Natural selection

Natural selection is the process by which individual organisms with favorable traits are more likely to survive and reproduce than those with unfavorable traits. Natural selection works on the whole individual, but only the heritable component of a trait will be passed on to the offspring, with the result that favorable, heritable traits become more common in the next generation. Given enough time, this passive process can result in adaptations and speciation (see evolution).

Natural selection is one of the cornerstones of modern biology. The term was introduced by Charles Darwin in his 1859 book The Origin of Species , by analogy with artificial selection, by which a farmer selects his breeding stock.

An example: antibiotic resistance

File:Antibiotic resistance.gif

Figure 1: Schematic representation of how antibiotic resistance is enhanced by natural selection. The top section represents a population of bacteria before exposure to an antibiotic. The middle section shows the population directly after exposure, the phase in which selection took place. The last section shows the distribution of resistance in a new generation of bacteria. The legend indicates the resistance levels of individuals.

A well-known example of natural selection in action is the development of antibiotic resistance in microorganisms. Antibiotics have been used to fight bacterial diseases since the discovery of penicillin in 1928 by Alexander Fleming. However, the widespread use and especially misuse of antibiotics has led to increased microbial resistance against antibiotics, to the point that the methicillin-resistant Staphylococcus aureus (MRSA) has been described as a 'superbug' because of the threat it poses to health and its relative invulnerability to existing drugs.

Natural populations of bacteria contain, among their vast numbers of individual members, considerable variation in their genetic material, primarily as the result of mutations. When exposed to antibiotics, most bacteria die quickly, but some may have mutations that make them a little less susceptible. If the exposure to antibiotics is short, these individuals will survive the treatment. This selective elimination of maladapted individuals from a population is natural selection in action.

These surviving bacteria will then reproduce again, producing the next generation. Due to the elimination of the maladapted individuals in the past generation, this population contains more bacteria that have some resistance against the antibiotic. At the same time, new mutations occur, contributing new genetic variation to the existing genetic variation. Spontaneous mutations are very rare, very few have any effect at all, and usually any effect is deleterious. However, populations of bacteria are enormous, and so a few individuals will have beneficial mutations. If a new mutation reduces their susceptibility to an antibiotic, these individuals are more likely to survive when next confronted with that antibiotic. Given enough time, and repeated exposure to the antibiotic, a population of antibiotic-resistant bacteria will emerge.

Recently, several new strains of MRSA have emerged that are resistant to vancomycin and teicoplanin . This is an example of what is sometimes called an 'arms race', in which bacteria continue to develop strains that are less susceptible to antibiotics, while medical researchers continue to develop new antibiotics that can kill them. A similar situation occurs with pesticide resistance in plants and insects.

Background and context

Until the early 19th century, the established view was that differences between individuals of a species were uninteresting departures from their Platonic ideal (or typus) of created kinds. However, growing awareness of the fossil record led to the recognition that species that lived in the distant past were often very different from those that exist today. Naturalists of the time tried to reconcile this with the emerging ideas of uniformitarianism in geology - the notion that simple, weak forces, acting continuously over very long periods of time could have radical consequences, shaping the landscape as we know it today. Most importantly perhaps, these notions led to the awareness of the immensity of geological time, which makes it possible for slight causes to produce dramatic consequences. This opened the door to the notion that species might have arisen by descent with modification from ancestor species.

In the early years of the 19th century, radical evolutionists such as Jean Baptiste Lamarck had proposed that characteristics (adaptations) acquired by individuals might be inherited by their progeny, causing, in enough time, transmutation of species (see Lamarckism). Even today, many people believe that adaptations are actively acquired, often learned, through some intentional process. By contrast, Darwin postulated that adaptation is a passive process in which the selective culling by nature of maladapted individuals results in an increase of the fittest individuals (see survival of the fittest). In other words, Darwin suggested that adaptation may be an unintended effect, rather than an intended cause. He realised that this simple and apparently inevitable process might be powerful enough to explain the evolution of the astounding ways in which organisms are adapted to their environments and the origins of the millions of species that exist.

