Friday, March 27, 2009

The Modern Evolutionary Synthesis

In the years following the publication of the Origin of Species in 1859, Charles Darwin’s theory of evolution became widely accepted throughout most of the scientific community. Other naturalists, including such notable figures as Charles Lyell, Joseph Hooker, Asa Grey, and especially Thomas Henry Huxley quickly came to accept Darwin’s assertion that what he called “descent with modification” had in fact occurred.

However, scientific opinion was much more divided on the subject of natural selection, Darwin’s proposed mechanism for evolution. To understand why, consider the three preconditions Darwin proposed as the necessary prerequisites for natural selection. They are:

• Variation: There must be significant differences between the members of an evolving population. These variations need not be extreme, as illustrated by the relatively large changes that animal and plant breeders have accomplished, using relatively slight differences in physical appearance and behavior.

• Inheritance: The distinct variations noted above must be heritable from parents to offspring.

• Fecundity: Living organisms have a tendency to produce more offspring than can possibly survive. Among those individuals that do survive, those that also reproduce pass on to their offspring whatever characteristics made it possible for them to survive and reproduce.

Given these prerequisites, then the natural outcome is:

• Non-Random, Unequal Survival and Reproduction: Survival and reproduction are almost never random. Individuals survive and successfully reproduce because of their characteristics. It is these demographic characteristics that form the basis for evolutionary adaptations.

Considering these four ideas, we can ask the question, “What is the ultimate source of the new characteristics that are preserved and promulgated from generation to generation?” The answer is, “The ultimate source of all new characteristics is the ‘engines of variation’ – that is, those processes that produce the natural variation between individuals that Darwin emphasized as being absolutely necessary for the operation of natural selection". In a nutshell:
Variation between individuals is the key to evolution by natural selection.

However, in the Origin, Darwin summarized his presentation of his views on variation with this statement:
"Our ignorance of the laws of variation is profound."

Neither Darwin nor any of his contemporaries (that he knew of) had a coherent theory of heredity or variation. However, this was not an insuperable obstacle to Darwin. Instead of giving up his argument, he simply accepted as a given that many important traits of animals and plants are heritable (pointing again to the observable facts of inheritance in domesticated animals and plants). He also proposed that, although he had no explanation of how they arose, variations among the members of a species do indeed occur, and can provide the raw material for natural selection.

There were therefore two reasons why Darwin’s proposed mechanism of natural selection was not widely accepted, even among scientists:

• Many of Darwin's contemporaries (and, in fact, Darwin himself) believed in Lamark's assertion that acquired characteristics could be inherited through use and disuse. This process directly contradicts the blind and purposeless process of natural selection, and therefore held the door open for purpose in evolution.

• The consensus among naturalists was that inheritance worked by "blending" the characteristics of parents, which would cause any incipient adaptations to be diluted out of existence.

This second objection to Darwin's mechanism of natural selection was almost fatal to his theory. In an influential review of the Origin, written in 1867 by Fleeming Jenkin (a very well-respected English engineer and designer of the first trans-Atlantic telegraph cable), Jenkin pointed out that blending inheritance would eliminate variation within a few generations:
“However slow the rate of variation might be, even though it were only one part in a thousand per twenty or two thousand generations, yet if it were constant or erratic we might believe that, in untold time, it would lead to untold distance; but if in every case we find that deviation from an average individual can be rapidly effected at first, and that the rate of deviation steadily diminishes till it reaches an almost imperceptible amount, then we are as much entitled to assume a limit to the possible deviation as we are to the progress of a cannon-ball from a knowledge of the law of diminution in its speed.”

If (as most naturalists of Darwin's time believed) all traits were blended from generation to generation, all of the distinctiveness of each variation would be lost and the population would remain essentially unchanged. Darwin got around this objection by proposing that large numbers of new variations (i.e. mutations) occur with each new generation. He called these “continuous variations,” but did not propose a mechanism for how they might be produced.

Mendelian Genetics

However, at about the same time that Darwin was working out his ideas on natural selection and evolution, Gregor Mendel was working out a revolutionary new theory of genetics. Mendel was born in 1822 in Moravia, a province of the Austrian Empire (now part of the Czech Republic). Because he was a peasant's son, Mendel was expected to return to the family farm after finishing his education. However, Mendel was not satisfied with all that he had learned. The university, instead of answering his questions, instilled in him an insatiable curiosity about nature.

