1. All complex organisms depend on coevolved mutualistic interactions
2. Coevolution shapes defenses and counterdefenses
3. Coevolution of competitors further structures the web of life
4. Species coevolve as a geographic mosaic
5. Coevolution may sometimes foster speciation
6. Coevolution may result in predictable webs of interaction
7. The coevolutionary process is pervasive in human endeavors
Coevolution is reciprocal evolutionary change among interacting species driven by natural selection. It is the evolutionary process by which many predators and prey, parasites and hosts, competitors, and mutualists adapt to each other in the constant struggle for life. It is also a process that can sometimes lead to new species, as different populations of interacting species coevolve in different ways in different geographic regions. Through its effects on adaptation and speciation, coevolution continually reshapes the web of life. Moreover, human society is increasingly altering the coevolutionary process through manipulation of ecological relationships among species within and among ecosystems, alteration of the genetic structure of crop plants, and development of novel strategies for mitigation of human diseases.
coevolution. Reciprocal evolutionary change in interacting species driven by natural selection
coevolutionary cold spot. Geographic regions in which one of a set of interacting species does not occur or in which the interaction, although occurring, does not result in reciprocal evolutionary change
coevolutionary hot spot. Geographic regions in which interactions between two or more interacting species result in reciprocal evolutionary change
local adaptation. Adaptation of populations to the local physical environment or to the local populations of other species with which they interact
Coevolution has been a major part of the process of evolution at least since the beginnings of complex life on Earth. In fact, many of the major events in the history of life are a direct result of the coevolutionary process that has created mutualistic symbioses among species. All complex organisms rely on mitochondria for cellular respiration. Those mitochondria are ancient bacteria that coevolved with their hosts and eventually became obligate organelles within the cells of all eukaryotic life. Every multicellular organism therefore has two genomes, a nuclear genome and a mitochondrial genome, as a direct result of this ancient coevolutionary process. Most animal species harbor one or more other coevolved symbionts that are necessary for their survival and reproduction. Among the most common are gut symbionts that aid digestion and nutrition. In many insect species, for example, coevolved symbionts provide one or more essential amino acids that are missing in the diet.
Plants harbor yet other ancient coevolved partners: obligate symbionts called chloroplasts. These organelles drive photosynthesis, and few plant species can survive without them. Many plants also rely on mycorrhizal fungi that attach to the roots of plants and aid in nutrition. Legumes and a few other plant taxa have coevolved relationships with rhizobial bacteria that convert atmospheric nitrogen into a form usable by the plant. In addition, the leaves of some herbaceous plants and some trees are laced with endophytic fungi, whose coevolved relationships with plants are only now being explored in depth. A majority of plants also rely on animal pollination for reproduction. Hence, survival and reproduction in most plant species require interactions with multiple other species, and many of these interactions are highly coevolved.
In general, all major biological communities are based on coevolved mutualistic relationships that form the underpinnings for community structure and succession. Most terrestrial communities rely on lichens (which are mutualistic symbioses between fungi and algae), mycorrhizal fungi, rhizobial bacteria, and chloroplasts to create the organic base on which other microbial life and animal species rely. Take away those coevolved mutualisms, and terrestrial life as we know it would disappear.
The same central ecological role for coevolved interactions holds for marine communities. Coral reefs, which harbor so much of the diversity of marine life, are a result of coevolved mutualistic symbioses between corals and the dinoflagellates (sometimes called zooanthellae) they harbor. The phenomenon called “coral bleaching,” which has increasingly devastated coral reefs worldwide, results from the loss of the dinoflagellate symbionts brought about by multiple forms of environmental change. Much deeper in the ocean, deep-sea vents harbor species that rely on complex symbioses that oxidize sulfide to produce energy, much as plants use the sun’s energy for photosynthesis. As new molecular tools continue to be developed to study microbial species, it is becoming increasingly evident that coevolved symbioses with microbial taxa permeate oceanic and terrestrial communities worldwide.
