, from a primitive wolf-like ancestor (see Figure 2.1). The same applies to plants, with many quite different vegetables sharing a common ancestral root. As Mark Twain quipped, ‘Broccoli is merely a cabbage with a college education’. Therefore, many organisms are the way they are today because their traits have been selected not by nature but by human beings, a process known as artificial selection.  30 Principles of Evolutionary Change Figure 2.1 Artificial selection has created a great range of domestic dog breeds from a primitive wolf-like ancestor in only a few centuries. Natural Selection In The Origin of Species, published in 1859, Darwin used the evidence of artificial selection to propose a theory of evolution that he called natural selection. This analogy, however, can be mis- leading. In artificial selection, there is a guiding hand; someone is deciding which traits are desirable and which ones are not. There is also an ultimate goal (e.g. to have plumper turkeys) and each gener- ation selected is a step on the way to achieving this goal. In evolution, there is no omnipotent being choosing which organism should survive and which should be consigned to oblivion, and there is no ultimate goal that the selection process is trying to achieve (see Dawkins, 1986; Stewart-Williams, 2018). Most biologists consider that natural selection plays the most crucial role in evolution; but that is not to say they believe it is the only thing that has led to life being the way it is today. Pro- cesses as diverse as disease, climate change and extra-terrestrial collisions have played key roles in pruning the tree of life into its present shape, but natural selection (and, as we will see in Chapter 3, sexual selection) leads not simply to change, but to adaptive change. NATURAL SELECTION AND SURVIVAL OF THE FITTEST As we saw in Chapter 1, natural selection is based on differential reproductive success of heritable characteristics that vary in a population. So individuals that happen to have genetically influenced advantageous characteristics in a given environment will be ‘favoured’ in terms of the number of surviving offspring they produce. Since individuals that survive to reproductive age are generally viewed as being physically fitter than those that don’t (for example, they may have superior running ability or more acute hearing), this led to the term ‘survival of the fittest’ being used as shorthand for natural selection. Interestingly Darwin did not use this phrase in the original version of The Origin of Species but incorporated it into later editions at the suggestion of his friend and contemporary Alfred Russel Wallace from a suggestion by another contemporary Herbert Spencer (Wallace, 1864; Spencer, 1864).  Mendel and Post-Mendelian Genetics 31 In many ways ‘survival of the fittest’ is an unfortunate term since fitness means different things to different people (Dawkins, 1982; Dickins, 2011). In the years since The Origin of Species first appeared, fitness has been used in a number of different ways ranging from its original ‘robust and hearty’ type of meaning to a technical measurement of fecundity, often leading to debates where the protagonists have argued at cross-purposes (Dawkins, 1982; Cartwright, 2016). During the twentieth century, evolutionists began to use the term fitness as a measure of how successful an individual is at reproducing, that is its lifetime reproductive success. Hence if you have three sur- viving children in your life your fitness will be three. Later on, in Chapter 7, we’ll see that in recent years the notion of fitness has been expanded beyond consideration of an individual’s own offspring. Mendel and Post-Mendelian Genetics Although Darwin solved the mystery of the mechanism by which evolution takes place, his solution was only partial. The problem was that Darwin had only a sketchy idea of the mechanism of inher- itance, the way in which traits are passed from parent to offspring. Again, as we saw in Chapter 1, it was Gregor Mendel’s work on pea plants that led to the mechanism of inheritance – genes – being discovered. Here we expand on Mendel’s findings and discuss developments in genetics that have occurred since his time. Mendel’s Findings The results of Mendel’s experiments gave rise to three important findings that ultimately led to the development of the science of genetics. First, they showed that traits are caused not by a single gene, rather that genes operate in pairs; in sexually reproducing species, a single copy of each gene is passed to the offspring in each of the parental gametes (the male and female sex cells). Second, his work revealed that the relationship between the genes that an individual pos- sesses (now called the genotype) and that individual’s physical structure (the phenotype) is rather more complex than it might appear. Two pea plants, for instance, might both have yellow peas, but the genes that specify that colour might be different; their phenotype is the same, but their genotype is different. This arises because one member of the gene-pair is dominant to the other, which is technically called recessive. Dominant simply means that the existence of a single copy is enough for that trait to be expressed, irrespective of the nature of the other copy. Mendel found that for pea plants, the yellow pea gene is dominant to green. This means that if a plant inherits a single copy of the yellow gene from either of its parents, then it will have yellow peas, even though it might also have a gene for green peas. The only way to obtain green peas is for the plant to inherit a copy of the green gene from both of its parents. In the notation used by geneticists the dominant gene is represented by the initial letter of the trait it specifies, thus the gene for yellow peas is denoted as Y. Rather confusingly, a lower-case version of the same letter is used to denote the recessive gene; thus in this case y is used for green. We can now see why two plants with different genes can have the same phenotype when it comes to a particular trait. A pea plant that has the genotype YY will have yellow seeds, as will a pea plant that is Yy; each has a different genotype, but the same phenotype. When an organism has two similar copies of a gene (such as YY or yy), it is said to be homozygous (homo = ‘alike’) for that trait; when it has two different copies (such as Yy), it is said to be heterozy- gous (hetero = ‘different’). Finally, his experiments showed that inheritance is particulate, rather than the result of a process of blending (see Chapter 1). For traits controlled by single pairs of genes such as seed 32 Principles of Evolutionary Change colour in pea plants, the results are always one thing (e.g. yellow) or the other (e.g. green), with no half-measures permitted. Previously the problem with particulate theories was that they could not account for the contemporary observations (such as traits skipping generations; see Box 2.1). Mendel’s discovery that traits are determined by paired genes went a long way towards solving this problem. One of the curious things about this discovery is that nowhere in Mendel’s work did he state it explicitly (Hartle and Orel, 1992), although it is true that it is a natural conclusion to draw from his experiments. Box 2.1 Mendel’s Demonstration of Colour Dominance in Pea Plants One of the mysteries of inheritance was why characteristics found in one parent were often not found in any of the offspring. If plants which breed pure for yellow peas are crossed with plants which breed pure for green peas, all of the resultant offspring will have yellow peas. If you then cross each member of this second generation with its siblings then some of the resultant offspring will once again have green peas. Again this observation is impossible for a blend model to ex- plain, since yellow should never produce green when crossed with yellow. Table 2.1 Mendel’s demonstration of colour dominance in pea plants Generation 1 Phenotype yellow peas green peas Genotype YY yy Gamete Y y Leads to … Y Y y Yy Yy y Yy Yy Generation 2 Phenotype yellow peas (all offspring) Genotype Yy (all offspring) Gamete Y and y Leads to … Y y Y YY Yy y Yy yy Generation 3 Phenotype yellow peas green peas Genotype YY; Yy; Yy yy Ratio of phenotypes 3 1 Note: It was the ratio of 3:1 (actual figures 6022:2001) in favour of yellow over green in the third generation that led Mendel to realise that the underlying genotypes YY, Yy, yY and yy were produced in the third generation.  Mendel and Post-Mendelian Genetics 33 Box 2.1 (cont.) It was observations such as this that led Mendel to formulate his theory. In Table 2.1 we can see that pea plants that are pure breeding for yellow and green seeds are both homozygous. When these are all crossed with each other all possible offspring are heterozygous, but, because yellow is dominant, all are phenotypically yellow. When these are all crossed to produce a third generation, green re-emerges (the homozygous form yy) in the ratio 1:3 with yellow. Notice that the green colour, lost to the previous generation, is expressed since it does not have a copy of the ‘masking’ dominant gene (Y). Box 2.2 Mendel’s Original Laws of Genetics (Using Modern Terminology) 1. Inheritance is particulate (i.e. parental genetic material is discrete and does not blend togeth- er), with each parent making an equal contribution to the progeny. 2. Characteristics are influenced by genes occurring in pairs (one contributed from each parent). The complete set of genes of an individual is known as its genotype. 