Mendel observed that some offspring of some organisms had traits that were similar to only one parent, rather than being intermediate between both. He explained this phenomenon by assuming that heredity was determined by tiny, discrete “particles of inheritance” that were passed from the parents to the offspring via the reproductive cells. This would explain how some traits could remain unblended in the next generation.

Such thinking stemmed from Mendel's university education in physics. Ever since Isaac Newton revolutionized the science of physics, all of nature has been considered by physicists to be subject to "natural laws" based on the existence of and interactions between small, indestructible particles of matter. The goal of a physicist is to learn about the laws that determine the behavior of the particles, and to use such laws to predict the behavior of material objects subject to natural forces, such as gravity. Central to this intellectual tradition is the idea that an investigator can often work out these laws through careful observation and experimentation.

Mendel believed that these same methods could be used to study inheritance in living things. Having become established as a monk in an Augustinian monastery in the city of Brünn (now Brno, in the Czech Republic). Over a period of seven years he studied the inheritance of various characters in garden peas. In his landmark paper, "Experiments in Plant Hybridization” ("Versuche über Pflanzen-hybriden" in the original German, published in 1866), Mendel tells how he used the garden pea plant to study the laws of heredity.

Mendel's techniques differed from those of other investigators in three ways:

(1) He looked at one trait at a time;

(2) He followed this trait from generation to generation over eight years; and

(3) He used larger numbers of organisms in his studies. At the end of his experiments, he had carefully observed over 29,000 plants.

In his most famous set of experiments, Mendel studied 22 varieties of plants of the same species: the common garden pea (Pisum sativum). He studied a total of seven different traits, each with two alternative forms, including seed shape, color, and seed coat color; pod shape and color, flower position on the stem, and stem height.

For example, in one series of experiments, Mendel crossed pea plants that produced round seeds with pea plants that produced wrinkled seeds, and then observed what kinds of seeds were produced as the result of this cross over two generations.

Mendel observed that the two forms of each of these traits did not blend with each other. Among the offspring of the first cross, only one form of each trait showed up; the alternative form seemed to be lost. For example, when peas with round seeds were crossed with peas with wrinkled seeds, the first generation of offspring only produced round seeds (as shown in the Punnett square, above).

However, in the second generation, the seemingly lost form showed up again. In our previous example, wrinkled seeds showed up again in the second generation of offspring, comprising approximately one-fourth of all of the offspring of that cross. Mendel explained this result by saying that the lost form of each trait was actually latent or cancelled by the expressed form. He called the prevailing form of a trait dominant and the latent form of a trait recessive. Mendel's definitions of dominance and recessiveness are sometimes called Mendel's Law of Dominance:
Dominant traits mask the appearance of recessive traits whenever dominant and recessive traits are combined in one individual.

In our example, the gene for seed shape has two different forms. One form produces round seeds; the other form produces wrinkled seeds. Different gene forms that produce different forms of a trait are called alleles (from the Greek allos for "other"). In this example, the allele that codes for round seeds is dominant to the allele that codes for wrinkled seeds.

Mendel observed that dominant and recessive forms of a trait did not become blended. Instead, the recessive form of the trait reappeared in an unaltered form in the second generation. Based on this observation, Mendel formulated his Law of Segregation, which states that:
The different forms of a trait remain separate and unblended from generation to generation.

Mendel was so convinced of the validity of his conclusions that his subsequent work with other plants, some of which failed to support his hypothesis, did not discourage him. As he wrote in 1866,
"It requires indeed some courage to undertake a labour of such far-reaching extent; this appears, however, to be the only right way by which we can finally reach the solution of a question the importance of which cannot be overestimated in connection with the history of the evolution of organic forms."

Late in his life, Mendel's time was mostly spent fighting political battles for the monastery and peasants of his village. In his lifetime, Mendel witnessed a complete change in his homeland. In his later years, the focus was no longer on agricultural advances but on political advances. The rise of the Hapsburg dynasty and the consolidation of the Austro-Hungarian Empire forced different values on the people. The days of intellectual freedom, when a monk could study agriculture in a monastery garden without interference by the government, were drawing to a close. Shortly before his death in 1884, Mendel said to a future abbot of the monastery:
"Though I have suffered some bitter moments in my life, I must thankfully admit that most of it has been pleasant and good. My scientific work has brought me a great deal of satisfaction, and I am convinced that it will not be long before the whole world acknowledges it."

