Speciation and Macroevolution
Dolph Schluter
Since the Big Bang, not much has happened in the universe more interesting than the diversification of life on earth. Most of life’s current diversity is wrapped up in the genetic and phenotypic differences between species, between the communities of species they form, and between the higher taxonomic groups that species make up, such as families and phyla. For this reason the study of the origin of species—speciation—and its consequences tells us a great deal about how the extraordinary diversity of life arose, how it is distributed across the globe, how it is presently maintained, and how it has changed through billions of years of earth’s history.
To Charles Darwin, how new species evolve was the “mystery of mysteries.” In his 1859 masterpiece, On the Origin of Species by Means of Natural Selection, he took the first big steps to demystifying the process. Darwin recognized that in nature there was no sharp discontinuity between the differences one sees among populations within species and the differences observed between closely related species: “I look at the term species as one arbitrarily given, for the sake of convenience, to a set of individuals closely resembling each other, and that it does not essentially differ from the term variety, which is given to less distinct and more fluctuating forms.” The origin of a species is not a sudden instant in the history of life but one that (usually) results from a steady accumulation of differences. Those differences, he explained, are the product of gradual evolution by natural selection of variation present in populations.
We now recognize that Darwin’s solution was incomplete. One reason is that our concept of species has evolved. In Darwin’s day, species were designated according to the magnitude of morphological differences: “the amount of difference is one very important criterion in settling whether two forms should be ranked as species or varieties.” This means that except for the magnitudes involved, morphological criteria grouped populations into species just as species were grouped into genera, and genera into families. Under this concept, speciation is the evolution of differences in ordinary phenotypic traits sufficiently large to warrant the taxonomic designation species. Darwin appreciated that matings between different species often produced inviable or sterile offspring, but he decided that this was not universal and was less reliable than morphological differences for classifying species.
The focus of speciation study changed with the development of the biological species concept (see chapter VI.1). In 1937, Theodosius Dobzhansky defined speciation as the evolution of “isolating mechanisms,” traits that reduce gene flow between populations. Subsequently, Ernst Mayr defined species as “groups of actually or potentially interbreeding natural populations, which are reproductively isolated from other such groups.” Reproductive isolation doesn’t just mean hybrid sterility and inviability—it includes any genetically based difference that acts as a barrier to the movement of genes between populations. The mechanical, chemical, behavioral, and ecological traits that reduce interbreeding all contribute to reproductive isolation. Reproductive isolation also includes traits that inhibit fertilization after mating and any evolved behavioral, ecological, and genetic factor causing hybrids to be relatively unsuccessful. Speciation research today is focused on answering the question, How does reproductive isolation evolve?
In the hierarchy of categories that we use to classify life’s diversity, the evolution of reproductive isolation is a feature unique to the species category. No equivalent process takes place during the evolution of a new genus or a new family. Speciation therefore has special significance in the evolution of diversity.
Another reason for its importance is that speciation occupies a juncture between the scales of processes—small and large—that have produced the patterns of diversity we see today. Speciation is undoubtedly a microevolutionary process. It is the outcome of accumulated genetic divergence between populations, and all intermediate stages are represented among contemporary populations in nature. Speciation is a macroevolutionary event as well, because species are the basic unit for measuring large-scale changes in life’s diversity over long spans of time. This dual interpretation of speciation makes sense because once gene flow is sufficiently restricted, new species can evolve independently of others. The completion of a speciation event changes patterns of biodiversity in many ways. A new species can coexist with its closest relatives without their collapsing to a hybrid swarm. Thus speciation affects the numbers of species coexisting in ecological communities. A new species in its defined geographic range also incrementally affects patterns of species diversity across the globe by influencing the number of species present in each region. Each new species also contributes to the diversity of the larger clade of species—all related by descent—to which it belongs. A speciation event may thus affect the probability of long-term persistence of its whole lineage. To the extent that traits of a new species are shared by its close relatives, and tend to be passed on from ancestor to descendant species (with some modification), the evolution of a new species alters the frequency distribution of traits represented in earth’s biota. A speciation event is thus the seed of long-term patterns in the history and future evolution of life.