3. Genes exist in at least two or more alternate forms that are called alleles. Where there are only two potential alleles at each locus then three types of gene combinations become possible. In the case of Mendel’s pea plants, combinations of genes for pea colour exist as YY, Yy and yy. Where an individual has identical genes at a specific locus (e.g. YY and yy) it is said to be homozygous for that characteristic. In contrast, where an individual has different genes at a specific locus (e.g. Yy and yY) it is said to be heterozygous. The complete description of an individual’s characteristics is known as its phenotype. 4. Dominant alleles override recessive alleles in their expression in the phenotype. Recessive genes can be expressed in the phenotype only when they occur in a double dose (the homozy- gous recessive condition, e.g. yy). 5. Only one of a pair of parental alleles is passed on to each of the offspring. Genes for different characteristics are passed on individually rather than being attached to each other (in the lan- guage of genetics they are segregated). This is sometimes known as Mendel’s first law. 6. Phenotypic features that occur together in the adult will not necessarily appear together in the offspring. This is called independent assortment and is a result of segregation. This is sometimes known as Mendel’s second law. In 1910 the Cambridge geneticist Reginald Punnett devised a simple way of illustrating the potential genotypes that can result from various breeding experiments – the Punnett Square. In the case of pea colour, yellow peas (Y) are dominant to green peas (y). Hence if we cross a heterozy- gous yellow pea plant (Yy) with another heterozygous yellow pea plant (Yy) we can create a simple Punnett Square (see Figure 2.2) that demonstrates the genotype ratios produced thus: 34 Principles of Evolutionary Change Y y Figure 2.2 Punnett Square to demonstrate pea colour. Note that both YY and Yy genotypes have a yellow phenotype, giving us the 3:1 ratio as in Box 2.1. YY Yy Y y Yy yy Following a large number of breeding experiments with pea plants Mendel proposed a number of conclusions which have become known as ‘Mendel’s laws of genetics’. Mendel’s laws are summarised in Box 2.2. It is important to realise that Mendel never saw a gene (nor did he use the term which came into existence in 1905); rather, he postulated that they must exist, based on their phenotypic effects in experiments such as that sketched in Box 2.1. Modifications of Mendel’s Laws Although Mendel’s ‘laws’ provided a framework for understanding the mechanism of inheritance, the picture was by no means complete. First of all, Mendel had no idea of how these units of in- heritance were physically realised, or where they were located. Second, it soon became clear that there are exceptions to most of his original suggestions. To Mendel, sources of variation came from inheriting a mixture of parental genes. However, other processes of which Mendel was unaware have been uncovered since his time and these lead to a more complex modern-day understanding of genetics which includes other sources of variation. Genes and Chromosomes During the 1930s it was discovered that genes are located in the nucleus of an organism’s cells at specific locations on larger bodies called chromosomes. Each individual has a number of chro- mosomes which is typical of its species. In the case of humans the number is 46 (23 homologous pairs, one from each parent). One pair of chromosomes is the sex chromosomes – in humans, XX for females XY for males. The remaining 22 pairs are known as autosomal chromosomes and they have nothing to do with sex at all. The discovery of chromosome pairs provided a physical home for Mendel’s paired genes. This means that a Yy pea will have its Y gene on one homologous chromo- some, and the y on the other. The number of chromosomes varies widely between species. Fruit flies of the genus Drosophila (a group of minute flies that have been used extensively to understand inheritance) have 8, dogs have 78 and some species of plants more than 250. A gene for a particular characteristic occurs at a specific point on the chromosome called its locus (plural loci). The purpose of the Human Genome Project was to identify the loci of all of the genes in our species; we may not know all the effects that a gene might have on the phenotype, but we now know where it is. Generally a locus is home to more than one alternative form of a gene; when this occurs the alternate forms are called alleles. In the above example we saw that the colour-determining gene can have two alternate forms, Y and y; these are alleles and as such will reside at the same locus on a chromosome.  Mendel and Post-Mendelian Genetics 35 Chromosome pairs become separated during sexual reproduction so that each sperm or ovum produced has half of the normal complement of genes (it is known as a haploid cell). The process of producing cells with half the number of genes/chromosomes is called meiosis. Meiosis ensures that when sperm and ovum fuse to form a fertilised egg or zygote the number of genes an offspring has is restored (i.e. it is diploid). Note that when body cells divide they double the number of chromosomes they contain just prior to forming two new cells so that each new cell has the nor- mal number of genes – a process known as mitosis. Factors Affecting the Transmission of Genes During the early years of the twentieth century, evidence was uncovered that demonstrated that genes are not always passed on independently. Different genes may be linked together when they are passed on to offspring. If two genes are found on the same chromosome they are far more likely to be passed on together to offspring than if they appear on separate chromosomes. This means that genes may be linked together and a chromosome may be considered as a ‘linkage group’ of genes. The fact that genes occur in linkage groups does not, however, mean that all of the genes on parental chromosomes will be passed on together to the offspring. Prior to gamete (eggs and sperm) forma- tion, homologous chromosomes pair up together and exchange genes at specific points. Figure 2.3 Human chromosomes. Each one comprises two identical chromatids joined at the centromere, which divides each chromatid into a long and a short arm. This exchange of genes is known as crossing over. In this way genes become recombined and the genes that are passed on to offspring are said to have undergone recombination. Due to re- combination each individual is not only genetically unique, but also has unique chromosomes (with the sole exception of identical twins). The closer together two genes are found on a chromosome, 36 Principles of Evolutionary Change the more likely they are to be passed on together during recombination. This means the degree of linkage (the likelihood of genes being passed on together) is an indication of how close their loci are on a chromosome. If we imagine a chromosome as a chain necklace with each gene represented by a link, then homologous chromosomes would be like having two chain necklaces with an identical number of links laid side by side. Although each link on one necklace is identical in length to its coun- terpart on the other, they may be made of different materials (representing the different genes on each). Now imagine breaking each necklace in a number of places and exchanging groups of links between them. Each necklace retains its original length but now differs at certain points. Note that the closer two links were on the original necklace, the more likely they will be to find themselves together on the new one. Today we know that the chain necklace view of genes is really quite a simplification since many characteristics are made up of parts of DNA taken from all over a chro- mosome. Other Exceptions to Mendel’s Laws Linkage and recombination are the main reasons that the simple ratios Mendel predicted in breeding experiments frequently do not occur. And there are other reasons why Mendel’s foundations of ge- netics have had to be modified over the years. Crossing over is a form of chromosome mutation that is common because it occurs every time a gamete is formed (there are other, somewhat less com- mon, forms of chromosome mutation as well). Individual genes may also be subject to mutations but these are rare compared with chromosome mutations. A gene mutation involves the chemical structure of the gene being altered (by radiation, for example) but because genetic material is chem- ically quite stable the chances of a given gene mutating are one in thousands. However, given that each multicellular species has a very large number of genes (around 20,300 in the case of humans) this means that individual organisms will have quite a large number of mutations (perhaps as many as 100 in you and me). You need not worry about your 100 mutated genes, however, since the vast majority of these are neutral, that is, they have no phenotypic effect. When a gene mutation does have a phenotypic effect it is more likely to be detrimental than beneficial, but just occasionally a gene mutation can have a beneficial effect and this may be an important event for evolution (see below). In a sense then, each of us is a mutant; however, being a mutant is one of the reasons every- body is unique. At the level of the gene pool, mutations can be viewed as a source of variation in the population. Thus in modern-day genetics there are considered to be three sources of variation in a population that natural selection can work on: Mendelian variation due to the mixing together of parental genes; recombination (also called chromosome mutation where genes are exchanged between chromo- somes); gene mutations. As stated earlier, Mendel was fortunate in studying pea plants since they have a number of char- acteristics that could occur in either of two contrasting forms. He was also fortunate in that the characteristics he chose demonstrate a simple dominant/recessive relationship. Like a tossed coin which can only land heads or tails, peas are either green or yellow, they are round or wrinkled and their flowers are either red