Evolution by Mutation

Mendel's belief that his work would eventually be recognized was not mistaken. In 1900, only fourteen years after his death, his work was simultaneously rediscovered by three different geneticists – Carl Correns, Erich Tschermak, and Hugo de Vries – working in three different countries. They each realized that Mendel's particulate theory of inheritance fit patterns of inheritance they were observing.

It is interesting to speculate what Darwin would have thought had he known about Mendel's work. Genes that did not blend in each generation were the answer to Darwin's dilemma, and could have put him onto the right track as early as 1866, the year Mendel's most important paper was published. A copy of the journal containing Mendel's paper was found in Darwin's library at Down House, but it had apparently not been opened or read.

There is an even deeper irony: the rediscovery of Mendel's work led geneticists to reject natural selection as the mechanism for evolution, in favor of mutations. Hugo de Vries, one of the rediscoverers of Mendel's work, proposed that mutations (i.e. changes in the phenotype of an organism, occurring in just one generation) were the primary "engine" of evolutionary change. De Vries did his pioneering work in genetics using the evening primrose (Oenothera lamarkiana), which is now known for having sudden, large mutations (called "macromutations") in its overall phenotype.

De Vries argued that these kinds of mutations were the basis for the changes in phenotype to which Darwin referred in the Origin of Species, and that therefore natural selection was neither necessary nor likely as a cause of evolutionary change. Indeed, DeVries asserted that macromutations were responsible for the "origin of species", and that natural selection played little or no role at all in this process. The mutational theory of evolution promoted by DeVries and other pioneering Mendelian geneticists was accepted by most of the prominent geneticists at the turn of the century, and led to widespread public testimonials that "Darwinism was dead":
“Today, at the dawn of the new century, nothing is more certain than that Darwinism has lost its prestige among men of science. It has seen its day and will soon be reckoned a thing of the past. A few decades hence when people will look back upon the history of the doctrine of Descent, they will confess that the years between 1860 and 1880 were in many respects a time of carnival; and the enthusiasm which at that time took possession of the devotees of natural science will appear to them as the excitement attending some mad revel.” - Eberhard Dennert, At the Deathbed of Darwinism (1904)

The Hardy-Weinberg-Castle Genetic Equilibrium Law

However, like Mark Twain, reports of Darwinism's death were "greatly exaggerated." In the second decade of the 20th century, three other researchers, again working separately and mostly unbeknownst to each other, proposed a theory that would eventually lead to the re-establishment of natural selection as the prime mover of evolution.

G. H. Hardy, Wilhelm Weinberg, and William Castle all proposed a mathematical theory that describes in detail the conditions that must be met for evolution to not occur. This theory, often called the Hardy-Weinberg Equilibrium Law lays out the conditions that must be met for there to be no changes in the allele frequency in a population of interbreeding organisms over time.

Recall Mendel's definition of alleles: different forms of the same gene that produce different variations of a trait. In the context of evolution, alleles are what code for the phenotypes that change over time in an evolving population. Therefore, changes in the alleles present in a population will produce changes in the phenotypes present in that population. This, in a nutshell, is the genetic definition of evolution:
Evolution is the result of changes in allele frequency in a population over time.

What Hardy, Weinberg, and Castle all realized is that for allele frequencies to not change in a population, five conditions must be met:

There must be no mutations (i.e. alleles cannot change into other, different alleles).

There must be no gene flow (i.e. individuals cannot enter or leave the population).

The population must be very large (i.e. random accidents cannot significantly alter allele frequences).

Survival must be random (i.e. there can be no natural selection).

Reproduction must also be random (i.e. there can be no sexual selection).

Notice that the Hardy-Weinberg Equilibrium Law seems to say only that there are conditions under which evolution can't happen. Aren't we interested in those conditions in which evolution can happen? Yes, but notice what the Hardy-Weinberg Equilibrium Law gives us: it outlines exactly what processes are essential to prevent evolution, and therefore by negation shows us how evolution can happen.

