What Is the Modern Synthesis of Evolution?

The Modern Synthesis is the theoretical framework, assembled between roughly 1918 and the 1940s, that unified Charles Darwin’s idea of natural selection with Gregor Mendel’s laws of heredity. Before the synthesis, these two pillars of biology seemed to many researchers to be in conflict. The synthesis resolved that conflict and established the core logic that still organizes evolutionary biology today: populations evolve when the frequencies of inherited gene variants shift over generations, driven by natural selection, mutation, genetic drift, and gene flow. Yet in the decades since, discoveries in molecular biology, developmental genetics, and genomics have raised persistent questions about whether the synthesis needs expanding or rethinking.

The Problem the Synthesis Had to Solve

At the turn of the twentieth century, evolutionary biology was surprisingly fractured. Darwin’s mechanism of natural selection was under sustained attack, and the rediscovery of Mendel’s laws around 1900 initially made things worse rather than better.¹ The problem was that Mendelian genetics described traits controlled by discrete units of inheritance, while natural selection seemed to require the smooth, continuous variation you see in traits like height or body size. Mendelians and “biometricians” (the statistical camp that tracked continuous variation in populations) talked past each other for years. Meanwhile, mutation theorists like Hugo de Vries argued that evolution proceeded in large, sudden leaps, making gradual natural selection unnecessary. The field needed a unifying framework, and it took a generation of mathematical and empirical work to produce one.

How the Synthesis Came Together

The foundation was mathematical. In 1918, R.A. Fisher showed that continuous variation in traits like height could be explained by Mendelian inheritance if you assumed that many genes, each with a small effect, contributed to the trait.² This was a breakthrough because it dissolved the apparent contradiction: Mendel’s discrete genetics could produce the smooth gradients Darwin’s theory required. Fisher, along with J.B.S. Haldane and Sewall Wright, built the mathematical discipline of population genetics through the 1920s and 1930s. Their equations showed how natural selection, random drift, mutation, and migration interact to change gene frequencies in populations.

The mathematics alone would have remained an abstraction without empirical reinforcement. Theodosius Dobzhansky brought the theory to life with studies of wild fruit fly populations, showing that natural populations harbored immense genetic variation and that selection acted on it in real time. Ernst Mayr contributed the concept of geographic speciation, arguing that new species form when populations become physically isolated and diverge genetically. Mayr also championed the biological species concept, defining species as groups of populations that can interbreed and are reproductively isolated from other such groups.³ George Gaylord Simpson extended the synthesis to paleontology, arguing that the same microevolutionary forces visible in living populations could explain the large-scale patterns seen in the fossil record. Julian Huxley’s 1942 book, Evolution: The Modern Synthesis, gave the framework its name.

What the Synthesis Actually Claims

The Modern Synthesis rests on a few interlocking ideas. Populations, not individuals, are the units that evolve. Evolution is a change in the frequency of gene variants within a population over time. Natural selection is the primary mechanism driving adaptive evolution, acting on heritable variation produced by random mutation. Speciation typically occurs through the gradual geographic isolation of populations. And the processes visible in present-day populations (microevolution) are sufficient, given enough time, to account for the large-scale patterns of life’s history (macroevolution). As Fisher, Haldane, and Wright established, Mendelian inheritance is exactly what natural selection needs to work efficiently, because discrete genes maintain variation rather than blending it away.⁴

A subtle but important point often gets lost: the synthesis did not claim that natural selection explains everything. Genetic drift, the random fluctuation of gene frequencies especially in small populations, was built into the framework from the beginning, largely through Wright’s work. The synthesis acknowledged drift but treated adaptive natural selection as the dominant creative force. This emphasis on selection as the primary engine would later draw significant challenges.