For all these reasons, speciation has played a large role in many of the processes that brought about the modern diversity of life and its distribution over the face of the earth. But what do we really know about these processes? What drives the origin of species, and what determines their geographic distribution? What is the connection between speciation and the evolution of ordinary phenotypic traits? Are the composition of species communities, and long-term trends in local and regional species numbers, affected by the mechanism of speciation? Why do some lineages speciate more often than others? What determines the overall rates of speciation and extinction? Are the factors that determine the success of some lineages over the long term, and the decline of others, the same as those that drive microevolution and the origin of species? In this section, we review what is known about the causes and global consequences of speciation.
SECTION THEMES
Species Diversity Patterns
Measuring species numbers over time and their distribution over the earth requires that we know what a “species” is. Speciation researchers today focus on the biological species concept, where (nearly complete) reproductive isolation is key. However, this is not always a practical criterion for classifying organisms to species. For example, how does one decide whether two populations are “actually or potentially interbreeding,” and thus belong to the same biological species when their ranges do not overlap, or the populations are known only from fossils, or their individuals reproduce asexually? Chapter VI.1 discusses these issues and some alternative species criteria that have also been proposed, their connections to one another, and their implications. Debates over species concepts are largely resolved by recognizing that different species criteria emphasize different stages and features along the continuum of changes that take place during the origin of species.
Centuries of exploration and survey have established that species are extremely unevenly distributed across the face of the earth and through time. The causes of this unevenness continue to challenge biologists. The most prominent spatial pattern is the latitudinal diversity gradient, whereby more species occur in the tropics than in the temperate zone. This pattern is seen in both sexual and asexual species, suggesting similar underlying causes. Chapters VI.2 and VI.3 describe these and other patterns of species diversity in space and time, the possible mechanisms that produce them, and their consequences.
Gene Flow
Divergence of populations is a tug-of-war between the forces that generate differences (mutation, genetic drift, natural and sexual selection) and the main process that erodes differences: gene flow. Speciation marks the point at which barriers to gene flow are strong enough to resist the effects of gene flow. Yet, if there is gene flow, it is difficult for reproductive isolation to evolve in the first place. The amount of gene flow between two populations is strongly influenced by their geographic distributions (see chapter VI.3), being least when the populations are fully separated by a geographic barrier to movement (allopatric), and most when they overlap in distribution (sympatric).
How does gene flow retard speciation? First, it slows or prevents genetic divergence between populations when divergence is not directly favored by selection, such as when it occurs by genetic drift, or when separate populations adapting to similar selection pressures by chance experience and fix different advantageous mutations. In such cases, gene flow moves alleles among populations, eroding genetic differentiation. Gene flow is less destructive when selection is divergent, favoring different alleles in different populations, because an allele that flows from a population in which it is favored to another in which it is not favored will eventually be removed by selection, provided selection is sufficiently strong. Second, even if selection is divergent, gene flow slows speciation by breaking up associations between alleles at different genes. For speciation to proceed, genes responsible for reproductive isolation, such as those that influence mate preferences, must become associated with the genes under divergent selection. The individuals in one population must prefer to mate with other individuals from the same population, rather than with individuals from another population. Gene flow will break up these associations, bringing mate preference genes from one population to the other, breaking down reproductive isolation between them. Consideration of this problem has led theorists to make predictions about the specific circumstances under which speciation with gene flow can nevertheless occur (see chapter VI.3). When it does occur in the face of gene flow, it should leave detectable marks on the types of genetic differences that evolve, on the strength of different kinds of evolved barriers to gene flow (see chapter VI.4), on genome-wide patterns of genetic differentiation (see chapter VI.9), and on spatial distributions of species (see chapters VI.2 and VI.3).
Paradoxically, episodes of gene flow between already well-differentiated species can sometimes be a creative rather than a homogenizing process, resulting in new genetic combinations that have novel phenotypes and represent brand-new “hybrid” species (see chapters VI.6 and VI.9)
Mechanisms of Speciation
What are the forces that generate differences and bring about the evolution of reproductive isolation during speciation? For a long time the answers to this question came mainly from theory and laboratory experiments, which evaluated the plausibility of speciation by genetic drift, founder events, divergent natural selection, and other processes. Only recently have we been able to test these ideas in nature and say with confidence how real species in nature have formed.