That is, if any of the five conditions for maintaining a Hardy-Weinberg equilibrium are not met, then evolution must be occurring. And, of course, virtually none of these conditions is ever permanently met in any known natural population of organisms:

• Mutations occur at a slow but steady rate in all known populations.

• Many organisms, especially animals, enter (immigration) and leave (emigration) populations.

• Most populations are not large enough to be unaffected by random changes in allele frequencies.

• Survival is virtually never random.

• Reproduction in organisms that can choose their mates is also virtually never random.

Therefore, according to the Hardy-Weinberg Equilibrium Law, evolution (defined as changes in allele frequencies over time) must be occurring in virtually every population of living organisms. In other words,
Evolution is as ubiquitous and inescapable as gravity.


The Hardy-Weinberg Equilibrium Law provided more than just a "null hypothesis" for genetic evolution. It also provided a mathematical basis for a much more comprehensive theory of evolution now known as the modern evolutionary synthesis (often called "neo-darwinism"). During the 1930s and 40s, R. A. Fisher, J. B. S. Haldane, Sewall Wright, and Theodosius Dobzhansky developed mathematical models for fitness, selection, and other evolutionary processes. These models were then applied to demographic data derived from artificial and natural populations of organisms in a rigorous (and ongoing) series of empirical tests of the validity of the neo-darwinian model for genetic evolution. As a result of their work, Darwin's theories of natural and sexual selection were combined with Mendelian genetics, biometry and statistics, demography, paleontology, comparative anatomy, botany, and (more recently) molecular genetics and ethology to produce a "grand unified theory" of the origin and evolution of life on Earth.

The Genetical Theory of Natural Selection

Ronald Aylmer Fisher built on the pioneering theoretical work of Hardy, Weinberg, and Castle by providing mathematical models that further undermined the Mendelian geneticist's theory of evolution via mutation. He did this by showing that continuous variation could provide the basis for natural selection as proposed by Darwin. In his most important work, The Genetical Theory of Natural Selection (published in 1930) Fisher showed that traits characterized by continuous variation (i.e. those that approximate a normal, or bell-shaped, distribution) were both common and could provide all the raw material necessary for Darwinian natural selection. This is because such traits, although being continuous in populations, do not blend from parents to offspring. Instead, as Mendel first showed, they are produced by unblending "particles" of inheritance (i.e. Mendelian "genes"). In other words,
Mendelian inheritance conserves, rather than eventually destroying, the genetic variation that exists in natural populations.

Fisher is perhaps best known for what he called the Fundamental Theorem of Natural Selection. Using a series of essentially mathematical arguments, Fisher showed that the rate of change via natural selection was a direct function of the amount of variation in a population. That is,
The more variation among alleles that exists in a population, the faster natural selection can causes changes in the allele frequencies in that population.

Conversely, the less variation among alleles that exists in a population, the slower natural selection can causes changes in the allele frequencies in that population.

R. A. Fisher's work formed the basis for a mathematical theory of evolution in which the process of natural selection is modeled mathematically in the same way that Newton modeled the force of gravity. Indeed, Fisher pointed out several times that the mathematics of natural selection were similar in many ways to such physical models as the ideal gas laws and the second law of thermodynamics. According to his mathematical models, alleles that were positively selected would increase in frequency in populations in much the same was as gas molecules spread out in an expanding balloon.

To many evolutionary biologists, this meant that natural selection would inevitably result in "fixation" of alleles that were not selected against. That is,
Any allele that results in increased survival and reproduction should, if given enough time, eventually become the only allele for that particular trait in a particular population.

This presented a problem to evolutionary biologists that was almost as severe as the “mutationism” of the early Mendelians. It implied that the inevitable result of natural selection would be the eventual elimination of all non-adaptive variation in natural populations. This would then cause natural selection to grind to a halt (or to become reduced to essentially the rate of production of new genetic mutations, which is slow in the extreme, much slower than the observed rate of evolution). Fisher suggested that constant environmental change would cause different alleles to be selected for and against, and that therefore fixation might not ever happen. However, this argument seemed to be "tacked on" to his argument for the relationship between the amount of variation in populations and the speed of evolutionary change via natural selection.