Evolutionary change can also happen without any selection, as a result of genetic drift or gene flow. However, adaptive change needs more than this, because it is very unlikely that favorable characteristics will consistently become more common in successive generations simply as a result of random fluctuations in occurrence. Favorable characteristics that can be attributed to genes that become more common through evolution by natural selection are called adaptations.

For Darwin, natural selection was synonymous with evolution by natural selection; other mechanisms of evolution such as 'evolution by genetic drift' were not explicitly formulated at that time, and Darwin realised that: "I am convinced that [it] has been the main, but not exclusive means of modification." Now, scientists use natural selection mainly to describe the mechanism. In this sense, natural selection includes any selection by a natural agent, including sexual selection and kin selection. Sometimes, sexual selection is distinguished from natural selection, but a more useful distinction is between sexual selection and ecological selection.

Definitions of natural selection

Scientists use several, slightly different definitions of natural selection. This section explains the different uses.

In each generation, only some individuals will produce offspring themselves, and of those that reproduce, some will leave more offspring than others. We can think of this as the "natural” process of selection of individuals to reproduce. Individuals with beneficial traits are more likely to be 'selected' than individuals with other traits. When those traits have a heritable component, they tend to become more common in the next generation. The mechanism of selection of individuals in a population does not "know" which traits are heritable; in this sense, the mechanisms of selection are "blind".

However, the term natural selection is often used to encompass the consequence of blind selection as well as the mechanisms to describe the complete process that leads to the enrichment of the beneficial characteristics in the next generation.

Nevertheless, it is sometimes helpful to distinguish clearly between the mechanisms of selection and the effects of selection. When this distinction is important, scientists define "natural selection" specifically as "those mechanisms that contribute to the selection of individuals that reproduce", without regard to whether the basis of the selection is heritable. This is sometimes referred to as 'phenotypic natural selection.'

This article discusses natural selection in this sense of being the mechanisms of selection of individuals to reproduce. Of particular importance is selection according to traits by which individuals differ from each other. It discusses the effects of this selection on the genetic characteristics of a population when some components of beneficial traits are heritable. Finally, the article ventures into the consequences of these effects for evolution.

Overview

Natural selection acts on the phenotype. The phenotype is the overall result of an individual's genetic make-up (genotype), the environment, and the interactions between genes and between genes and the environment. Often, natural selection acts on specific traits of an individual, and the terms phenotype and genotype are sometimes used narrowly to indicate these specific traits.

Some traits are determined by just a single gene, but most are affected by many different genes. Variation in most of these genes has only a small effect on the phenotypic value of a trait, and the study of the genetics of these quantitative traits is called quantitative genetics.

The key element in understanding natural selection is the concept of fitness. Natural selection acts on individuals, but its average effect on all individuals with a particular genotype is the fitness of that genotype. Fitness is measured as the proportion of progeny that survives, multiplied by the average fecundity, and it is equivalent to the reproductive success of a genotype. A fitness value of greater than one indicates that the frequency of that genotype in the population increases, while a value of less than one indicates that it decreases. The relative fitness of a genotype is estimated as the proportion of the fitness of a reference genotype. Related to relative fitness is the selection coefficient, which is the difference between the relative fitness of two genotypes. The larger the selection coefficient, the stronger natural selection will act against the genotype with the lowest fitness.

Natural selection can act on any phenotypic trait, and any aspect of the environment, including mates and conspecifics, can produce selective pressure. However, this does not imply that natural selection is always directional and results in adaptive evolution; natural selection often results in the maintenance of the status quo through purifying selection. The unit of selection is not limited to the level of individuals, but includes other levels within the hierarchy of biological organisation, such as genes, cells and relatives. There is still debate, however, about whether natural selection acts at the level of groups or species, (i.e. selection for adaptations that benefit the group or species, rather than the individual). Selection at a different level than the individual, for example the gene, can result in an increase in fitness for that gene, while at the same time reducing the fitness of the individuals carrying that gene (see intragenomic conflict for more details). Overall, the combined effect of all selection pressures at various levels determines the overall fitness of an individual, and hence the outcome of natural selection.

