Science Wiki

Modern Synthesis (20th century) (also called the modern evolutionary synthesis, the neo-Darwinian synthesis, or simply the synthesis) was the unification, in the 1930s and 1940s, of Charles Darwin's theory of evolution by natural selection with the Mendelian genetics of discrete hereditary factors, using the mathematical models of population genetics. The term comes from Julian Huxley's book Evolution: The Modern Synthesis (1942). The synthesis resolved a dispute that had lasted since 1900 between the Mendelians, who studied discrete traits and emphasized large mutations, and the biometricians, who studied continuous variation and held that natural selection acts on small differences. It also joined the work of geneticists with that of naturalists, systematists, and paleontologists.

The synthesis established that evolution occurs in populations and is a change in the frequencies of alleles in the gene pool over generations, and that the processes of mutation, natural selection, genetic drift, and gene flow, together with genetic recombination, are sufficient to explain microevolution and, by extension, many patterns of macroevolution. It remains the core of evolutionary theory, although it has been extended, and some biologists argue that it needs a further revision.

Darwin and the problem of heredity

Darwin proposed natural selection in On the Origin of Species (1859), but he lacked a correct theory of heredity. He accepted a role for the inheritance of acquired characters and proposed pangenesis. Fleeming Jenkin argued in 1867 that under blending inheritance a new favorable variant would be diluted by crossing, so that its effect would be halved each generation. This objection was resolved by particulate inheritance, in which hereditary factors are not blended.

Mendelism and biometry

Gregor Mendel published his experiments on pea plants in 1866, and his work was rediscovered in 1900 by Hugo de Vries, Carl Correns, and Erich von Tschermak. William Bateson and other early Mendelians studied discrete variants and thought that new species arise by large mutations (saltation), so that selection has only a minor role. The biometricians, led by Karl Pearson and W. F. R. Weldon, measured continuous traits such as size and thought that evolution proceeds by selection on small, continuous variation. Each school thought that the other's view was incompatible with its own, and the dispute delayed the acceptance of natural selection. Early experimental work, such as that of Wilhelm Johannsen on pure lines of beans and the work of Thomas Hunt Morgan and colleagues on Drosophila, showed that mutations occur in the laboratory, that genes lie on chromosomes, and that small mutations exist. The results helped to join the two views.

Reconciling Mendelism and continuous variation

Ronald Fisher, J. B. S. Haldane, and Sewall Wright developed the mathematical theory in the 1910s to the 1930s. Fisher showed in 1918 that the correlations between relatives that were measured by the biometricians can be explained by Mendelian inheritance at many loci, each with a small effect. A continuous trait is therefore the result of polygenic inheritance, with the combined effects of many genes and of the environment. The total phenotypic variance V_P can be divided into genetic and environmental components:

where V_A is the additive genetic variance, V_D is the dominance variance, and V_I is the variance from epistasis. The narrow-sense heritability h² = V_A / V_P determines the response to selection through the breeder's equation:

where R is the change in the mean trait in the next generation and S is the selection differential.

The Hardy-Weinberg principle

The Hardy-Weinberg principle, formulated independently in 1908 by G. H. Hardy and Wilhelm Weinberg, is the null model of population genetics. For a locus with two alleles at frequencies p and q (with p + q = 1), random mating gives genotype frequencies of:

In the absence of mutation, selection, migration, and drift, and with random mating, allele frequencies do not change, and genotype frequencies reach these proportions after one generation. The principle shows that Mendelian inheritance by itself preserves genetic variation, which answered the argument that dominant traits should spread through a population and that variation would be lost. Deviations from Hardy-Weinberg proportions indicate that one of the assumptions is violated, but they do not show which one.

Selection

The population geneticists showed that selection on small differences is effective. For a single locus with relative fitnesses w_AA = 1, w_Aa = 1 − hs, and w_aa = 1 − s, the change in the frequency p of allele A in one generation is:

where w̄ is the mean fitness. A selection coefficient of a fraction of a percent changes allele frequencies over hundreds of generations, a short time on a geological scale. Haldane's 1924 to 1932 papers on "A Mathematical Theory of Natural and Artificial Selection" computed the rate of spread of an advantageous allele, for example in the evolution of industrial melanism in the peppered moth. Fisher's The Genetical Theory of Natural Selection (1930) argued that evolution is driven mainly by selection acting on abundant variation in large populations, and it included the fundamental theorem. A new beneficial mutation with a small advantage s is most likely to be lost by drift while it is rare. Its probability of fixation is about 2_s_ in a large population, a result obtained by Haldane.

Wright's shifting balance theory

Sewall Wright emphasized that populations are often subdivided into small, partly isolated groups. In his shifting balance theory, genetic drift in small demes can move a population across a "valley" of low fitness on the fitness landscape to the slope of a different peak, after which selection and migration spread the new combination. The role of drift and of such a process in evolution has been controversial. Fisher thought that large populations and selection were more important, and Wright thought that population structure and interactions among genes were important. The variance in allele frequency among subpopulations is measured by Wright's fixation index F_ST.