In addition to the intimate associations found in symbiotic mutualisms, many species are involved in short-term mutualisms that are fundamental to the structure and maintenance of biological communities. The majority of plants in terrestrial communities from the polar regions to the tropics rely not only on pollinators to fertilize their ovules but also on seed dispersers to distribute their seeds to new sites for germination. In marine communities, some fish species are involved in mutualisms with other fish species, called cleaner fish, that groom parasites from the skin of host fish.
As these mutualisms coevolve, natural selection hones the complementarity of the traits involved in the interaction (e.g., deep floral corollas in some flowers and long tongues in some pollinators), and other species evolve to converge on that common set of traits. The result is the formation of webs of mutualistic species. For example, in interactions between plants and pollinators, plants have evolved to provide pollen, nectar, or resins as rewards to attract pollinators, and a wide range of animal species have converged in the shapes of their mouthparts to extract the rewards from flowers. Plant species in different families have often converged on traits that attract particular groups of pollinators, just as some birds, bats, or insects have converged on traits that allow them to visit flowers with particular shapes.
Hence, mutualisms often accumulate groups of phylogenetically related and unrelated species over time, as species converge through natural selection on similar traits. Similarly, many unrelated plant groups have converged on similar fruit sizes, shapes, and colors to attract frugivorous birds, and birds from multiple avian families have converged on the traits that allow them to exploit these fruits as a major part of their diet. The combination of convergence and complementarity of traits therefore sometimes creates a coevolutionary vortex that continues to collect new species into the interaction over millions of years and shapes the structure of ecological communities.
Not all interactions, of course, are mutualistic. Almost all species are attacked by parasites and predators. These interactions drive yet other forms of coevolution as species evolve defenses and counterdefenses against each other. Ongoing coevolution holds true for even the simplest forms of life. Bacteria are attacked by a great diversity of viruses known as bacteriophages. Estimates of the molecular diversity of oceanic bacteria and phages indicate that phage diversity may be at least as great as bacterial diversity. Multiple experiments have shown that bacteria and phages undergo very sophisticated forms of coevolution involving molecular changes in cell walls of bacteria, driven by a relentless battle between the ability of bacteria to prevent phages from breaking through the cell wall and the ability of phages to thwart those defenses and penetrate the cell wall. In experimental studies within laboratory ecological microcosms, bacteria and phages have been observed repeatedly to undergo rapid coevolution through natural selection on new mutations within the period of only a few weeks. These experiments suggest that even the simplest forms of life undergo continual coevolution with enemies.
Coevolution with enemies is equally common among multicellular species, and the signature of that process can be seen in the ecological lifestyles and the traits of species. The most common way of life on Earth is parasitism. There are more known species of parasites than there are of all other kinds of species. In turn, every species that has been studied in detail has been shown to have traits that have evolved to thwart attack by parasites or predators. Tens of thousands of chemical compounds found in plants are thought to have evolved as defenses against enemies. Similarly, all animals have physiological defenses that thwart attack by the many parasitic species that are constantly attempting to gain access to animal tissues. In most cases, we do not know many of the details about how coevolution shapes these interactions, but we do know that they are pervasive. Humans, for example, are subject to at least 1400 known diseases caused by pathogens and parasites.
Coevolution with predators or parasites can sometimes lead to multiple rounds of increasingly high levels of investment in defenses and counterdefenses through the process of directional natural selection. Such coevolutionary arms races are not sustainable in the long term but can nevertheless result in highly exaggerated traits. One of the most visual examples is that of camellia fruits that have evolved to be much larger in some populations than in others as a defense against camellia weevils that must eat through the fruit to get to embedded seeds on which they lay eggs. In response, some camellia weevils have evolved a highly elongated head that is longer than the entire length of the rest of the body, allowing the insects to penetrate the fruit to the seeds.