Mayr’s Proximate-Ultimate Distinction

One of the less visible but deeply influential contributions to come out of the synthesis era was Ernst Mayr’s 1961 distinction between proximate and ultimate causes in biology. Proximate causes are the immediate “how” mechanisms: how does a bird navigate? Ultimate causes are the evolutionary “why” explanations: why did navigation evolve? Mayr argued these were complementary, not competing, and that confusing the two led to muddled thinking.⁵

This framework became a standard organizing principle in biology for decades. But it also had consequences that are still debated. Critics argue that the clean separation between proximate and ultimate causes encouraged biologists to treat development as a “proximate” black box, something that merely executes the instructions of genes shaped by selection, rather than as a process that actively shapes evolutionary outcomes. Several ongoing debates in biology, including arguments about the role of development in evolution and the importance of niche construction, trace back to disagreements about whether Mayr’s dichotomy helps or hinders progress.⁶

The Neutral Theory Challenge

The first major conceptual challenge to the synthesis arrived in the late 1960s from an unexpected direction: the molecules themselves. As protein sequencing became possible, Motoo Kimura proposed that the vast majority of evolutionary changes at the molecular level are not driven by natural selection at all. Instead, most molecular changes are selectively neutral, meaning they make no difference to the organism’s survival or reproduction, and they spread through populations by random drift.⁷

This was a direct challenge to the synthesis’s emphasis on selection as the dominant evolutionary force. If most DNA-level evolution is selectively invisible, then natural selection governs what organisms look like and how they behave, but drift governs much of what their genomes actually contain. The neutral theory was fiercely contested by the mainstream at first, but supporting evidence accumulated steadily.⁸ Kimura’s framework drew on Wright’s tradition of emphasizing finite population size and random drift as major evolutionary factors.⁹

Today, virtually all molecular biologists accept that neutral evolution plays a major role at the DNA level, even as debate continues about exactly how much. The practical result is that the synthesis’s portrait of evolution as a process dominated by natural selection turned out to be a strong description of visible traits, but an incomplete one for what goes on inside the genome.

Punctuated Equilibrium and the Tempo of Evolution

In 1972, paleontologists Stephen Jay Gould and Niles Eldredge proposed punctuated equilibrium, another challenge that came from the synthesis’s own turf. The standard synthesis view assumed that evolution is typically gradual: populations slowly accumulate changes, and given enough time, these small shifts produce new species and major new forms. But the fossil record kept stubbornly refusing to show these smooth transitions. Most species appear in the record looking pretty much the same throughout their existence, then disappear and are replaced by related but distinct forms.

Gould and Eldredge argued that this pattern is real, not an artifact of gaps in the fossil record. They proposed that most evolutionary change is concentrated in rapid speciation events that are geologically instantaneous, even if they take thousands of years in real time. Between these bursts, species remain stable. Large-scale evolutionary trends, they argued, reflect the differential success of some species lineages over others, not the slow, directional transformation of a single lineage.¹⁰

Punctuated equilibrium generated fierce argument throughout the 1980s and 1990s. Today, most evolutionary biologists accept that both gradual change and punctuational patterns exist, and the interesting question is how often and under what circumstances each occurs. The idea’s lasting contribution was to force the field to take the fossil record’s tempo patterns seriously rather than explaining them away.

Evo-Devo and the Genetic Toolkit

Perhaps the most surprising post-synthesis discovery is that animals as different as fruit flies, mice, and humans share a remarkably similar set of genes controlling body plan development. This “genetic toolkit” of regulatory genes was in place before the Cambrian explosion that produced most modern animal body plans.¹¹ What changes between species is often not the toolkit genes themselves, but when, where, and how much those genes are expressed during development.

This insight is the foundation of evolutionary developmental biology, or “evo-devo.” Rather than evolving new genes for every new body form, animals tend to rewire the regulatory networks that control when existing genes turn on and off. Changes to the switches and circuits that control gene expression can produce dramatic differences in morphology without requiring a flood of new mutations to structural genes.¹²

Evo-devo raised uncomfortable questions for the Modern Synthesis. The synthesis had treated development as largely irrelevant to evolutionary theory, a topic for embryologists rather than population geneticists. But if the regulatory architecture of development channels the kinds of variation that can appear, then development is not just executing genetic instructions but actively shaping which evolutionary paths are accessible. Some regulatory circuits are ancient and deeply conserved, making them resistant to change, while others are flexible and permit rapid evolutionary modification. This mix of constraint and flexibility was not something the synthesis’s gene-frequency mathematics was designed to capture.