or white. Today we realise that many characteristics do not show complete dominance or complete recessiveness. In addition, many genes have more than one phenotypic effect; this is known as pleiotropy and many characteristics depend on more than  Mendel and Post-Mendelian Genetics 37 one gene, that is, they are polygenic. Finally, the way that a gene is expressed in the phenotype may be altered by a modifier gene or by other portions of the DNA. So although Mendel laid its foundation stones, the builders of modern-day genetics have modified his original blueprints on many occasions. Box 2.3 The Evolution of Our Species – from Ape to Early Archaic Homo sapiens Most experts today suggest that chimpanzees and humans diverged from a common ancestor around 7 million years ago. The earliest of the hominin (human-like) species Sahelanthropus tchadensis (‘Saharan hominin from Chad’) was uncovered in central Africa in 2002 and dates back to around 6–7 million years before present (YBP, Humphrey and Stringer, 2018). Sahelan- thropus had features intermediate between humans and chimpanzees. Although the cranial ca- pacity was relatively small, the position of the foramen magnum (the hole in the skull through which the spinal cord passes) suggests that it walked upright on two feet. This suggests that one of our earliest major hominin features was bipedalism. It is not known whether Sahelanthropus was a direct ancestor of ours, but it was certainly the most ancient hominin to be discovered since our split from a common ancestor with chimpanzees. Further ape-like fossil remains were discovered by paleoanthropologist Tim White and co-workers between 1992 and 1994 and were named Ardipithecus ramidus. It is possible that this was a descendant of Sahelanthropus, although this is an area of debate. In either event several subsequent fossil finds suggest that there may have been at least two different species of Ardip- ithecus (Lovejoy, 2009; White et al., 2010). Ardipithecus existed around 4.4 million years ago in Ethiopia. It had ape-like, long and strongly built arm bones but human-like, smallish canine teeth. Plant fossils found with Ardipithecus suggest that it was very much a woodland creature. In contrast, the ‘next stage’ of human evolution Australopithecus (various species found in a number of areas of Africa; McHenry, 2009; Humphrey and Stringer, 2018), which begins to appear in the fossil record around 4.2 million years ago, is associated with drier savannah areas, as is generally the case for hominins thereafter. The Australopithecines eventually developed into two broad categories – a large ‘robust’ form which had a massive jaw, and a smaller, lightly built, ‘gracile’ form. Although the Australopithecines were clearly able to walk upright they had relatively shorter legs and greater wrist and ankle mobility than ourselves, allowing them to retain good tree-climbing abilities. It is believed that the Homo line descended from a species of gracile Australopithecine (possibly Australopithecus afarensis; see fossil skull below) around 2.5 million years ago – again within Africa. The first of these was Homo habilis and it had a larger cranial capacity (and hence a larger brain) than Australopithecus. H. habilis used stone choppers and scrapers to remove meat from the bones of its prey (H. habilis was clearly a meat eater but there are debates over whether or not it was a hunter or scavenger). Although H. habilis probably spread to many areas of Africa it did not leave that continent. In contrast, the next species on the path to ourselves – Homo erectus, which began to appear in the fossil record around 1.9 million years ago – migrated north and eventually spread throughout much of the Eurasian continent. (Note that some experts reserve the name Homo erectus for Asian specimens and Homo ergaster 38 Principles of Evolutionary Change Box 2.3 (cont.) for African ones.) H. erectus had a brain size some 50 per cent larger than H. habilis and smaller molar teeth suggesting less reliance on uncooked plant food. It was distinctly more human-like than any previous hominin, having a less protruding face and making use of more complex stone tools. Artefacts and fossils from the Malay Archipelago suggest that H. erectus built rafts or boats and constructed shelters with stone bases. H. erectus existed for at least 1.5 million years but at some stage less than 400,000 years ago ‘archaic’ forms of our own species Homo sapiens began to emerge. Although archaic H. sapiens had brains some 20 per cent larger than H. erectus, they were still 20 per cent smaller than modern H. sapiens. Anatomically modern H. sapiens probably have existed for around 150,000 years but precisely where they first evolved is a hotbed of debate. From this brief description of the journey from ape to human we should not assume that a series of simple step-like changes occurred when one species evolved into another so that the former species no longer existed. Speciation (formation of a new species) is thought to occur when one population becomes geographically isolated from another for a lengthy period of time and is subjected to different selection pressures. After many generations the separated population becomes so different from their ancestors that they would no longer be able to breed together to produce viable offspring should they meet up once more. This is the point at which they are generally considered to be separate species. This is likely to have been the case during human evolution and indeed the fossil record shows that a number of hominin species are likely to have coexisted (Tattersall and Matternes, 2000; McHenry, 2009; Humphrey and Stringer, 2018). A number of Australopithecines probably coexisted with each other and with H. habilis. Also there is clear evidence that H. sapiens coexisted with H. erectus/H. ergaster, the latter only disappear- ing from the fossil record around 40,000 years ago. Human evolution from the first hominins to modern humans. The arrows indicate the two times when Homo erectus and Homo sapiens migrated out of Africa (after Goldsmith and Zim- merman, 2001, 278). Figure 2.4 A range of early hominin skulls. From top left to bottom right (with years before present): Homo sapiens neander- thalensis (300,000 YBP); Sahelanthropus tchadensis (6–7 million YBP); Austra- lopithecus afarensis (3.7 million YPB); Homo ergaster (1.9 million YBP).  Modern Genetics 39 Box 2.3 (cont.) Modern humans Figure 2.5 Human evolution from the first hominins 0 Neanderthals Homo sapiens to modern humans. The arrows indicate the two times when Homo erectus and Homo sapiens 1 Homo erectus migrated out of Africa. Million Years Ago Homo habilis 2 Australopithecus (several species) 3 Australopithecus afarensis, A. africanus Australopithecus 4 anamensis Ardipithecus ramidus Modern Genetics Genes and the Structure of DNA In order to comprehend the role that genes play in nature it is important to understand what a gene is and to appreciate the role that it plays in helping to create an organism. So what is a gene and what is its relationship to a chromosome? A chromosome consists of a lengthy string of genes. Physically it is a double strand of the chemical deoxyribonucleic acid or DNA, the structure of which was discovered by Watson and Crick in 1953. DNA is a truly giant molecule, sometimes as long as two inches, and resides within the nucleus of the cell. In structure, DNA is the shape of a twisted ladder (the famed ‘double helix’) with each of the ‘rails’ – known as the backbone – being made up of alternating units of phosphoric acid and deoxyribose sugar. Each link of acid and sugar is connected to a base that may come in one of four forms: adenine, thymine, cytosine and guanine (A, T, C and G); each link of sugar, acid and base is called a nucleotide (see Figures 2.3 and 2.6). The ‘rungs’ of the ladder are made up of paired bases. The properties of these bases are such that adenine can pair up with thymine, and cytosine can pair up with guanine. The bases code for the production of amino acids that are the building blocks of proteins. Each amino acid is coded for by a triplet or codon of bases (the amino acid lysine, for example, is coded for by the codon of AAG). We can think of each codon as a different word and a gene as a sentence made up of these words. It is the precise sequence of the 1 billion ‘words’ that makes up human DNA that the Human Genome Project was set up to uncover (see Box 2.4). The use of three-letter words from an alphabet of four letters leads to 64 (4 × 4 × 4) different possibilities. This is more than ample for amino acid production since only a little over 20 amino acids exist in nature. This means that a number of slightly different three-letter words have the same meaning (like synonyms). For example, in addition to AAG coding for lysine, so too does AAA. Although there are only around 20 different amino acids in existence, by stringing these together in various sequences hundreds of thousands of different proteins may be created. So, to return to the original question of what is a gene; a gene is a portion of DNA that, via its sequence of codons, codes for protein synthe- sis (see Box 2.4). And proteins, of course, are the main building blocks of the body and brain. This 40 Principles of Evolutionary Change at least is the classic view of what a gene is. Recently, however, it has been discovered that the vast majority of our DNA does not code for protein formation but that much of this non-coding DNA plays an important role in controlling the activity of the protein production portions. In fact it is now known that genes which code for proteins make up a mere 2 per cent of our DNA. The remaining 98 per cent used to be thought of as ‘junk DNA’, serving no purpose. It is now known, however, that such non-coding DNA (‘introns’) play an important role in