Alternatively, some parasites and hosts coevolve through natural selection that favors rare genetic forms of the host species, resulting in fluctuating genetic polymorphisms in the coevolving species. Parasites evolve adaptations to the most common genotypes in a host population, and natural selection therefore favors hosts that have rare genotypes to which the parasite is not adapted. This process of frequency-dependent natural selection (i.e., selection favoring the least frequent genotype) results in the maintenance of multiple genetic forms within the coevolving populations. The remarkable allelic diversity found in the mammalian immune system is thought by biologists to be maintained by this form of coevolutionary selection acting on hosts and their parasites.
Genetic novelty among individuals within populations can be enhanced by the recombination of genes that accompanies sexual reproduction, and one of the leading hypotheses for the evolution of sexual reproduction is that it is favored by coevolution between parasite and hosts. According to this hypothesis, sexual females are more likely than asexual females to produce offspring with rare genotypes, and these genetically different offspring of sexual females are, on average, more likely to be resistant to parasites than the genetically identical offspring of asexual females. Hence, one of the most important aspects of the ecology of organisms, the process of sexual reproduction, may result at least in part from the process of coevolution with fast-evolving parasites.
Some parasites or predators interact locally with multiple host species, creating more complex forms of coevolution. One major hypothesis for this form of multispecific coevolution is called coevolutionary alternation. According to the hypothesis, natural selection favors parasites or predators that attack the currently least-defended host or prey species. Selection acting on a host or prey population that is being strongly attacked favors individuals that have higher levels of defense, leading eventually to an overall increase in the level of defense in that population. Those increased defenses in turn favor parasites or predators that attack other, less-defended, species. The result is a constantly changing mix over time of which host or prey species are attacked.
The best-known potential example of this form of coevolution is that between European cuckoos and the birds they parasitize. Cuckoos lay their eggs in the nests of other bird species, and the unsuspecting hosts raise the cuckoo nestlings. Cuckoos differ across Europe in the bird species they parasitize, and there appear to be geographic differences in preference for host species. The bird species that serve as hosts differ among populations in the defenses they harbor. In addition, some host populations are heavily attacked but have low defenses, suggesting that they are locally new hosts for the cuckoos. Some populations of other bird species have high defenses but are rarely attacked, suggesting that these populations may have formerly been hosts but have now been abandoned in favor of local host species with lower defenses. In general, these interactions appear to coevolve through a continual reshuffling of preference in the cuckoos and levels of defense in the host species.
Not all interactions with parasites lead either to escalating arms races, fluctuating polymorphisms, or coevolutionary alternation. In some interactions between parasites and hosts, the relationship may coevolve toward attenuation of the level of antagonism, especially if parasites are transmitted directly from mothers to daughters at birth. This mode of transmission can favor parasites that are less virulent than other parasites, because selection favors parasite genotypes that do not kill their host. If a female host dies before she can reproduce, that lineage of parasite is not transmitted and becomes extinct. Over time, then, less virulent lineages tend to spread in populations under these conditions of transmission. Although some biologists in the past have assumed that parasites will tend to evolve toward decreased virulence over time, much empirical work and mathematical theory in coevolution has shown that assumption to be false. There is nothing inevitable about the direction of evolution in interactions between parasites and hosts. The trajectory of coevolutionary selection depends on the ecological and genetic conditions in which the interaction occurs.
The web of life gains additional structure from coevolution among competing species. Wherever species share resources that are insufficient to support all individuals, competition favors those that are either more efficient than others in garnering limited resources or those individuals that use alternative resources. Through competition for limited resources, coevolutionary selection favors divergence among species in the use of those resources. The divergence can take multiple forms, including the use of different foods, the use of different habitats, or the use of the same foods or habitats but at different times of year.
The outcome of such competition is called character displacement. Its effects can be seen through studies of geographic variation in the morphologies of species that co-occur with competitors in some parts of their geographic range but not others. Competing animal species often differ in body size or other morphological characters to a greater degree in regions where the species occur together than in regions where they do not overlap. In fact, most published studies of competition have focused on how species diverge in morphological characters associated with food choice when in competition with other, closely related species.