The Four Primary Evolutionary Processes

When a population does not meet the assumptions of the Hardy-Weinberg model, its allele frequencies change by one or more of the following processes:

1. Mutation. Mutation is the original source of new alleles. It is random with respect to the needs of the organism, in the sense that the chance of a mutation does not depend on whether it would be beneficial. This was shown by the Luria-Delbrück experiment (1943). Mutation is not random in all respects, because the rates differ among sites and types of changes. The rate in humans is about 10⁻⁸ per base pair per generation. Once a mutation arises, its fate is determined by selection and drift, and the process of selection is not random.
2. Natural selection. Selection is the difference in survival and reproduction among genotypes. It includes directional selection, which changes the mean of a trait, stabilizing selection, which reduces variance, disruptive selection, which increases it, and balancing selection, which maintains polymorphism, as in heterozygote advantage and negative frequency-dependent selection.
3. Genetic drift. Drift is the random change in allele frequencies caused by the sampling of gametes in a finite population. The variance of the change in one generation is _p_(1 − p) / (2_N__e), where N_e is the effective population size. Drift is strong in small populations, and it can fix neutral or slightly deleterious alleles, especially after a population bottleneck or a founder event. Selection is effective when |_s_| is greater than about 1/(2_N__e).
4. Gene flow. Migration of individuals or gametes between populations changes allele frequencies and makes populations more similar. In a simple continent-island model, a migration rate m changes the allele frequency of the island by Δ_p_ = −_m_(p − p_m), where p_m is the frequency in the source. Gene flow can prevent divergence and local adaptation, and its restriction is usually needed for speciation.

Genetic recombination and sexual reproduction shuffle alleles into new combinations, but they do not by themselves change allele frequencies. Non-random mating, such as inbreeding, changes genotype frequencies and not allele frequencies.

Dobzhansky and Mayr

Theodosius Dobzhansky's Genetics and the Origin of Species (1937) connected the mathematical models to the genetic variation that he and others found in natural populations of Drosophila, for example the chromosomal inversion polymorphisms of Drosophila pseudoobscura. It showed that wild populations have large amounts of genetic variation, that this variation is the material of evolution, and that differences between species are made of the same kinds of changes as differences within species. Ernst Mayr's Systematics and the Origin of Species (1942) did the same for systematics and the geographic variation of animals. Bernhard Rensch in Germany and G. Ledyard Stebbins in botany (Variation and Evolution in Plants, 1950) made similar arguments, and Stebbins included the role of polyploidy and hybridization in plants.

Species and speciation

Mayr defined the biological species concept: species are groups of actually or potentially interbreeding natural populations that are reproductively isolated from other such groups. Reproductive isolation results from prezygotic barriers (such as differences in habitat, timing, behavior, or gametes) and from postzygotic barriers (such as hybrid inviability or sterility). Dobzhansky and Hermann Joseph Muller explained how hybrid incompatibilities can evolve without a fitness valley, by negative interactions among genes at different loci, now called the Bateson-Dobzhansky-Muller model.

The synthesis gave particular weight to allopatric speciation, in which a population is divided by a geographic barrier, so that gene flow stops and the populations diverge by mutation, selection, and drift. Mayr also developed the idea of peripatric speciation, in which a small peripheral population diverges after a founder event. Later work showed that sympatric and parapatric speciation occur, as in polyploid plants and in some cases of ecological divergence, but allopatric speciation is still considered to be the most common mode in many groups of animals.

Paleontology and macroevolution

George Gaylord Simpson's Tempo and Mode in Evolution (1944) argued that the patterns of the fossil record, including the rates of evolution, adaptive radiations, and extinction, are consistent with the processes studied by population geneticists. He introduced the idea of adaptive zones and described how a lineage that enters a new zone can radiate rapidly. Simpson concluded that macroevolution does not require special mechanisms that are not found in microevolution. This view was challenged in 1972 by Niles Eldredge and Stephen Jay Gould, who proposed punctuated equilibrium, in which species show long periods of stasis that are interrupted by rapid change, concentrated around speciation. Whether macroevolution needs additional processes, such as species selection, is still debated. The contributions of zoologists and botanists were collected in the 1947 Princeton conference and in the journal Evolution, which was founded in the same year.

Population Thinking and Particulate Inheritance

The synthesis established several principles:

• Evolution is gradual in the sense that it is the result of small changes in allele frequencies, although the rate of change can differ. This does not exclude large effects of some mutations.
• Natural selection is the main cause of adaptation.
• Populations, and not individuals or types, evolve. This replaced typological thinking with "population thinking".
• Inheritance is particulate. Acquired characters are not inherited, and the genes are not changed by the use of the organism.
• Species are the units of the diversity of life, and they arise through the divergence of populations.