regulating the production of proteins (transcription; see below) by the other 2 per cent (‘exons’; see Chapters 6 and 13). In recent years the importance of these non-coding regions of the DNA has become better understood and they are considered to be a main source of individual differences between people. Such findings have led to debates about how we define what a gene is (Plomin et al., 2016; Plomin, 2018). Despite such recent discoveries and debates it is still considered that the main function of DNA in all species is the production of proteins. But how does DNA do this? When a specific pro- tein is required by a cell a portion of the double helix ‘unzips’ revealing a sequence of bases. Other, free-floating, bases within the cell attach to the exposed bases, forming a second type of molecule – messenger ribonucleic acid, or mRNA for short. This mRNA, once formed, detaches and travels to cell ‘organs’ called ribosomes where, via another form of RNA (transfer or tRNA), it forms the template for amino acid, and hence protein production. The formation of proteins from DNA is called transcription. DNA has one other main function – to make copies of itself, otherwise known as replication. Replication occurs when a strand of DNA fully unzips to form two separate strands. As before, free-floating bases attach to the exposed bases on the strand to form two new identical strands. Each species may, in theory, be defined by the specific proteins that its DNA produces. Our own species has perhaps as many as 100,000 different proteins that make up the body and all of the chemical reactions within (although the base-pair sequence has been determined the exact number and sequence of proteins that our DNA codes for is yet to be resolved, see Box 2.4). Given that a third of all protein-coding genes are expressed only in the brain (i.e. they only produce proteins here) many behavioural biologists believe that differences in personality and intellectual ability may, in part, be traced back to differences in the genetic code that we inherit from our parents (Plomin et al., 2016). Figure 2.6 Production of protein from DNA double helix in nucleus of cell. Note the DNA double helix (red and blue) and the first step of protein production (yellow and purple strands) being based on the DNA nucleotide sequence.  Modern Genetics 41 Box 2.4 The Human Genome Project – Unravelling the Code to Build a Person? Although a draft sequence of the human genome was originally published in February 2001, it was in April 2003 that the world woke up to hear the news that the entire human genome had been sequenced. Humans were not the first species to be sequenced – the DNA for 39 species of bacteria had already been determined as well as a species of yeast, a nematode worm, the Drosophila fruit fly and a mustard weed. But unravelling our own DNA was arguably a quantum leap forward compared with sequencing these ‘simpler’ species. So now we know the precise base-pair sequence for human DNA and, as a consequence, we also know how many genes it takes to build a person. This is a major triumph of molecular biology. Indeed some evolutionists claimed at the time that this would be the biggest scientific development the world had ever seen (Ridley, 1999; 2003). Others, however, suggested that this breakthrough raises as many questions as it answers. One important question that we might ask is what does knowing the human genome sequence really tell us about how we are built? Another is, given that our genes vary from person to person at hundreds of different loci, whose DNA has been sequenced? Turning to the second question first, the Human Genome Project (HGP) has really produced the average or consensus human sequence for 200 people. The first question, unfortunately, is not so easily answered. Genes specify the sequence of amino acids that link up to form large molecules called polypeptides – biochemically one gene codes for one polypeptide. Proteins are then created by linking a number of these polypeptides together. Humans are built from proteins (based on 3 billion nucleotide base pairs – of which we vary in 30 million pairs). You might think that knowing the base-pair sequence allows us to determine the polypeptide sequence and that this, in turn, allows us to determine the precise number of proteins that make up a human. Wrong. Polypeptides literally fold themselves up into various shapes in order to form proteins. But many polypeptides can fold themselves up in different ways resulting in different proteins. How precisely they fold themselves depends on the presence of a number of smaller molecules in a given cell, such as sugars, and on the presence of other proteins. This is why, although the base-pair sequence is now known, this information alone does not tell us the protein sequence. It also helps to explain why, although humans have perhaps 90,000–100,000 proteins, these are created from fewer than 22,000 genes (which incidentally is not enormously more than the fruit fly with 13,000 and the nematode worm with 18,000). In September 2011 the next step was an- nounced – to identify at least one protein for each of the exons (protein-producing genes). This project, called the Human Proteome Project (HPP), was launched at the Geneva World Con- gress and has subsequently led to a number of genes (and their effects on protein production) be- ing identified which are associated with specific diseases (Karczewski and Snyder, 2018). Such findings not only bring us a step closer to understanding the role that proteins play in body and brain development and function but also open the door to the notion of personalised medicine. This means that once specialist clinicians determine an individual’s entire genome sequence they may then be able to deliver specific medications and other treatments tailor-made for that individual (Banku and Abalaka, 2012; Plomin, 2018). Moreover, comparative DNA sequence analysis from distinct human populations has confirmed that we all originated from a series of migrations out of Africa. 42 Principles of Evolutionary Change Since genes code for the synthesis of the building blocks of the body they are frequently considered to be the blueprints that dictate how an organism is built (Plomin, 2018). As Oxford zoologist Richard Dawkins (1982) pointed out nearly 40 years ago, however, this analogy may be an oversimplification; a better one is that genes make up a recipe for development. In a recipe for baking a cake, for example, the ingredients are specified by instructions prior to baking. The quality and availability of ingredients and the temperature and humidity of the oven, however, will all have an effect on the final product. In this analogy the recipe consists of the genetic code that is inherited and the other variables are equivalent to environmental input during development which, in ourselves, range from peer pressure and parental attitude to upbringing, through diet and illnesses to education. As Dawkins rightly points out, once it is taken out of the oven there is ‘no one-to-one reversible mapping from words of recipe to crumbs of cake’ (Dawkins, 1982). To Dawkins it is practically impossible to partition out the specific effects of individual genes. Nor is it easy to deter- mine the precise effects of specific environmental variables on the outcome of a characteristic in an individual. Despite this it is possible to make an overall estimation of the relative contribution that genes make to individual differences (Plomin, 2018; see Chapters 6 and 13). Heritability of Characteristics As mentioned previously, animal breeders have long taken advantage of the emergence of individual differences in their livestock to breed for exaggerated features. The estimation of the extent to which we are able to breed for a characteristic is called its heritability. Canadian evolutionists Martin Daly and Margo Wilson define heritability as ‘the proportion of the observed phenotypic variance that can be attributed to correlated variance in genotypes’ (Daly and Wilson, 1983). Put simply, this is an estimation of the extent to which a characteristic in a population is due to genetic rather than environmental components. Daly and Wilson use egg production to illustrate the notion of heritabil- ity. In recent years egg producers have been able to increase the size of domestic hen eggs far more successfully than they have been able to increase the number of eggs produced by each hen. This is because there is more genetic variability available for egg size than egg number. Egg size therefore has a larger heritability than egg number. Note that estimations of heritability must always be qualified by specifying the population or, in effect, the gene pool. As Daly and Wilson correctly assert, in a different population of hens heritability for these two characteristics might be equal or even reversed. This might be the case, for example, in the wild-living equivalent of modern-day domestic hens, the Red Burmese junglefowl. In domestic hens it may be that egg producers have pushed their flocks to a point where it is very difficult to increase further the number of eggs that a hen can produce in a year (which is around 365, incidentally) but that there still remains a greater degree of genetic variation for egg size. This means that under current rearing conditions there is little variation that can affect egg number. If a number of gene mutations were to arise and rearing conditions were greatly altered, then egg num- ber might become more heritable. In contrast to domestic hens, for junglefowl the number of eggs laid is much smaller and varies with the seasons. Since junglefowl have not been selectively bred to produce so many eggs throughout the year, in populations of this sub-species it may be possible to push this number up greatly. So for Red Burmese junglefowl the