Since Darwin the role of natural selection has been of great interest, and we know more about its role than that of any other process. For example, natural selection on ordinary phenotypic traits may incidentally build reproductive isolation between populations as a by-product (see chapter VI.4). Such isolation can occur when separate populations adapt to different environments (ecological speciation) as different alleles favored in one environment but not the other gradually accumulate between populations. Alternatively, selection may build genetic differences among populations that experience similar selection pressures if the populations by chance experience and accumulate different sets of advantageous mutations (mutation-order speciation). Finally, there has been a long-standing interest in the role of genetic drift—speciation without any natural selection at all—but we still don’t know much about its importance.
When speciation involves natural selection, what are its mechanisms? As described in Section III: Natural Selection and Adaptation, adaptation to abiotic factors in the environment such as soil and climate is one possibility, and there are now good examples of this process. Biotic interactions with other species, including predator-prey, host-parasite, competition and mutualism, are also a major (perhaps the major) source of selection on populations. Furthermore, reciprocal evolutionary changes between interacting species—coevolution—might bring about rapid divergence between populations within each of the interacting species (see chapter VI.7; see additional examples in chapters VI.10 and VI.16). Strong natural selection can also result from internal genomic conflict. For example, reproductive isolation might evolve as a by-product of conflict resolution between different genetic elements within individuals (intragenomic conflict) or between the sexes (intergenomic conflict) (see chapters VI.4 and VI.8).
Natural selection might also contribute to speciation when it directly favors stronger prezygotic reproductive isolation between incipient species when their hybrid offspring have reduced fitness (see chapter VI.4). This process, called reinforcement, represents the only known circumstance in which natural selection directly favors the evolution of stronger reproductive isolation. Otherwise, as described earlier, the role of selection is indirect—reproductive isolation evolves as an incidental consequence of adaptation.
It has often been pointed out that the most conspicuous differences between closely related species are often in secondary sexual traits, such as in courtship or body coloration of males, rather than in ecological traits (see chapter VI.5). Since the exaggeration of such traits is caused by sexual selection, it seems likely that sexual selection is also frequently involved in speciation. Any such role would likely involve natural selection, too, because it is the process that leads to divergence of mate preferences or that favors the evolution of traits that ameliorate intergenomic conflict.
The genetic changes that underlie the evolution of reproductive isolation (speciation genes) are finally being discovered, and the hunt for them represents one of the most exciting directions in modern speciation research (see chapter VI.8). Sometimes, mutations of large effects on reproductive isolation are found, whereas reproductive isolation often results from the accumulated effects of many small-effect mutations. Genetic “signatures” of selection detected on speciation genes provide some of the best evidence that natural and/or sexual selection have been responsible for driving the mutations to high frequency, and hence for the evolution of reproductive isolation. We are beginning to learn why speciation genes are often clustered rather than dispersed within the genome and how the genome evolves collectively during the evolution of reproductive isolation (see chapter VI.9).
A surprisingly common mechanism of sudden speciation is via the evolution of polyploidy. Individuals that have more than two sets of chromosomes are occasionally formed (see chapter VI.9), and as a result, they instantly possess some degree of reproductive isolation from their diploid ancestors. This makes polyploidization the fastest mode of speciation known. Often, the polyploids are hybrids between two species. The process is most common in plants, but several examples from animals have recently been discovered.
Adaptive Radiation
When a group of organisms experiences a flurry of speciation events in association with adaptation of nascent species to different ecological niches, the result is adaptive radiation (see chapter VI.10). Classic examples include the finch radiations on the Galápagos and Hawaiian Islands. In both cases the species have evolved a wide diversity of beak sizes and shapes that enhance the ability of individuals to exploit particular resources, such as hard seeds, nectar from long-tubed flowers, or insects under bark. In the few adaptive radiations that have been studied intensively, it is clear that the same natural selection pressures that adapt populations to distinct niches also indirectly contribute to the buildup of reproductive isolation between populations. In these few studied cases, at least, there is a close connection between rapid speciation and adaptive evolution.