Epigenetic Inheritance

One of the synthesis’s core assumptions is that the only information reliably passed from parent to offspring is DNA sequence. Acquired characteristics, the idea associated with Lamarck, were firmly excluded. But recent research has complicated this picture. Chemical modifications to DNA and its associated proteins can alter gene expression without changing the DNA sequence itself, and in some cases, these modifications can be transmitted across generations.

This transgenerational epigenetic inheritance has been documented in plants, nematodes, and some mammals, involving mechanisms like DNA methylation, histone modifications, and non-coding RNAs.¹³ Environmental conditions can drive these epigenetic changes, meaning the environment can directly influence what gets passed to the next generation.¹⁴

The evolutionary significance of epigenetic inheritance remains hotly debated. In plants, the evidence is relatively strong. In mammals, it is less clear how widespread or stable these effects are, or how much they actually matter compared to conventional genetic inheritance.¹⁵ Enthusiasts see epigenetics as a channel of inheritance the synthesis missed entirely. Skeptics argue that most epigenetic marks are reset between generations and that the documented cases, while fascinating, remain marginal compared to DNA-based inheritance. The honest answer is that we do not yet know how much weight epigenetic inheritance carries in evolution at large.

Phenotypic Plasticity and Genetic Assimilation

The Modern Synthesis focused on genetic variation as the raw material for natural selection. But organisms also respond to their environments during their own lifetimes, producing different phenotypes from the same genotype. A plant grows taller in good soil and shorter in poor soil without any genetic change. This flexibility is called phenotypic plasticity, and a growing number of biologists argue it may play a creative role in evolution that the synthesis overlooked.

The idea is that when an organism responds to a new environment by developing a new trait through plasticity alone, natural selection can then act on the genetic variation underlying that plastic response. Over generations, what started as an environmentally triggered response can become genetically fixed, a process called genetic assimilation. Research in amphibians has provided some of the strongest support for this process, where environmentally induced changes appear to precede and facilitate the origin of genuinely novel traits.¹⁶ Detailed modeling suggests that during genetic assimilation, epigenetically induced phenotypic changes are incrementally replaced by multiple minor genetic mutations through selection.¹⁷

If plasticity regularly initiates evolutionary novelty, the synthesis’s picture of evolution as a process that begins with random mutation needs revising. The environment would not just select among existing variants but actively provoke the appearance of new ones.

Horizontal Gene Transfer and the Web of Life

The Modern Synthesis imagined evolution as a branching tree: species split off from common ancestors, and lineages diverge. Genome sequencing has revealed that, at least for microbes, the tree is tangled with horizontal connections. Bacteria routinely pick up genes from unrelated species, and this horizontal gene transfer has been a constant force reshaping microbial genomes throughout evolutionary history.¹⁸ The result is that different genes in the same organism can have wildly different evolutionary histories, which complicates any attempt to draw a single tree of life.¹⁹

Among eukaryotes (organisms with complex cells, including everything from amoebae to humans), horizontal gene transfer is less pervasive but far from absent. The endosymbiotic events that gave rise to mitochondria and chloroplasts involved massive transfers of genes from the engulfed bacteria to the host cell’s nucleus. Photosynthetic lineages among single-celled eukaryotes have been shaped by multiple rounds of endosymbiosis and gene exchange.²⁰

None of this was on the synthesis’s radar. The founders were primarily thinking about sexually reproducing animals and plants, where inheritance follows Mendelian rules and lineages branch cleanly. Microbes, which represent the overwhelming majority of life’s diversity, play by different rules, and the synthesis’s tree-like model of descent does not adequately describe their evolutionary history.

Niche Construction

A standard synthesis-era view of adaptation treats the environment as a given: organisms adapt to it, or they perish. But organisms also modify their environments in ways that alter the selection pressures they and other species face. Beavers build dams, earthworms transform soil chemistry, and humans reshape entire ecosystems. These activities, collectively called niche construction, create feedback loops between organisms and their environments that can drive evolution in directions the standard model does not predict.²¹

Proponents of niche construction argue that it deserves recognition as an evolutionary process on par with natural selection and drift, not just a background condition. Critics counter that niche construction is simply an aspect of phenotype that natural selection can already explain. The debate reflects a deeper disagreement about whether the synthesis’s one-way causal arrow, from genes through development to phenotype to selection, adequately captures how evolution actually works, or whether reciprocal causation between organism and environment needs to be built into the framework.