number of eggs laid may be highly heritable when compared to their domestic cousins. Studying the heritability of egg production certainly has practical implications – an egg producer’s livelihood may depend on it – but when we turn our attention to human cognitive and be- havioural characteristics the notion of a genetic basis for such features takes on social and political  Modern Genetics 43 connotations. The idea of human abilities being largely inherited has been misused on a number of occasions during the twentieth century. It is important then that we are clear about the limitations of heritability studies. To say that a characteristic is estimated to have a high degree of heritability does not mean that little can be done to modify it by changing the environment. Adult height may be a highly heritable characteristic in humans, but it may be substantially reduced when a child is reared with a highly restricted diet. So how might we estimate the degree of heritability of human abilities? One method is to examine the phenotypic variation in people who are genetically related to different degrees. In such studies comparing the correlation between monozygotic (identical) twins with the correlation between dizygotic (non-identical) twins on various traits has been of particular use (Mc- Farland, 1999; Plomin et al., 2016). The idea here is that, since identical twins share 100 per cent of their genes, whereas non-identical twins share 50 per cent of theirs, we would expect the correlation for any trait between the former to be much higher than between the latter if genes play a prominent role in the trait. Studies of the correlation between IQ scores in twins, for example, suggest a corre- lation of around 0.75 for identical twins and around 0.38 for non-identical twins (based on five stud- ies, see Alcock, 2001; see also Chapter 13). This finding has been taken by behavioural geneticists as evidence that intelligence shows quite a high degree of heritability (Alcock, 2001; Plomin and von Stumm, 2018). We should bear in mind, however, that 0.75 is not a perfect correlation (i.e. 1.0), thus leaving room for environmental influence. In contrast to intelligence, most studies of personal- ity traits suggest that about 40 per cent of the variation is due to genes, thus leaving around 60 per cent due to environmental differences between individuals (Plomin et al., 2016; Plomin, 2018; see also Chapters 6 and 13). In addition to comparing people who are genetically related on a number of correlational scales a recently developed technique known as genome-wide association studies (GWA) is currently revolutionising behavioural genetics. We will consider this new method in some Figure 2.7 In addition to having close physical similarities, twins have very similar scores on intelligence tests. 44 Principles of Evolutionary Change detail in Chapter 13, but for now, in a nutshell, GWA studies rapidly scan markers across the entire genome of a large number of individuals in order to establish genetic variations associated with specific traits (Plomin, 2018). Behavioural Epigenetics – a Return to Lamarckism? From our discussion of heritability above, it would appear that we can divide behavioural responses and internal states up into those which can be attributed to the environment and those which can be accounted for by our genes. In recent years however evolutionists have developed a new approach in an attempt to understand how the environment might feedback on the genes we inherit. This field, which is known as epigenetics, in a sense, opens the door to a degree of Lamarckian inheritance. During the last 20 years evidence has accumulated that demonstrates how various environmental influences from diet to environmental toxins to social experiences (good and bad) influence the expression of various specific genes (Pembrey et al., 2006; Moore, 2015; Plomin, 2018). The ex- pression of a gene means how often it is active (i.e. involved in protein production). It is now well established that certain life experiences affect which genes are active. This means that two people who share similar or even the same genes can end up with quite large differences, for example, in personality traits. One study of identical twins suggested that differences in the life experiences of a pair of identical twins can have a knock-on effect in terms of how risk aversive each becomes (Kaminsky et al., 2008). Such differences are believed to be related to changes in DNA methylation (a chemical reaction that turns off a gene) which, in turn, affects the development of neurones in the brain (note there are also other mechanisms by which epigenetics functions). Hence an early bad social encounter, for example, that only one of a pair of twins experiences can lead that member of the pair to become more risk aversive than the other later on in life (because this event leads to certain genes being less active during development). What has become known as behavioural epi- genetics therefore helps us to understand why identical twins, despite sharing a genome, are never really quite identical in personality. Behavioural epigenetics is, of course, not just concerned with why identical twins differ but with why we cannot determine how any individual’s personality will develop simply by reading their genetic code. Moreover, it demonstrates how nurture alone cannot explain the development of personality. Finally, it is worth noting that some epigenetic changes are passed on to the offspring in which case it is known as ‘transgenerational epigenetics’ or TGE. A good example of TGE is the finding that fathers who smoke at an early age are more likely to have heavier 9 year old boys (Pembrey et al., 2006). The discovery of transgenerational epigenetics might be seen as opening the door to at least some restricted forms of Lamarckian inheritance. Most ex- perts dispute this as they perceive TGE as one example of phenotypic plasticity (Cartwright, 2016). We return to a discussion of behavioural epigenetics when we consider evolutionary psychopathol- ogy and personality in Chapters 12 and 13, respectively. Gene Flow and Genetic Drift Up until now we have considered natural selection to be the cause of change in organisms over evo- lutionary time scales. Today natural selection is certainly seen as the prime mover in evolution. How- ever, other causes of change can occur in a population. When animals move from one population to another they may, by chance, have a genetic makeup different from the new population. If the new makeup confers an advantage in the new local environment then the population can alter quite rapid- ly. This process is called gene flow since new genes ‘flow’ into a new environment. Another force for  Levels of Selection – the Fittest What? 45 change called genetic drift consists of changes in a population that may build up by chance because they are selected neither for nor against. Genetic drift may operate in all populations but is believed to be a significant factor only in very small ones since each chance mating will have a bigger impact. Occasionally, however, genetic drift might have a major impact on a population due to a process called the founder effect. This means that when a new population is started by a small number of individuals, as might happen when a group of rodents colonises an island for the first time, then they are likely to have only a small proportion of the genes from the population they left behind. In these circumstances the founder effect may mean that genetic drift is initially of great importance. This means that under some conditions, processes other than natural selection can lead to evolutionary change, but under most circumstances natural selection is the prime mover in evolution. Levels of Selection – the Fittest What? As we saw earlier, natural selection is frequently referred to as the survival of the fittest. But the fittest what? Species? Group of individuals? Individuals themselves? Since Darwin first introduced the notion of survival of the fittest, people have frequently taken this to mean survival of the species. As we will see when considering the relationship between evolution and behaviour, nothing could be further from the truth. Group Selection In 1962 Scottish biologist Vero Wynne-Edwards published a book entitled Animal Dispersion in Re- lation to Social Behaviour. It was destined to be one of the most cited texts on evolution published in the twentieth century. But unlike Darwin’s opus a century before, Wynne-Edwards’ book became fa- mous for being wrong. In Animal Dispersion Wynne-Edwards proposed that an animal’s behaviour is shaped by evolution to aid the survival and reproduction of its group. If many twentieth-century evolutionary theorists had been decidedly vague about the level at which natural selection operates, Wynne-Edwards was very clear. For him it operated at the level of the group. Wynne-Edwards used this group selection theory to explain many aspects of animal social behaviour. And it appeared to make a great deal of sense. Why else should songbirds give alarm calls to warn other members of their group that predators were present? Why else should the runt of a litter give up its life for its littermates? And why else should animals come together so frequently from communal roosts to communal migrations unless they were assessing the size of the population and using this infor- mation to control breeding so as not to overexploit resources? Animals, unlike humans, Wynne-Ed- wards argued, help the survival of the group (and hence the species) by only breeding when times are right. When times are hard, the group as a whole holds back and therefore survives into future generations. In Wynne-Edwards’ eyes animals had evolved to be truly self-sacrificing or altruistic. Box 2.5 The Evolution of Our Species – the Emergence