Although competition for resources may be the cause of most instances of character displacement, some instances of displacement may result from other ecological causes. For example, if two prey species are attacked by the same predator species, then natural selection could favor character displacement among the prey species. A prey species that is different from other coexisting species may have a selective advantage because the local predators may become adapted to only the most common species.
In plants, some instances of character displacement are driven by selection to avoid hybridization with other plants that share the same pollinators. In such instances of reproductive character displacement, natural selection favors plants with flowers that minimize the chance that pollen from another species will reach their stigmas and produce hybrid offspring with lower Darwinian fitness. Reproductive character displacement may involve the evolution of flowering at different times of the year from closely related plants, or the evolution of floral shapes that exclude pollinators that visit other plant species.
As species coevolve through mutualistic and antagonistic interactions throughout their geographic ranges, natural selection may come to differ among environments. Traits of interacting species that are favored by selection in one environment may be ineffective in other physical or biotic environments. Over time, then, interacting species may exhibit complex geographic patterns in the traits shaped by this selection mosaic. One species may evolve to be larger than other competitors in one geographic region but evolve to be smaller than other competitors in other regions. Similarly, parasites and hosts, or predators and prey, may evolve different arsenals of defenses and counter-defenses in different populations. For example, some plant species are known to have evolved different mixes of chemical defenses in different geographic areas, and the specialist insects that attack these plants have evolved, in response, mixes of detoxification compounds that are customized to their local plant populations. The resulting geographic selection mosaic in local adaptation is the raw material that fuels the process of coevolution across the Earth’s constantly changing landscapes.
In addition, natural selection may sometimes be intensely reciprocal in only some environments. These coevolutionary hot spots may be embedded in a matrix of coevolutionary cold spots, where the interaction does not occur or is detrimental or beneficial to one species but not to other interacting species. A parasite, for example, may commonly cause death to its hosts in some environments but have relatively little effect on host fitness in other environments. Coevolving species may therefore commonly be a complex mix of populations that differ in the degree to which coevolution is driving the evolution of the interacting populations.
Gene flow, random genetic drift, and metapopulation dynamics can add to the geographic mosaic of coevolution by continually changing the geographic areas in which different combinations of coevolving traits occur. A trait evolved in one population may spread to other, but not necessarily all, populations of a species. In addition, traits can be lost in some local populations through random genetic drift. The continual remixing of genes and traits among populations adds to the geographic mosaic of coevolving species.
The geographic mosaic theory of coevolution argues that selection mosaics, coevolutionary hotspots, and trait remixing are common features of coevolving interactions. Evidence for the geographic mosaic of coevolution has been shown in an increasingly wide array of interacting species, including herbaceous plants and insect herbivores, conifers and seed-eating birds, and snakes and toxic salamanders. The persistence of some interactions over millions of years may be a result of the geographic mosaic of coevolution, which allows coevolution to proceed simultaneously in different directions and involve different traits in different populations.
In some cases, the geographic mosaic of coevolution may lead to speciation. As coevolving populations adapt locally to each other, they may diverge to such an extent that they become reproductively isolated from other populations. In fact, one of the major hypotheses for the remarkable diversification of life is that coevolution has repeatedly favored starbursts of speciation between interacting lineages as the species involved continue to evolve novel defenses and counterdefenses.
This process of diversifying coevolution may result in groups of closely related species that are specialized to interact with only one or a few other species. For example, crossbills are a group of birds specialized for extracting seeds from conifers before the cones open. The birds use their crossed bills to pry apart the scales of conifers so that they can then extract the seeds with their tongues. Different crossbill populations have evolved different beak shapes, each adapted to a different conifer species. Some are specialized on particular pine species, others on particular fir or spruce species. In response to attack by these specialist populations, some conifer species have evolved novel cone shapes and sizes that increase the difficulty for the crossbills when they are attempting to extract seeds. Moreover, populations of crossbills and conifers even within the same species differ geographically in their patterns of coadaptation. Overall, crossbills appear to have diverged into a group of specialist species each adapted to different conifer species, and sometimes even to particular populations of a conifer species, as a direct result of their ongoing coevolution with conifers.