Figure 2: The life cycle of a sexually reproducing organism. Various components of natural selection are indicated for each life stage.

Natural selection occurs at every life stage of an individual (see Figure 2), and selection at any of these stages can affect the likelihood that an individual will survive and reproduce. After an individual is born, it has to survive until adulthood before it can reproduce, and selection of those that reach this stage is called viability selection. In many species, adults must compete with each other for mates (sexual selection), and success in this competition determines who will parent the next generation. When species reproduce more than once, a longer survival in the reproductive phase increases the number of offspring (survival selection). The fecundity of both females (e.g. how many eggs a female bird produces) and males (e.g. giant sperm in certain species of Drosophila) can be limited (fecundity selection). The viability of produced gametes can differ, while intragenomic conflict (meiotic drive) between the haploid gametes can result in gametic or genic selection. Finally, the union of some combinations of eggs and sperm might be more compatible that others (compatibility selection).

"Ecological selection" and "sexual selection"

It is also useful to make a mechanistic distinction between ecological selection and sexual selection. Ecological selection covers any mechanism of selection as a result of the environment (including relatives (e.g. kin selection) and conspecifics (e.g. competition, infanticide)), while sexual selection refers specifically to competition between conspecifics for mates . Sexual selection includes mechanisms such as mate choice and male-male competition although the two forms can act in combination in some species, when females choose the winners of the male-male competition. Mate choice, or intersexual selection, typically involves female choice, as it is usually the females who are most choosy, but in some sex-role reversed species it is the males that choose. Some features that are confined to one sex only of a particular species can be explained by selection exercised by the other sex in the choice of a mate, e.g. the extravagant plumage of some male birds. Aggression between members of the same sex (intrasexual selection) is typically referred to as male-male competition, and is sometimes associated with very distinctive features, such as the antlers of stags, which are used in combat with other stags. More generally, intrasexual selection is often associated with sexual dimorphism, including differences in body size between males and females of a species.

"Selection for" versus "selection of"

Selection targets specific traits of an individual, and if such a trait has a heritable component, the trait will tend to become more common in the next generation. "Selection for" refers to the traits targeted by selection, the causes of selection whereas "selection of" is used to describe its effects. Selection for a specific trait therefore results in the selection of certain individuals .

Selection for a specific trait can also result in the indirect selection of other traits ("free riders"). This can happen when two or more traits are genetically linked through mechanisms such as pleiotropy (single gene that affects multiple traits) and linkage disequilibrium (non-random association of two genes).

Genetical theory of natural selection

Natural selection by itself is a simple concept, in which fitness differences between phenotype play a crucial role. However, the interplay of the actual selection mechanism with the underlying genetics is where the explanatory power of natural selection comes from.

Directionality of selection

When some component of a trait is heritable, selection will alter the frequencies of the different alleles (variants of a gene) involved. Selection can be divided into three classes, on the basis of their effect on the allele frequencies.

Positive or directional selection occurs when a certain allele has a greater fitness than others, resulting in an increase in frequency of that allele until it is fixed and the entire population expresses the fitter phenotype.

Far more common is purifying or stabilizing selection, which lowers the frequency of alleles which have a deleterious effect on the phenotype (that is, lower fitness), until they are eliminated from the population. Purifying selection results in functional genetic features (e.g. protein-coding sequences or regulatory sequences) being conserved over time because of selective pressure against deleterious variants.