The Extended Evolutionary Synthesis Debate

All of these challenges, from evo-devo and epigenetics to plasticity, niche construction, and horizontal gene transfer, converge in a formal proposal called the Extended Evolutionary Synthesis (EES). Proponents argue that the Modern Synthesis, while not wrong, is incomplete. It underestimates the role of development, non-genetic inheritance, and organism-environment feedback, and it needs to be expanded to accommodate them.²² The EES retains the fundamentals of evolutionary theory but shifts emphasis toward constructive developmental processes and reciprocal causation between organisms and their environments.

Others in the field see no need for a new label. They argue that the Modern Synthesis has always been a flexible framework that absorbs new findings, as it absorbed neutral theory and molecular biology before. On this view, evo-devo, epigenetics, and the rest are exciting additions to evolutionary biology but do not require restructuring the theory’s foundations. A philosophical analysis of the disagreement suggests the two frameworks have substantial conceptual differences but are not incommensurable paradigms; researchers on both sides are still working within recognizably the same scientific tradition.²³

The synthesis’s central claim, that gradual accumulation of small genetic changes within populations scales up to produce macroevolutionary patterns, is itself part of this reexamination.²⁴ How much the synthesis’s micro-to-macro logic holds up, versus how much macroevolution has its own dynamics, remains genuinely unsettled.

Why the Synthesis Still Matters in Practice

However the theoretical debate shakes out, the Modern Synthesis’s core framework has enormous practical consequences. One of the clearest is antibiotic resistance. The rise of drug-resistant bacteria is a textbook illustration of synthesis-era population genetics: random mutations conferring resistance arise in bacterial populations, and the selection pressure of antibiotic use drives those resistant variants to high frequency. But the story has a layer the synthesis did not anticipate. Resistance genes also spread by horizontal gene transfer, jumping between unrelated bacterial species on mobile genetic elements. Microbes have exploited every available channel of gene transmission to develop resistance mechanisms against every antibiotic put into clinical or agricultural use.²⁵ Understanding resistance requires both the synthesis’s population genetics and the post-synthesis recognition that horizontal gene transfer reshapes microbial evolution in ways vertical inheritance alone cannot explain.

A similar blend of old and new theory applies to gene-culture coevolution in humans. Cultural practices like dairying changed selection pressures on human populations, favoring the persistence of lactase production into adulthood in some lineages. Modeling this kind of feedback requires dual-inheritance theory, which treats cultural transmission and genetic transmission as two interacting inheritance systems.²⁶ The synthesis had no framework for this because it focused exclusively on genetic inheritance. Dual-inheritance theory extends Darwin’s framework by treating culture as a second channel of inherited information that interacts with genetics.²⁷

Where the Polygenic Problem Stands

One of Fisher’s central insights, that many traits are influenced by large numbers of genes each contributing a small effect, has been dramatically confirmed by modern genomics. Genome-wide studies of traits like height, body mass, and disease susceptibility consistently find that hundreds or thousands of genetic variants contribute, each nudging the outcome slightly. Quantitative geneticists have long known that adaptive variation is highly polygenic, but this is only now becoming fully appreciated by molecular biologists, and the field still lacks a satisfying framework for understanding how natural selection acts on such diffuse genetic architectures.⁴

This is not a minor gap. If most adaptive traits depend on subtle shifts across hundreds of genetic variants, then tracking evolution gene by gene misses the forest. Understanding how selection moves whole distributions of allelic effects, how developmental systems buffer or amplify those effects, and how environmental feedback alters the landscape is a problem that sits right at the intersection of classical synthesis population genetics and the newer concerns of evo-devo and systems biology. It is a place where the synthesis’s mathematics remain indispensable, even as the questions outgrow the original framework.