of Modern Homo sapiens Currently two competing views exist as to where and when modern Homo sapiens emerged. The multi-regional hypothesis is the view that gradual changes in numerous Eurasian populations of Homo erectus led, via archaic Homo sapiens, to anatomically modern humans. In contrast,  46 Principles of Evolutionary Change Box 2.5 (cont.) Figure 2.8 This skeleton of a Homo erectus (H. ergaster), ‘Turkana boy’ (around 8 years old), is the most complete specimen of an archaic human ever discovered. It was found in Kenya near Lake Turkana in 1984 and dated to around 1.6 million years ago. the out-of-Africa hypothesis suggests that a later African population of H. erectus led (again via archaic H. sapiens) to modern H. sapiens and that these then gradually spread out of Africa displacing earlier hominins. If the former argument is correct, then our direct ancestors lived outside of Africa for well over a million years, but if the latter is correct, then we all share a recent common ancestor that lived in Africa between 100,000 and 200,000 years ago. It also suggests that our ancestors have only recently colonised the rest of the world (recently on a geographical timescale, that is). Which hypothesis is correct? The advocates of the multi-regional hypothesis rely largely on their interpretation of the fossil record. The proponents of the out-of-Africa theory also rely on an interpretation of the fossil record but additionally they make use of another, more powerful tool – the molecular clock. Recent developments in molecular biology have led to the new sub-field of molecular genetics. Molecular geneticists have turned their attention to the genes that exist out- side of a cell’s nucleus in order to create their molecular clock. You may be surprised to discover that genes exist outside of a cell’s nucleus. In the cytoplasm (the jelly-like material surrounding the central nucleus) of each human cell there are thousands of minuscule lozenge-shaped bodies called mitochondria. The mitochondria are responsible for providing energy for the cell via the controlled breakdown of sugars from food and are believed to have evolved from bacteria that invaded living cells around 2 billion years ago. Each mitochondrion has its own small single ring of genes that is passed on through the maternal line directly in the ova. This means that  Levels of Selection – the Fittest What? 47 Box 2.5 (cont.) mitochondrial genes remain uncontaminated by sexual reproduction or selection pressures and the only changes that occur to their DNA sequence arise through random mutations (Dawkins, 2004; Oppenheimer, 2004). Since the mutation rate is well known for mitochondria (somewhat higher than for nuclear DNA), then, by comparing the variability in base-pair sequences between people from different geographical regions, molecular geneticists have been able to estimate when our most recent common ancestor existed. It is these changes in mitochondrial DNA that constitute the molecular clock and the figure that molecular geneticists have determined through this technique is around 172,000 years ago (Ingman et al., 2000). The molecular clock is con- sidered such a powerful tool because, whereas a fossil might have left descendants, all of our mitochondrial DNA certainly had ancestors (Goldsmith and Zimmerman, 2001). Since the mito- chondrial DNA is only passed down via the female line the earliest common human ancestor of present-living humans has been called ‘Mitochondrial Eve’ or sometimes ‘African Eve’. Hence today many evolutionists support the later out-of-Africa argument and claim that African Eve was indeed the common ancestor of us all (Aiello, 1993; Lahr and Foley, 1994; Meredith, 2011; Humphrey and Stringer, 2018). Such arguments make great intuitive sense and they appeal to our sense of fair play. Unfor- tunately natural selection has nothing to say about fair play. It is merely the name given to describe the process by which some individuals are better able to pass on copies of their genes to future gen- erations than are others. Imagine a group where everybody breeds only when it is good for the group, producing fewer offspring at times when food availability is low, for example. Now imagine an indi- vidual in that population which has a mutant gene that makes it attempt to produce as many offspring as possible, selfishly ignoring the greater good of the group. It is this selfish individual which is most likely to pass on its genes under such conditions, not the altruistic ones (Dawkins, 1976). Wynne-Edwards’ thesis may well have done a service to the understanding of the relation- ship between behaviour and evolution in that it led other evolutionists to scratch their heads and decide whether it stood up to scrutiny. One evolutionist who scratched his more than able head and found the thesis to be wanting was George Williams. In 1966 Williams published a book on evo- lution called Adaptation and Natural Selection. In it he showed that when animals do cooperate it is almost always between close relatives. If animals were acting altruistically to their kin then they were really acting to promote copies of their own genes in relatives. If this is the case then it is not real altruism but ultimately selfish behaviour from the point of view of the gene. Williams’ argu- ment drew upon the most clear-thinking evolutionary theorists of his day and in bringing together their arguments (and adding a few of his own) Adaptation and Natural Selection severely dented Wynne-Edwards’ notion of group selection. Inclusive Fitness One of the foundation stones of Williams’ argument was laid by another evolutionary theorist of the mid 1960s, Bill Hamilton, who proposed a theory which revolutionised the very way that we look at the relationship between evolution and behaviour. Hamilton (1964a; 1964b), like Williams, held 48 Principles of Evolutionary Change the view that selection worked on individuals. Clearly individuals may help to increase the survival of their offspring by looking after them. You don’t need to be an evolutionary theorist to notice that parental care is common throughout the animal kingdom. Animals, in effect, invest time and effort in raising their offspring because having surviving offspring is what natural selection is all about (note that this suggests for many species individuals investing less time and effort in their offspring were less likely to pass their genes on). What Hamilton did was simply to expand this argument to consider the way that animals might treat other relatives. Since we share, on average, 50 per cent of our genes with each of our offspring, an act of heroism saving two children would save 100 per cent of our genes. However, we also share 25 per cent of our genes with each of our nephews, nieces and grandchildren and 12.5 per cent with each of our cousins and so on. Thus an act of heroism which saved, say, four grandchildren or eight cousins would also save on average 100 per cent of our genes. These great acts of heroism are extreme examples used for the sake of illustration. The main point is that an individual can also pass on copies of its genes indirectly by giving aid to relatives other than direct offspring. Hamilton called the proportion of genes shared between two family members the coeffi- cient of relatedness or ‘r’. This value may vary from 1 to 0. Identical twins share all of their genes and have an r of 1, siblings and offspring have an r of 0.5, while for grandchildren, nephews and nieces it is 0.25 and for cousins 0.125. Using this information it becomes clear that there are now two ways in which an animal can aid the transfer of copies of its genes into the next generation: directly via offspring or indirectly through giving aid to other relatives. If fitness had come to mean the number of surviving offspring an individual produced, Hamilton now added the number of other relatives that an individual helped to survive, each of which is weighted by its r, to the equation to come up with a value that he called inclusive fitness. This means that inclusive fitness is equal to the sum of direct and indirect fitness. Since the 1960s experts interested in the social behaviour of animals have used Hamilton’s notion of inclusive fitness to help explain some of the remarkable acts of apparent altruism within groups of closely related individuals. Apparent altruism directed at relatives has been called kin-selected altruism by Maynard Smith (1964). So for Hamilton, parental care is just an extreme and highly common form of kin-selected altruism. This highly influential argument will be explored further in Chapter 7. Box 2.6 Multilevel Selection Theory Although today most evolutionary psychologists subscribe to individual- or gene-level selection theory, the notion of group selection is by no means dead and buried. A number of evolutionists have attempted to revive a form of group selectionism under the banner of multilevel selection theory (MLST). Multilevel selection is particularly associated with American evolutionists El- liot Sober and David Sloan Wilson (see e.g. Wilson and Sober, 1994; Sober and Wilson, 1999). Multilevel selection is a complex theory that proposes that groups, in addition to individuals, might be considered to be ‘vehicles’ in the language of Richard Dawkins. Although Sober and Wilson have sought to revive the group as a unit of selection, they do not propose that this replaces either gene- or individual-level selection – but rather that natural selection can act at all three levels (hence ‘multilevel’). Drawing on both logical and empirical studies, Sober and Wilson argue that, when individual group members cooperate, the group can then outcompete other groups in terms of reproductive success. What sets this apart from Wynne-Edwards’ group  Levels of Selection – the Fittest What? 49 Box 2.6 (cont.) selectionism