Finally, a number of forms of balancing selection exist, which do not result in fixation, but maintain an allele at intermediate frequencies in a population. This can occur in diploid species (with two pair of chromosomes) when individuals with a combination of two different alleles at a single position at the chomosome (heterozygote) have a higher fitness than individuals that have two of the same alleles (homozygote). This is called heterozygote advantage or overdominance. Maintenance of allelic variation can also occur through disruptive or diversifying selection, which favors genotypes that depart from the average in either direction (that is, the opposite of overdominance), and can result in a bimodal distribution of trait values. Finally, it can occur through frequency-dependent selection, where the fitness of one particular phenotype depends on the distribution of other phenotypes in the population (see also Game theory).

Selection and genetic variation

A portion of all genetic variation is functionally neutral; i.e., it produces no phenotypic effect or significant differences in fitness. Previously, this was thought to encompass most of the genetic variation in non-coding DNA, but recent studies have shown that large parts of those sequences are highly conserved and under strong purifying selection; i.e. they do not vary as much from individual to individual, indicating that mutations in these regions have deleterious consequences). When genetic variation does not result in differences in fitness, selection cannot directly affect the frequency of such variation. As a result, the genetic variation at those sites will be higher than at sites where selection does have a result.

Genetic linkage

Genetic linkage occurs when two alleles are in close proximity to each other. During the formation of the gametes, recombination of the genetic material results in reshuffling of the alleles. However, the chance that such a reshuffle occurs between two alleles depends on the distance between those alleles; the closer the alleles are to each other, the less likely it is that such a reshuffle will occur. Consequently, when selection targets one allele, this automatically results in selection of the other allele as well; through this mechanism, selection can have a strong influence on patterns of variation in the genome.

Mutation-selection balance

Natural selection results in the reduction of genetic variation through the elimination of maladapted individuals and, through that, of the mutations that causes the maladaptation. At the same time, new mutations occur, resulting in a mutation-selection balance. The exact outcome of the two processes depends both on the rate at which new mutations occurs and on the strength of the natural selection. Consequently, changes in the mutation rate or the selection pressure will result in a different mutation-selection balance.

Selective sweep

Selective sweeps occur when an allele becomes more common in a population as a result of positive selection. As the prevalence of one allele increases, linked alleles (those nearby on the chromosome) can also become more common, whether they are neutral or even slightly deleterious. This is called genetic hitchhiking. A strong selective sweep results in a region of the genome where the positively selected haplotype (the allele and its neighbours) are essentially the only ones that exist in the population.

Whether a selective sweep has occurred or not can be investigated by measuring linkage disequilibrium, i.e., whether a given haplotype is overrepresented in the population. Normally, genetic recombination results in a reshuffling of the different alleles within a haplotype, and none of the haplotypes will dominate the population. However, during a selective sweep, selection for a specific allele will also result in selection of neighbouring alleles. Therefore, the presence of strong linkage disequilibrium might indicate that there has been a 'recent' selective sweep, and this can be used to identify sites recently under selection.

Background selection

Background selection is the opposite of a selective sweep. If a specific site experiences strong and persistent purifying selection (perhaps as a result of mutation-selection balance), linked variation will tend to be weeded out along with it. Background selection, however, acts as a result of new mutations, which can occur randomly in any haplotype. It therefore produces no linkage disequilibrium, though it reduces the amount of variation in the region.

Evolution by means of natural selection

A prerequisite for natural selection to result in adaptive evolution, novel traits and speciation, is the presence of heritable genetic variation that results in fitness differences. Genetic variation is the result of mutations, recombinations and alterations in the karyotype (the number, shape, size and internal arrangement of the chromosomes). Any of these changes might have an effect that is highly advantageous or highly disadvantageous, but large effects are very rare. In the past, most changes in the genetic material were considered neutral or close to neutral because they occurred in noncoding DNA or resulted in a synonymous substitution. However, recent research suggests that many mutations in non-coding DNA do have slight deleterious effects. Overall, of those mutations that do affect the fitness of the individual, most are slightly deleterious, some reduce the fitness dramatically and some increase the fitness.