Although it is relatively easy to understand how coevolution may shape interactions between a pair of species, it is a greater challenge to understand how coevolution shapes more complex webs of interacting species. In species-rich biological communities, some species may interact with many other species, whereas others may interact with few species, creating tangled webs of interaction. Recent studies, however, suggest that coevolution may shape webs of dozens or even a hundred or more interacting species in predictable ways. Research on food webs of predators and prey has shown that these interactions are commonly compartmentalized into multiple smaller subwebs of species that interact more often with one another than with other species within the larger web. This is expected from coevolutionary theory because coevolution between antagonistic species can lead to groups of species that share similar defenses and counterdefenses, and natural selection favors prey that interact with fewer rather than more predators.
In contrast, mutualistic interactions between free-living species such as plants and their pollinators or frugivores tend to group into few subwebs. Instead, they form more of what is called a nested structure, where a core group of generalist species interacts with each other and a group of specialist species interacts preferentially with the most generalist species. This structure of specialization is consistent with coevolutionary theory because natural selection on mutualistic interactions among free-living species can result in a coevolutionary vortex that draws more species into the interaction over time without breaking it up into many subwebs. Hence, natural selection often favors mutu-alists that interact with multiple other mutualists, and it appears to favor a few species that specialize on the core species within the mutualist network.
These mechanisms by which natural selection shapes large webs of coevolving species are among the least-understood aspects of coevolution. Studies of coevolving webs, however, are becoming an increasingly important part of evolutionary biology and ecology because they are central to understanding how coevolution has shaped the overall organization of biodiversity within and among ecosystems.
Human society has increasingly altered the coevolutionary process as we have transformed all the major ecosystems on Earth. The wholesale movement of species among continents by humans has created new interactions among species at a global scale unprecedented in the Earth’s history. Those species are now coevolving with native species, but we have little understanding of how these biological communities will stabilize over the coming centuries.
Even more directly, we have increasingly co-opted the coevolutionary process in our attempt to increase our food supplies and minimize diseases. Much of the development of agricultural crops has been driven by cycles of selective breeding for new varieties that are resistant to pests, followed by the evolution of new pests that can overcome that resistance, which is then followed by subsequent selective breeding for yet newer resistant varieties. These same kinds of cycles of human-induced coevolution are occurring in the development of new antibiotics against diseases that afflict humans and domesticated animals.
Some aspects of human-induced coevolution closely match those found in the natural process of coevolution, whereas others do not. We can, however, use studies of coevolution within natural populations as a guide for the development of new strategies to slow down our coevolutionary arms races with agricultural pests and human diseases. Hence, a science of applied coevolutionary biology is beginning to emerge and is likely to grow in importance as human society continues its attempt to manipulate the ecological structure of biodiversity worldwide.
Brodie, E. D. III, and E. D. Brodie, Jr. 1999. Predator–prey arms races. Bioscience 49: 557–568. This article presents a concise explanation of the complex ways in which natural selection shapes the evolution of defenses and counter-defenses through coevolution.
Bronstein, J. L., R. Alarcón, and M. Geber. 2006. The evolution of plant–insect mutualisms. New Phytologist 172: 412–428. This article presents an overview of some of the most common mutualisms found in terrestrial environments and the role of coevolution in shaping these interactions.
Pennisi, E. 2007. Variable evolution. Science 316: 686–687. A leading science writer interviews multiple coevolutionary biologists on our current understanding of the coevolutionary process.
Sotka, E. E. 2005. Local adaptation in host use in marine invertebrates. Ecology Letters 8: 448–459. This summary of studies discusses local adaptation and the potential for coevolution in a diverse group of marine species.
Thompson, J. N. 2005. The Geographic Mosaic of Coevolution. Chicago: University of Chicago Press. This volume presents a synthesis of current understanding of the coevolutionary process, including a wide range of examples of how coevolution continues to reshape interactions among species.