is the recognition that gene- and individual-level selection also occur. In this sense it appears to be a half-way house argument. This view is not without support from evolutionary psychologists and sociobiologists. E. O. Wilson, for example, has been favourable to multilevel selection theory (Wilson, 2005; 2012; 2015; Wilson and Holldobler, 2005). The gene-centred evolutionists such as Richard Dawkins (1994) and Daniel Dennett (1994) are yet to be convinced about multilevel selection, however. Dawkins in particular has argued that ultimately genes are the currency of natural selection – even though in the short term the group may appear to be the unit of selection. In his view what Sober and Wilson are advocating is not really a form of group selection. To Dawkins true group selection should see groups spawning other groups, which then compete with other groups still for survival. The winning group then passes on its ‘genes’ to the next generation. This is not real- ly what Sober and Wilson have in mind with their model of multilevel selection. To them MLST rejects inclusive fitness theory (IFT) as an explanation for altruistic behaviour. Unlike IFT which focuses on the relationship between natural selection and inclusive fitness, the MLS approach focuses on group traits and group fitness (Gardner, 2015). The problem here is that we can also think of group traits and group fitness as composites of individual traits and fitness. In summary, multilevel selection theory may have kept the argument about the level of selection on the table, but it has a long way to go if it is to convince mainstream evolutionary psychologists. Reciprocal Altruism By focusing on the individual’s relationship with family members Hamilton and Williams may have helped to explain away many examples of apparent altruism in the animal kingdom. But not all exam- ples of animals giving aid to others involve relatives. Among the primates there are well-documented cases of individuals aiding non-kin. An example of this is that unrelated vervet monkeys which form regular grooming pairs will come to each other’s aid in combat (Seyfarth and Cheney, 1984; see Figure 2.9). Until this and other examples of altruism could be explained, group selection still had a foothold in evolutionary theory. In the early 1970s a student of Hamilton by the name of Robert Trivers came up with an argument as to how non-relatives could engage in apparently altruistic acts without resorting to a simple group selection explanation. He called his idea reciprocal altruism. Reciprocal altruism might be likened to ‘you scratch my back and I’ll scratch yours’ but it is just a little more complicated than this. In Trivers’ model there are a few prerequisites. It is necessary for animals to live in quite stable groups, to be relatively long-lived and to be capable of spotting cheats. It is also important that the cost of the altruistic act is low compared to the benefit. To quote Trivers: Whenever the benefit of an altruistic act to the recipient is greater than the cost to the ac- tor, then as long as the help is reciprocated at some later date, both participants will gain. (Trivers, 1971) Trivers’ reciprocal altruism appears to work quite well for humans. Imagine, for example, that we are living on the plains of the Serengeti in Africa and that I have just killed a wildebeest. There is more meat than I could possibly eat before it either goes off in the heat of the African sun or is stolen by a clan of hungry hyenas. You may be starving but I can save your life at very little cost by giving 50 Principles of Evolutionary Change Figure 2.9 Vervet monkeys frequently form grooming pairs. you meat that is left over when I’ve had my fill. A week later our roles might easily be reversed and you may then save my life with your meat. In this way, if individuals regularly meet each other and enter into such a reciprocal arrangement then we may all gain. Of course, if there is net benefit to both parties from this reciprocation this raises the ques- tion can we really call it ‘altruism’? In order to avoid this philosophical question today reciprocal altruism is often referred to as direct reciprocity: ‘direct’ because the recipient of the ‘altruism’ later repays the initial actor. In recent years this concept has been extended to include indirect reciprocity, whereby third parties may provide benefits due to a person’s reputation for ‘altruistic’ behaviour (Colquhoun et al., 2020). In other words, developing a kind reputation can lead to benefits either to the individual with the reputation or to their kin. A special case of indirect reciprocity is known as costly signalling theory. Here an individual sends out signals of desirable characteristics that are too costly to fake (e.g. strength, health and kindness). We will return to reciprocity and costly signalling theory when we consider social behaviour in Chapter 8. Reciprocated acts of kindness are arguably one of the main features of human society, but isn’t this asking rather a lot of animals? Just how common reciprocation (both direct and indirect) is in the animal kingdom has become an area of great debate in recent years (Clutton-Brock, 2009; Colquhoun et al., 2020). Furthermore if aid is given when reciprocated aid is anticipated is this real- ly a form of altruism at all? We will further explore both kin altruism and reciprocal altruism when we turn our attention to social behaviour in Chapters 7 and 8. The Selfish Gene The reasoning that Williams, Hamilton and Trivers put forward to explain cooperation and altruism  The Selfish Gene 51 and even promoted behavioural responses which might benefit others because they share copies of our genes, then perhaps we shouldn’t be focusing on the individual as the entity that selection acts on at all. Perhaps we should be focusing on the gene itself. This is the conclusion that Richard Dawkins came to in the mid 1970s. His book The Selfish Gene (1976) drew on the ideas of Williams, Hamilton and Trivers but it also made an original contribution to evolutionary theory. Whereas pre- vious works had suggested that we should be focusing on genes if we want to explain behaviour and physical traits, Dawkins explicitly proposed that the unit of selection is the gene itself. In order to explain his thesis he introduced a number of new terms into the debate, in particular the replicator and the vehicle. Replicators are any entities which are able to make copies of themselves and ve- hicles are the entities which, on a geological timescale, briefly carry the replicators. In the context of life on earth we can think of replicators as genes and vehicles as organisms, including ourselves. But why the ‘selfish gene’? When we talk about people behaving selfishly we are using purposive and emotive language. In the context of his theory Dawkins uses the term selfish in a very specific way. Genes are considered selfish since alleles in the past which affected bodies to promote copies of themselves at the expense of others are the ones that are with us today. Genes that promot- ed copies of other alleles in the past were quickly removed from the population. A gene which is a particularly good replicator will leave many copies of itself and may continue for an indeterminately large number of generations. In this way, while the vehicle may be considered a transient survival machine, the gene which is most ‘selfish’ may in theory be immortal via copies of itself that it leaves. So selfish in this context merely means affecting the organism to make one’s own replication likely with no purposive state intended. But why did Dawkins reach such a radical conclusion? By the early 1970s both theoretical and observational work was making this conclusion highly likely to Dawkins. As far as theory was concerned, as Ridley has put it: Given that genes are the replicating currency of natural selection, it is an inevitable, al- gorithmic certainty that genes which cause behaviour that enhances the survival of such genes must thrive at the expense of genes that do not. (Ridley, 1996) As far as observational work was concerned, since Hamilton’s classic work of the 1960s it had become clear that many behaviour patterns make a great deal of sense from a gene-centred perspec- tive. Why, for example, should social insects lay down their lives for each other? Why should naked mole rats give up the right to breed to their ‘queen’ but brutally tear members of other colonies to pieces should they be unfortunate enough to wander into the wrong tunnel? And why should vam- pire bats regurgitate blood into the mouths of their starving colony mates? All of these and many other examples of extreme social behaviour could be explained only by recourse to the gene-focused theories of Williams, Hamilton and Trivers whereby copies of the genes for such responses ulti- mately benefit from the behaviour. From the gene-centred perspective the individual organism only makes sense as a transient survival machine (which is sometimes expendable). You may feel unhappy to be called a transient survival machine. If you do, then you are not alone. Dawkins’ notion of the selfish gene has not been universally accepted. In fact it has received criticism from a number of quarters. Much of the criticism has come from individuals who may not have a very clear understanding of how Dawkins uses the term ‘selfish gene’ (e.g. Midgley, 1979; Hayes, 1995); others have moral or political misgivings about this way of looking at life (e.g. Rose et al., 1984); still others have purely theoretical reservations (e.g. Daly and Wilson, 1983; Wilson, 2005). Dawkins dealt with many of his critics in a series of stylish rebuttals (Dawkins, 1979a; 1979b; 1989; 2006) and over the last 40 years the gene-centred view has had a major impact not only 52 Principles of Evolutionary Change 2018). Indeed, selfish