Adaptive radiations are particularly prevalent where ample resources are available, and few competing lineages take full advantage of them (ecological opportunity). Even under such conditions, however, some lineages diversify more readily than others, as though they have intrinsic differences that affect their abilities to speciate rapidly, or to adapt to and usurp, novel resources. One reason might be that the fortunate lineages possess key traits that increase their evolvability or their propensity to speciate (key evolutionary innovations; see chapter VI.15). For example, it has been proposed that the huge diversity of angiosperm plants is attributable to the evolution of the flower. Adaptation of flower structures to different suites of pollinators in different environments might speed the evolution of premating reproductive isolation. Another hypothesis is that the evolution of traits permitting certain insects to consume plant tissue is behind the astonishing diversity of phytophagous insects, such as herbivorous beetles, found today. Plants are incredibly abundant and diverse in their leaf structures, chemistries, and life histories, which favors niche specialization and diversification by insects that exploit them. Such hypotheses are challenging to test, but great strides are being made.
Evolutionary Rates
Adaptive radiations represent episodes of particularly fast evolution and speciation. In contrast, study of patterns of evolution in the fossil record and in phylogenetic trees has found that evolution is often slow. Lineages frequently undergo long periods in which little evolution seems to take place—at least in easily identified morphological traits (perhaps rates are not so slow in other aspects, such as at genes involved in fighting disease). The hypothesis of punctuated equilibria was an extreme statement about rates of evolution in nature: that evolution hardly ever occurs except in the relatively brief periods during which speciation also takes place. The rest of the time, so the hypothesis goes, species exhibit stasis, changing little. This conjecture prompted a great deal of research that continues to examine the true relationships among speciation, time, and trait evolution (see chapters VI.11 and VI.12). A key question is whether the punctuated equilibrium is a caricature of evolutionary patterns in the fossil record. Sustained directional changes in traits might indeed occur infrequently and episodically, but the rest of the time evolution might be better described as oscillating rather than static, or at least not sustained and directional, with fluctuations of varying amplitude taking place through time. Chapter VI.11 describes additional patterns, including the paradoxical observation that measured rates of evolution appear faster the shorter the time interval over which change is measured, and the observation that rates of phenotypic evolution appear to be highest early in a clade’s history.
Macroevolutionary Trends
The traits that a species possesses—such as the mean body size of its individuals or its geographic range size—can influence the rate at which it subsequently produces new species and the probability that it will go extinct (see chapter VI.12). If a relationship between a trait possessed by species and speciation or extinction rates holds consistently across multiple lineages and over time, the result will be a large-scale increase in the prevalence of that particular trait in nature. We can think of this process as species selection—the macroevolutionary analogous of natural selection on individuals within populations. The notion of species selection has been controversial, and many researchers regard it as a weak force compared with ordinary natural selection within species. For example, a large-scale trend toward larger body size in the fossil record appears to be mainly the result of ordinary natural selection within species accumulated over a long time span (see chapter VI.12). However, species selection need not oppose ordinary natural selection, and it may indeed generate trends in the absence of any net direction to evolution within species. Current research focuses on the evidence for species selection driving macroevolutionary trends, and we now have good examples of the process (see chapters VI.12 and VI.14).
One of the most striking discoveries from the fossil record is that long-term success and failures of lineages are not necessarily determined by the same factors that drive evolution within populations (see chapter VI.13). Dinosaurs possessed exquisite adaptations to the many environments in which they occurred, yet they were wiped out en masse by catastrophic environmental changes never before experienced during their many millions of years of history. Indeed, it does not often happen that the lineages that dominated the earth prior to mass extinctions recover and reassume their dominant positions afterward. More typically, previously minor components of life’s ensemble proliferate subsequently and become the new dominants, which in turn results in wholesale changes to the frequency distribution of different kinds of traits represented in nature. These changes may often be due more to chance (species drift) than selection, though some species selection seems to occur during mass extinctions, consistently favoring lineages with certain traits such as a broad diet and a large geographic range (see chapter VI.13).
The resolution of these microevolutionary and macroevolutionary forces has left its mark on the composition of life on earth and on the communities of species seen today (see chapter VI.16). The macroevolutionary processes of speciation and extinction, of species selection and species drift, acting over long spans of time created the biodiversity that has assembled into ecological communities and continues to influence how local assemblages change through time. Natural selection on variation within species has produced the myriad adaptations of species to one another and to the abiotic environments they encounter across their geographic ranges. As species adapt to one another, the strength of their interactions changes and the flow of energy and materials is altered, producing consequent changes in the properties and dynamics of the surrounding ecosystem. Thus, microevolution on species within communities generates new species and modifies the traits that species possess and so provides the material that drives macroevolutionary changes. These changes, in turn, will affect the course of future evolution.