The exuberant tail of the peacock is thought to be the result of sexual selection by females. This peacock is an albino - it carries a mutation that makes it unable to produce melanin. Selection against albinos in nature is intense because they are easily spotted by predators or are unsuccessful in competition for mates, and so these mutations are usually rapidly eliminated by natural selection

By the definition of fitness, individuals with greater fitness are more likely to contribute offspring to the next generation, while individuals with lesser fitness are more likely to die early or they fail to reproduce. As a result, alleles which on average result in greater fitness become more abundant in the next generation, while alleles which generally reduce fitness become rarer. If the selection forces remain the same for many generations, beneficial alleles become more and more abundant, until they dominate the population, while alleles with a lesser fitness disappear. In every generation, new mutations and recombinations arise spontaneously, producing a new spectrum of phenotypes. Therefore, each new generation will be enriched by the increasing abundance of alleles that contribute to those traits that were favored by selection, enhancing these traits over successive generations.

X-ray of the left hand of a ten year old boy with polydactyly

Some mutations occur in so-called regulatory genes. Changes in these can have large effects on the phenotype of the individual because they regulate the function of many other genes. Most, but not all, mutations in regulatory genes result in non-viable zygotes. For example, mutations in some HOX genes in humans result in an increase in the number of fingers or toes or a cervical rib. When such mutations result in a higher fitness, natural selection will favor these phenotypes and the novel trait will spread in the population.

Established traits are not immutable: an established trait may lose its fitness if environmental conditions change. In these circumstances, in the absence of natural selection to preserve the trait, the trait will become more variable and will deteriorate over time. The power of natural selection will also inevitably depend upon prevailing environmental factors; in general, the number of offspring is (far) greater than the number of individuals that can survive to the next generation, and there will be intense selection of the best adapted individuals for the next generation.

Speciation

Speciation requires selective mating, which result in a reduced gene flow. Selective mating can be the result of, for example, a change in the physical environment (physical isolation by an extrinsic barrier), or by sexual selection resulting in assortative mating. Over time, these subgroups might diverge radically to become different species, either because of differences in selection pressures on the different subgroups, or because different mutations arise spontaneously in the different populations, or because of founder effects - some potentially beneficial alleles may, by chance, be present in only one or other of two subgroups when they first become separated. When the genetic changes result in increasing incompatibility between the genotypes of the two subgroups, gene flow between the groups will be reduced even more, and will stop altogether as soon as the mutations become fixed in the respective subgroups. As few as two mutations can result in speciation: if each mutation has a neutral or positive effect on fitness when they occur separately, but a negative effect when they occur together, then fixation of these genes in the respective subgroups will lead to two reproductively isolated populations. According to the biological species concept, these will be two different species.

History of the principle

The modern theory of natural selection derives from the work of Charles Darwin in the nineteenth century.

Between 1842 and 1844, Charles Darwin outlined his theory of evolution by natural selection as an explanation for adaptation and speciation. He defined natural selection as the "principle by which each slight variation [of a trait], if useful, is preserved". The concept was simple but powerful: individuals best adapted to their environments are more likely to survive and reproduce. As long as there is some variation between them, there will be an inevitable selection of individuals with the most advantageous variations. If the variations are inherited, then differential reproductive success will lead to a progressive evolution of particular populations of a species, and populations that evolve to be sufficiently different might eventually become different species.

Darwin thought of natural selection by analogy to how farmers select crops or livestock for breeding (artificial selection); in his early manuscripts he referred to a 'Nature' which would do the selection. In the next twenty years, he shared these theories with just a few friends, while gathering evidence and trying to address all possible objections. In 1858, Alfred Russel Wallace, a young naturalist, independently conceived the principle and described it in a letter to Darwin. Not wanting to be scooped, Darwin contacted scientific friends to find an honorable way to handle this potentially embarrassing situation, and two short papers by the two were read at the Linnean Society announcing co-discovery of the principle. The following year, Darwin published The Origin of Species, along with his evidence and detailed discussion. This became a topic of great dispute; evolutionary theories became the primary way of talking about speciation, but natural selection did not predominate as the mechanism by which it happened. What made natural selection controversial was doubt about whether it was powerful enough to result in speciation, and that it was 'unguided' rather than 'progressive', something that even Darwin's supporters balked at.