gene theory is considered by some to be one of the foundation stones which led to the development of evolutionary psychology (Pinker, 1997; Laland and Brown, 2011; Work- man, 2014). It would be a mistake, however, to assume that all evolutionists interested in behaviour today are dyed-in-the-wool selfish-geners. The debate concerning the level at which natural selec- tion selects continues. As we saw in Box 2.6 on multilevel selection some experts today consider there may be a number of levels at which selection occurs. And even those who remain unconvinced by group level arguments prefer to focus on the individual. As Daly and Wilson (1983) have put it, ‘what we observe are individual organisms behaving, and it is their behavior that we wish to understand’. For most examples of behaviour, then, individual selection may be equated with gene selection since both benefit from the act. When it comes to considering altruistic behaviour, how- ever, as we will see in Chapter 7, genes may sometimes benefit even when there is a net cost to the individual. This chapter has been largely about the relationship between natural selection, genes and behaviour. In the next chapter we introduce a new and powerful driving force which may help to explore the relationship between these three areas further: sexual selection. Summary By the mid nineteenth century the notion that organisms change was in the air. Many scientists considered the notion of evolution seriously but the mechanism for this was lacking. In 1859 Charles Darwin introduced just such a mechanism – natural selection – to the scientific commu- nity and the public at large. Natural selection is based on heritable variation and differential repro- ductive success. Hence individuals with characteristics which allow them to survive and outbreed others pass on such characteristics to future generations. Mendel’s work on breeding in pea plants, once rediscovered in the twentieth century, led to a modern understanding of the particulate nature of genetic inheritance. Mendel demonstrated that genes act in pairs to determine an individual’s characteristics via a dominance–recessiveness rela- tionship. In Mendelian genetics, genes for different characteristics are passed on individually rath- er than being attached to each other. Mendel called this segregation. Features that occur together in the adult do not necessarily appear together in the offspring. This independent assortment is a result of segregation. Genes are found at specific locations on larger bodies called chromosomes that occur in pairs (one from each parent). The position of a gene is called its locus, and alternate genes that may occur at a given locus are called alleles. Individuals with identical genes at the same locus on each of a pair of chromosomes are said to be homozygous for that characteristic; conversely, where individuals have different genes at the same locus, they are said to be heterozygous. The complete set of an individual’s genes is known as its genotype, and the description of all of its characteristics (both physical and behavioural) is called its phenotype. During sexual reproduction, sex cells or gametes are produced which have half the number of chromosomes of normal cells due to a process called meiosis. Fusion of two gametes (a sperm and an ovum) leads to a fertilised egg which thereby has the normal number of chromosomes once  Summary 53 more. New body cells are formed which have the normal number of genes by a process called mitosis. Mendelian genetics has been modified by discoveries of the twentieth century. Genes that occur on the same chromosomes are often passed on together – this is called linkage. Due to linkage, assortment is not entirely independent. Mutations and recombination are important sources of variation of which Mendel was unaware. Additionally, many characteristics are polygenic, that is a number of genes are involved in determining the trait. Alternatively, individual genes may be pleiotropic, which means that they have more than one phenotypic effect. The expression of a gene may be altered by another gene, called a modifier gene. The heritability of a characteristic is an estimation of the extent to which it may be bred for and is a central concept to the field of behavioural genetics. The recent technique of genome-wide association is currently revolutionising human behavioural genetics. Epigenetics is the field of research that considers heritable changes which do not involve changes to the DNA sequence. The relatively new field of behavioural epigenetics considers how the ex- pression of genes might be influenced by lifetime experiences and how this, in turn, might affect behaviour. Although natural selection is the main mechanism responsible for adaptive change, other pro- cesses may also lead to change. Examples of this are gene flow – the movement of individuals with different genes into new areas – and genetic drift – random changes that build up over time. Physically, chromosomes consist of deoxyribonucleic acid (DNA) which contains four alternating nucleotides (adenine, thymine, cytosine and guanine – A, T, C and G). These nucleotides code in three-letter codons for amino acids – the building blocks of proteins. In this way we can think of genes as functioning to code for protein production. During the latter half of the twentieth century, evolutionists debated the level at which natural se- lection operates. Wynne-Edwards considered that individuals acted for the good of the group (i.e. group selection). Other experts, such as George Williams and Bill Hamilton, disagreed, arguing that individuals who act for the good of themselves and their relatives are more likely to pass their genes on to future generations (i.e. individual selection). Hamilton also suggested that individuals may pass on copies of their genes indirectly by giving aid to their relatives. The likelihood of them doing so may be related to the number of genes they share by common descent – their coefficient of relatedness (r). Maynard Smith called this kin selection. Kin selection has been used to explain the existence of self-sacrificing behaviour or altruism in ourselves and other species. Robert Triv- ers further suggested that examples of altruistic behaviour might be explained by a process of reciprocation – reciprocal altruism. The work of Williams, Hamilton and Trivers has led evolutionists to reconsider the level at which selection operates. In his book The Selfish Gene Richard Dawkins made explicit the notion of the gene as the unit of selection and introduced the concepts of the replicator and the vehicle. In the case of organic life the replicator is the gene and the vehicle the organism. Debates concerning individual versus gene selection continue. For most purposes, selection pressures which act on individuals will also act on genes directly. In the case of altruistic and selfish behaviour, however, this may not always be the case. 54 Principles of Evolutionary Change Questions 1. Old psychology and biology textbooks frequently suggested that animals’ behaviour has evolved to aid the ‘survival of the species’. Why is this notion problematic? 2. The notion of the selfish gene has been quite heavily criticised. Why do you think that people are unhappy with this term? You should be able to conceive of at least three different forms of criti- cism. 3. The ‘out-of-Africa’ hypothesis of human evolution suggests all modern human ethnic groups arose from a quite recent common ancestor. In contrast, the ‘multi-regional’ hypothesis suggests Homo erectus evolved into Homo sapiens separately in different geographical locations in our more distant past. If the multi-regional hypothesis were found to be correct what would this say about notions of a common human nature? Might it be argued that the multi-regional hypothesis is politically incorrect? 4. Robert Trivers’ notion of reciprocal altruism has had a profound effect on our understanding of why humans help each other. But is reciprocal altruism really altruism? Present an argument for why it should be considered as altruism and an argument taking issue with this. 5. The relatively new field of epigenetics suggests that some individual life experiences can affect the traits of their offspring. Might this finding suggest Darwin got it wrong? Further Reading Cartwright, J. (2016). Evolution and Human Behaviour: Darwinian Perspectives on the Human Condition. London: Macmillan/Palgrave. Comprehensive coverage of the relationship between evolution and human behaviour from an experienced behavioural biologist. Dawkins, R. (1976; 1989; 2006; 2016). The Selfish Gene. Oxford: Oxford University Press. A clearly argued account of the relationship between evolution and animal behaviour which also made an original contribution to evolutionary thinking by arguing that natural selection operates at the level of the gene. Humphrey, L. and Stringer, C. (2018). Our Human Story. London: Natural History Museum. Accessible account of hominin evolution from 7 million years ago up to present day. Laland, K. N. and Brown, G. R. (2011). Sense and Nonsense: Evolutionary Perspectives on Human Behaviour. Oxford: Oxford University Press. A critical review of the various approaches that have been developed in order to explore the relationship between human behaviour and evolution. Includes sociobiology, behavioural ecology and evolutionary psychology. Plomin, R. (2018). Blueprint: How DNA Makes Us Who We Are. London: Allen Lane/Penguin Books. Presents the history and recent developments in behavioural genetics to a general audience. In particular, considers the implications of genome-wide association studies and their impact on our understanding of the transmission of human traits. For those who would like to explore the technical detail further there is a lengthy notes section.