Adaptive Landscapes and Genetic Drift

A solution to this problem was provided by Sewall Wright, who discovered a process that has eventually become known as genetic drift. Wright, who worked primarily with domesticated animals in controlled breeding programs, proposed that in small populations of organisms, random sampling errors could cause significant changes in allele frequencies in those populations. He showed mathematically that the smaller a population was, the greater the effect of such random events on its allele frequencies. In other words,
Evolution can proceed by at least three primary mechanisms: natural selection, sexual selection, and random genetic drift.

Wright's discovery of genetic drift solved the problem that Fisher's Fundamental Theorem posed: how can natural selection be prevented from shutting itself down as the result of fixation? Wright proposed that allele frequencies could be visualized as forming what he came to call an "adaptive landscape". In an adaptive landscape, allele frequencies formed a series of hills and valleys, in which the top of a hill represented the highest an allele frequency could reach via natural selection. According to Fisher, there is an iron-clad rule operating here: if an allele is on a slope, it can only go up the slope via natural selection.

But this means "you can't get there from here": if a trait is at the top of one adaptive peak, it can't go down through a valley to get to the top of another, even higher (i.e. more adaptive) peak. What Wright showed was that "you can get there from here" if you drift there. That is, if a population becomes very small, it is possible for it to "drift" from one adaptive peak to another, without sliding down into the valley in between. This means that natural selection doesn't get "stuck"; populations at one adaptive peak can make it to another, even higher adaptive peak, so long as they drift randomly to it.

The Causes of Evolution

John Burdon Sanderson Haldane (usually referred to as J. B. S. Haldane) solidified the revolution in theoretical population genetics begun by Hardy, Weinberg, Castle, Fisher, and Wright. In his most important book, The Causes of Evolution, published in 1932, he showed that genetic mutations could provide the raw material for Darwinian natural selection. Furthermore, he showed mathematically that such mutations could do this even when their frequency in a population was initially so low that they would be "invisible" to statistical analysis. He also showed how dominance could evolve in populations by means of natural selection, even when the original expression of an allele was initially recessive.

Haldane is also remembered for two quips that are often repeated by evolutionary biologists. The first concerns a question posed to him by an Anglican minister, who asked him (supposedly at a dinner party) what his study of nature had led him to conclude about the principle concern of the Creator. Without batting an eyelash, Haldane replied: "An inordinate fondness for beetles," referring to the fact that there are more species of beetles on Earth than any other kind of organism.

During another conversation (supposedly in a pub), Haldane was confronted with the observation that natural selection should result in pure selfishness on the part of individuals, and therefore no one should be willing to risk his own life to save another. To this Haldane replied,
"I would be willing to risk my life to save two brothers or eight cousins."

This quip is based upon the observation that brothers share an average of one-half of their genetic material, whereas first cousins share an average of one-eighth. Therefore, saving two brothers or four cousins would result in the same genetic contribution to the next generation as that represented by one's own genome. This quip was later cited by one of the founders of what is now know as the theory of kin selection in which natural selection is considered to act at the level of genes, rather than individuals. I will discuss this idea in more detail in a later blog post.

R. A. Fisher, J. B. S. Haldane, and Sewall Wright are usually recognized as having laid the theoretical foundation for modern evolutionary theory. However, many evolutionary biologists and historians of science consider that the "modern evolutionary synthesis” was initiated by Theodosius Dobzhansky with the publication of his most famous book, Genetics and the Origin of Species published in 1937.

Genetics and the Origin of Species

Dobzhansky combined the Mendelian genetics, the mathematical models of Fisher, Haldane, and Wright, and the observations of evolution and natural selection in the wild in a theory that reinstated natural selection as the primary engine of evolution. He emphasized both the scientific aspects of evolutionary theory, and the implications of evolutionary theory for education and society in general. In a famous essay entitled "Nothing in biology makes sense except in the light of evolution” he showed how modern synthetic evolutionary theory provides a comprehensive explanation for the origin and evolution of life on Earth.

Dobzhansky also grounded the "modern evolutionary synthesis" in empirical investigation. Using the common fruit fly (Drosophila melanogaster). Dobzhansky and his colleagues showed that the patterns of variation and natural selection predicted by Fisher actually occurred in controlled populations of living organisms under laboratory conditions.