Darwin's ideas were inspired by the observations that he had made on the Voyage of the Beagle, and by the economic theories of Thomas Malthus, who noted that population (if unchecked) increases exponentially whereas the food supply grows only arithmetically; thus inevitable limitations of resources would have demographic implications, leading to a "struggle for existence", in which only the fittest would survive. However, similar ideas go back to ancient times; the Ionian physician Empedocles said that many races "must have been unable to beget and continue their kind. For in the case of every species that exists, either craft or courage or speed has from the beginning of its existence protected and preserved it". Several eighteenth-century thinkers wrote about similar theories, including Pierre Louis Moreau de Maupertuis in 1745, Lord Monboddo in his theories of species alteration, and Darwin's grandfather Erasmus Darwin in 1794–1796. In the 6th edition of The Origin of Species Darwin acknowledged that others — notably William Charles Wells in 1813, and Patrick Matthew in 1831 — had proposed similar theories, but had not presented them fully or in notable scientific publications. Wells presented his hypothesis to explain the origin of human races in person at the Royal Society, and Matthew published his as an appendix to his book on arboriculture. Edward Blyth had also proposed a method of natural selection as a mechanism of keeping species constant. However, these 'precursors' had little influence on evolutionary thought.

Within a decade of The Origin of Species, most educated people had begun to accept that evolution had occurred in some form or another. However, of the many ideas of evolution that emerged, only August Weismann's saw natural selection as the main evolutionary force. Even T.H. Huxley believed that there was more "purpose" in evolution than natural selection afforded, and neo-Lamarckism was also popular. After reading Darwin, Herbert Spencer introduced the term survival of the fittest; this became popular, and Wallace marked up an entire edition of The Origin of Species, replacing every instance of natural selection with Spencer's phrase. Although the phrase is still often used by non-biologists, modern biologists avoid it because it is tautological if fittest is read to mean functionally superior.

Modern evolutionary synthesis

Only after the integration of a theory of evolution with a complex statistical appreciation of Mendel's 're-discovered' laws of inheritance did natural selection become generally accepted by scientists. The work of Ronald Fisher (who first attempted to explain natural selection in terms of the underlying genetic processes), J.B.S. Haldane (who introduced the concept of the 'cost' of natural selection ), Sewall Wright (one of the founders of population genetics), Theodosius Dobzhansky (who established the idea that mutation, by creating genetic diversity, supplied the raw material for natural selection), William Hamilton (who conceived of kin selection), Ernst Mayr (who recognised the key importance of reproductive isolation for speciation) and many others formed the modern evolutionary synthesis. This propelled natural selection to the forefront of evolutionary theories, where it remains today.

Impact of the idea

Darwin's ideas, along with those of Adam Smith and Karl Marx, had a profound influence on 19th-century thought. Perhaps the most radical claim of the theory of evolution through natural selection is that "elaborately constructed forms, so different from each other, and dependent on each other in so complex a manner" evolved from the simplest forms of life by a few simple principles. This claim inspired some of Darwin's most ardent supporters—and provoked the most profound opposition. The radicalism of natural selection, according to Stephen Jay Gould , lay in its power to "dethrone some of the deepest and most traditional comforts of Western thought". In particular, it challenged beliefs in nature's benevolence, order, and good design, the belief that humans occupy a summit of power and excellence, belief in an omnipotent, benevolent creator, and belief that nature has any meaningful direction, or that humans fit into any sensible pattern.