Not all biologists agreed about the relative importance of these factors. Some, such as Richard Goldschmidt, argued for the importance of large mutations, and in the 1960s and 1970s the neutral theory of molecular evolution of Motoo Kimura argued that most molecular variation is neutral, which suggested a larger role for drift than that of the early synthesis.

Molecular Genetics, Kin Selection, and Genomics

The synthesis has been extended by discoveries since the 1940s:

• Molecular genetics. The identification of DNA as the genetic material and the determination of its structure (1953) gave a physical basis for mutation and inheritance. Sequence data allowed the study of evolution at the molecular level, with molecular clocks, phylogenetic trees, and the neutral theory of molecular evolution.
• Kin selection and sociobiology. W. D. Hamilton's theory of inclusive fitness (1964) explained the evolution of altruism, and George C. Williams and others argued that selection acts mainly at the level of genes and individuals, as developed by Richard Dawkins in a gene-centered view of evolution.
• Evolutionary developmental biology. The discovery of conserved regulatory genes, such as the Hox genes, shows that large changes of form can result from changes in the regulation and timing of the expression of existing genes, and need not need many new genes.
• Horizontal gene transfer and symbiosis. Horizontal gene transfer among bacteria and archaea, and the origin of mitochondria and plastids by endosymbiosis, showed that inheritance is not only vertical in the history of life, and they complicate the idea of a tree of life in prokaryotes. This does not change the principles of selection and drift.
• Genomics. Genome sequencing has shown the importance of gene duplication, transposable elements, and genome rearrangements, and has allowed the study of selection by comparisons of genomes.

The extended evolutionary synthesis

Some biologists propose an "extended evolutionary synthesis" (EES) that gives more weight to developmental bias and constraint, phenotypic plasticity, niche construction, and non-genetic forms of inheritance, such as epigenetic and cultural inheritance. Proponents argue that these processes are causes of evolution and not only products of it. Others hold that these topics are already compatible with the existing framework, and that the standard theory is able to include them. The discussion concerns the relative importance of these processes and the way in which they should be modeled. It does not question the occurrence of evolution by natural selection.

Two points are often discussed:

• Epigenetic inheritance. Chemical modifications of DNA and histones, such as DNA methylation, can be inherited through cell divisions and, in some plants, fungi, and nematodes, through generations. Transgenerational epigenetic inheritance in mammals is less well established, and in many cases the marks are reset in each generation. Whether epigenetic variants make a significant contribution to long-term evolution is under study. The claim that the original synthesis "dictated" that acquired characters cannot be inherited is correct as a general statement of the rejection of Lamarckism, but it concerned the genetic material, and it is not contradicted by the existence of limited non-genetic effects.
• Plasticity and genetic assimilation. Plastic responses can allow a population to survive in a new environment and can influence the direction of later genetic change, for example by genetic assimilation or the Baldwin effect. The frequency and importance of these processes are debated.

How did Ronald Fisher's 1918 paper resolve the conflict between Mendelians and biometricians?

Early Mendelians argued that discrete genetic factors produced discontinuous jumps in phenotype, whereas biometricians observed continuous Gaussian distributions in traits like human height. In his landmark 1918 paper, Ronald Fisher demonstrated mathematically that if a quantitative trait is governed by many independent Mendelian loci—each contributing a small additive effect (V_A) combined with environmental variance (V_E)—the resulting phenotypic distribution converges to a continuous normal curve that obeys biometric parent-offspring regression laws.

How does the Bateson-Dobzhansky-Muller model explain the evolution of hybrid sterility without crossing a fitness valley?

If reproductive isolation required a single locus to mutate from AA to bb when the heterozygote Ab is sterile, neither population could evolve b without suffering reduced fitness. The Bateson-Dobzhansky-Muller model shows that an ancestral aabb population split by a geographic barrier can independently fix AAbb in population 1 and aaBB in population 2 via neutral or advantageous steps; negative epistatic incompatibility arises only when A and B meet for the first time in AaBb hybrids.

Why did Motoo Kimura's neutral theory of molecular evolution challenge the pan-selectionism of the early synthesis?

Whereas early synthesis theorists like Fisher held that almost all genetic polymorphisms and evolutionary substitutions were driven by natural selection, Motoo Kimura demonstrated in 1968 that the rate of amino acid and nucleotide substitution across diverging lineages was too rapid and clock-like to be sustained by selective deaths alone. Kimura showed that the majority of molecular substitutions are selectively neutral (|_s_| << 1 / 2_N__e) and reach fixation randomly via genetic drift at a rate equal to the underlying mutation rate μ.

• Evolution
• Population genetics
• Natural selection
• History of evolutionary thought
• Extended evolutionary synthesis
• Neutral theory of molecular evolution
• Evolutionary developmental biology