Most importantly, Dobzhansky showed empirically that the "continuous variation" that both Darwin and Fisher asserted were essential for natural selection actually occurred for many traits in nature. According to Dobzhansky, most traits are distributed in what is often referred to as a "bell-shaped curve". That is, for most traits there is an average value for the trait, which the majority of the members of the population share. There is also two "tails" to the bell-shaped curve, consisting of extreme versions of the trait.

Dobzhansky then went on to identify three different forms of natural selection, which depended upon which part of the bell-shaped curve of variation selection affected:

• Directional selection, in which selection against one extreme "tail" of the bell-shaped curve caused the average value for the trait to move over time;

• Stabilizing selection, in which selection against both extreme "tails" of the bell-shaped curve caused the average value for the trait to remain where it was; and

• Disruptive selection, in which selection against the average value of the bell-shaped curve caused the population to split into two diverging curves, corresponding to the two extreme versions of the trait.

The proponents of the "modern evolutionary synthesis" asserted that this last form of natural selection was the underlying explanation for the divergence of one species into two or more different species (for this reason, disruptive selection is sometimes referred to as "diversifying selection"). That is, Darwin's "mystery of mysteries" – the origin of species – was shown to have a mathematical basis which could be studied empirically and tested statistically, thereby making it a genuinely "scientific" study.

The Historical Importance of the "Modern Evolutionary Synthesis"

What, then, was the importance of the “modern evolutionary synthesis” to evolutionary theory? Perhaps J.B.S. Haldane said it best:
"The permeation of biology by mathematics is only beginning, but unless the history of science is an inadequate guide, it will continue, and the investigations here summarized represent the beginning of a new branch of applied mathematics."

The theory of evolution as Darwin first proposed it was essentially a qualitative theory; it had no mathematical basis, and could not be tested using statistical methods. Indeed, Darwin himself was a “mathophobe,” who had neither the training nor the inclination to provide a mathematical basis for his theories.

However, the founders of the modern synthesis were all well-versed in mathematics, as was Gregor Mendel. Indeed, R. A. Fisher not only provided the first solid mathematical framework for the theory of evolution by natural selection, he virtually founded the disciplines of biometry and statistics. Many of the statistical tests that are still used to test evolutionary hypotheses (indeed, hypotheses throughout the natural and social sciences) were first formulated by Fisher.

Providing a mathematical foundation for evolutionary theory literally meant converting evolution from “natural history” into a modern science. When a hypothesis can be tested by gathering numerical data (by counting or measuring objects and events), that data can then be statistically tested to determine if it verifies or falsifies that hypothesis. This is what happens in the other natural sciences, like chemistry and physics. Since the modern evolutionary synthesis, this is also what happens in evolutionary biology.


Mayr, Ernst & William Provine (eds.) (1998) The Evolutionary Synthesis: Perspectives on the Unification of Biology. Harvard University Press.


Bowler, Peter J. (1983) The Eclipse of Darwinism: Anti-Darwinian Evolutionary Theories in the Decades Around 1900. The Johns Hopkins University Press.

Darwin, Charles (1868) The Variation of Animals and Plants Under Domestication. John Murray. Available online here.

Dobzhansky, Theodosius (1937) Genetics and the Origin of Species. Columbia University Press.

Dobzhansky, Theodosius (1973) Nothing in biology makes sense except in the light of evolution. The American Biology Teacher, March 1973, volume 35, pages 125-129. Available online here.

Fisher, R. A. (1930) The Genetical Theory of Natural Selection. Oxford University Press.

Haldane, J. B. S. (1932) The Causes of Evolution. Princeton University Press.

Jenkin, Fleeming (1867). [Review of] The Origin of Species. The North British Review, June 1867, 46, pp. 277-318. Available online here.

Mendel, Gregor (1866) Experiments in Plant Hybridization. Verhandlungen des naturforschenden Vereines in Brünn, volume 4, pages 3-47. Available (in English) online here.

Provine, W. (1971) The Origins of Theoretical Population Genetics. University of Chicago Press.


As always, comments, criticisms, and suggestions are warmly welcomed!


Labels: , , , , , , ,


At 11/13/2009 08:03:00 AM, Blogger Unknown said...

Quite interesting article, tackled quite a lot in one fell swoop. I'm digesting it still. :)

I found this while looking for mathematical approaches to investigating evolution, perhaps even abiogenesis. Came upon the works of J.B.S Haldane and ended up here.


Post a Comment

<< Home