Social Theory

The social implications of the theory of evolution by natural selection also became the source of continuing controversy. Engels in 1872 wrote that "Darwin did not know what a bitter satire he wrote on mankind when he showed that free competition, the struggle for existence, which the economists celebrate as the highest historical achievement, is the normal state of the animal kingdom".. That natural selection had apparently led to 'advancement' in intelligence and civilisation also became used as a justification for colonialism and policies of eugenics —see Social Darwinism. Konrad Lorenz won the Nobel Prize in 1973 for his analysis of animal behavior in terms of the role of natural selection (particularly group selection). However, in Germany in 1940, in writings that he subsequently disowned, he used the theory as a justification for policies of the Nazi state. He wrote "... selection for toughness, heroism, and social utility...must be accomplished by some human institution, if mankind, in default of selective factors, is not to be ruined by domestication-induced degeneracy. The racial idea as the basis of our state has already accomplished much in this respect." . Others have developed ideas that human societies and culture evolve by mechanisms that are analogous to those that apply to evolution of species . (see article on Sociocultural evolution).

Energetic theory

In 1922, Alfred Lotka proposed that natural selection might be understood as a physical principle which can be energetically quantified. Through the work of Howard T. Odum this became known as the maximum power principle whereby evolutionary systems with selective advantage maximise the rate of useful energy transformation.

Information theory

Natural selection need not apply only to biological organisms. In computer-based systems (e.g., artificial life), simulating natural selection can be very effective in 'adapting' entities to their environments . By combining this with simulated reproduction and random variation it is possible for instance to 'evolve' problem-solving abilities of computer-based systems. However, whether such systems show that evolution by means of natural selection per se can generate complexity is contested. The mathematician and science fiction writer Rudy Rucker explored the use of natural selection to create artificial intelligence in his best-known work, The Ware Tetralogy, and in his novel The Hacker and the Ants.

Trivia

• In a letter to Charles Lyell in September 1860, Darwin regrets the use of the term 'Natural Selection', preferring the term 'Natural Preservation'.

Further reading

• For technical audiences
  • Gould SJ (2002). The Structure of Evolutionary Theory. Harvard University Press. ISBN 0-674-00613-5. 
  • Maynard Smith J (1993) The Theory of Evolution Cambridge University Press
  • Popper K (1978) Natural selection and the emergence of mind. Dialectica 32:339-55. See [6]
  • Sargent TD, Millar CD, Lambert DM (1998) The "classical" explanation of industrial melanism: Assessing the evidence. Evolutionary Biology 30:299-322.
  • Williams GC (1966) Adaptation and Natural Selection: A Critique of Some Current Evolutionary Thought Oxford University Press.
  • Williams GC (1992) Natural Selection: Domains, Levels and Challenges Oxford University Press.

• For general audiences
  • Dawkins R, (1996) Climbing Mount Improbable Penguin Books, ISBN 0-670-85018-7
  • Dennett D (1995) Darwin's Dangerous Idea: Evolution and the Meanings of Life Simon & Schuster ISBN 0-684-82471-X
  • Gould SJ (1997) Ever Since Darwin: Reflections in Natural History Norton, ISBN 0-393-06425-5
  • Jones S (2001) Darwin's Ghost: The Origin of Species Updated Ballantine Books ISBN 0-345-42277-5. Also published in Britain under the title Almost like a whale: the origin of species updated Doubleday. ISBN 1-86230-025-9
  • Lewontin RC (1978) Adaptation. Scientific American 239:212-30
  • Weiner, J. (1994) The Beak of the Finch: A Story of Evolution in Our Time, Vintage Books, ISBN 0-679-73337-X

• Historical
  • Kohm M (2004) A Reason for Everything: Natural Selection and the English Imagination London: Faber and Faber. ISBN 0-571-22392-3. For review, see [7] van Wyhe J (2005) Human Nature Review 5:1-4

External links

• Evolution by Natural Selection - An introduction to the logic of the theory of natural selection
• Natural selection: how evolution works Interview with Douglas J. Futuyma
• The Origin of Species by Charles Darwin - Chapter 4,Natural Selection
• Natural Selection- Modeling for Understanding in Science Education, University of Wisconsin
• natural selection from University of Berkeley education website

See also

• Co-evolution
• Gene-centered view of evolution
• Genetic algorithm
• Negative selection
• Ring species
• Unit of selection