The Princeton Guide to Evolution

Evolution of Communities

Mark A. McPeek

OUTLINE

1. What are communities?

2. Microevolutionary change and community evolution

3. Macroevolutionary change and community evolution

4. Geography of speciation and extinction

Changes in abundances of species over time, combinations of species that can and cannot live together, and the number of species that can live together in one place at one time are all influenced by the abilities of each species to deal with the abiotic environment and to interact with the other species they encounter. These abilities are shaped by evolution, and so evolution is the foundational process shaping the properties of biological communities. The genetic diversity of one species can influence the outcomes of species interactions. Also, as species adapt to one another, they alter many aspects of the ecosystem, because of the change in their ability to influence their environment. Speciation and extinction have the most dramatic effects, because these processes introduce new species and eliminate existing species, respectively. In addition, because different modes of speciation generate varying amounts of ecological difference among species, the diverse modes of speciation can result in assorted properties in the resulting communities and can generate a variety of long-term outcomes.

GLOSSARY

Community. A collection of interacting or potentially interacting species on both local and regional spatial scales.

Ecological Speciation. The process of speciation resulting from the adaptation of two or more populations of a species to different ecological environments.

Extinction. The loss of a species from some biological system. The ultimate extinction of a species occurs when the last remaining individual of that species dies.

Functional Group. A group of species within a community that interact with other members of the community in very similar ways. A functional group has a unique ecological role within the community.

Speciation. The process of forming a new reproductively isolated species from preexisting species.

Species Interaction. The mechanism by which one species affects the population growth rate of another species.

Species Richness. The number of species present in a community.

1. WHAT ARE COMMUNITIES?

A biological community is a collection of species that live together on both local and regional spatial scales and whose interactions influence one another’s distributions and abundances. For example, all the species that live together in a pond form a biological community, and all the species that live in a forest form a community. Communities may be very small—the bacteria, protozoans, and insects inhabiting the pool in a pitcher plant leaf—or they may be very large—all the species living together in Lake Superior.

Interactions among the species within a community influence their distributions and abundances across the regional collection of communities where they are found. Species influence one another’s birth and death rates via interactions that can take many forms. For example, some species interact as predator and prey, or pathogen and host. Other species may compete for limiting resources (e.g., light, water, or mineral nutrients) or for biological food, as when two predators compete for a shared prey. Species may mutually benefit one another, as plants and their pollinators do, or they may alter the physical environment in ways that facilitate the performance of others, such as when they recycle inorganic nutrients from detritus into forms that others can then use.

These species interactions shape patterns of species distributions and abundances locally and regionally and also shape the properties of these communities, such as species richness and diversity. Species abundances may change over time owing to their interactions, such as when numbers of lynx and hare cycle relative to each other over multiple years. Also, not all species are capable of living together. For example, different prey species are found in communities depending on the presence or absence of a top predator in the system, because in the former case, prey are killed too quickly.

The performance of species in these interactions is determined by their phenotypes. The success of a prey in avoiding an attacking predator will depend on how its morphology, physiology, and behavior influence how fast it can flee, while the success of a food competitor may depend on how fast it can take up the resource or how well its physiology performs when the resource is scarce, to allow it to survive. As the phenotypes of species evolve as a consequence of these species interactions (see chapter III.15), the properties of the communities in which they live will necessarily also change.

Two fundamental questions regarding communities are: What is the source of all their interacting species? And what are the consequences of their loss from the system? New species enter communities via speciation or immigration from other areas, and they are lost via extinction. Therefore, both micro- and macroevolutionary processes have a profound influence on the properties of communities. Thus development of a fundamental understanding of today’s communities must include questions about how past and ongoing evolutionary events endowed species with their current properties.

2. MICROEVOLUTIONARY CHANGE AND COMMUNITY EVOLUTION

The processes of microevolution—mutation, genetic drift, gene flow, and natural and sexual selection—can all change the ecological capabilities of individual species by changing the phenotypes that influence their responses to the environment—including properties like temperature, water, and nutrient availability—as well as their interactions with other species.

The genetic diversity of species can have strong influences on the properties of communities if that genetic diversity is also reflected in phenotypic diversity that is ecologically important. For example, many plant species defend themselves by producing noxious secondary compounds that deter feeding by herbivorous insects. Different genotypes within a species may produce these compounds in varying concentrations and mixtures. Thus, different herbivores will be deterred from feeding on each genotype, and so the herbivorous arthropod assemblages may vary dramatically in species composition and abundance among genotypes of the same plant species. These effects can propagate to higher trophic levels as well if the predators of these herbivores differ in their abilities to exploit the particular herbivore assemblages found on different plant genotypes. Genetic diversity within species has also been shown to affect the likelihood that different plants can live with one another, because the genetic variation in traits of the species determines how they interact.

While mutation, gene flow, and genetic drift can alter the distributions of ecologically important traits in populations over long periods of time, natural selection is the most rapid and powerful evolutionary force shaping the abilities of species to interact with their environment. There is now abundant evidence that natural selection operates in natural populations and that interactions among species are a prime cause of that natural selection (see chapters III.15 and VI.7). Adaptive evolutionary responses in one species can therefore alter species abundances and other community properties throughout the system. For example, as a prey species evolves better defenses against a predator, the abundance of prey should increase because better defenses lower its mortality rate, and the abundance of its predator can be expected to decrease because it then obtains food at a slower rate. Additionally, if a predator evolves in its abilities to capture different types of prey, the abundances of those prey should change as a result. Likewise, if one competitor species evolves increased abilities to take up a limiting resource, its abundance should increase, and the numbers of other competitors for that resource should decrease.

Recent studies have begun to test and confirm such expectations. Studies of protozoan evolution in pitcher plant communities have demonstrated that population growth rates do evolve in response to predators and competitors, and indirect effects propagate throughout the community. For example, the population growth rate of a species of Colpoda, a ciliated protozoan, evolved to increase in response to both predation by pitcher plant mosquitoes (Wyeomyia smithii) and competitive interactions with other protozoan species.

The adaptation of Trinidadian guppies (Poecilia reticulata) to different predation regimes is an exemplar of evolution in the wild. Guppy populations in high predation areas of streams evolve to mature at smaller body sizes and to have higher reproductive effort than those in low predation areas. In artificial stream experiments, invertebrate biomass was higher and algal biomass was lower with guppies from high predation populations than with guppies from low predation populations. These differences mirrored differences in the various guppy diets. Guppies from the two areas also caused differences in various ecosystem properties, including alterations in rates of gross primary productivity, nitrogen flux, leaf decomposition, and standing amounts of benthic organic matter, and these differences mirrored those seen in natural streams.

The impact of alewife (Alosa pseudoharengus) on zooplankton assemblages in New England coastal lakes is a prime example of the way a top predator shapes community structure. Alewife are anadromous fish that return to coastal New England lakes to spawn, and as a consequence of their large body size, they feed selectively on large zooplankton, which causes shifts in zooplankton assemblages to favor small-bodied zooplankton species. Dams placed on some rivers by European settlers trapped some alewife populations in the freshwater lakes, and these landlocked populations have evolved much smaller body sizes. These differences in body size, and the concomitant differences in the size of their feeding apparatus, cause predictable differences in zooplankton assemblages between lakes with anadromous and landlocked alewife: small-bodied zooplankton species are found in lakes with anadromous alewife, and larger-bodied zooplankton species live in lakes with landlocked alewife.

Recent laboratory studies have also shown that adaptive evolutionary responses of species to one another can shape short-term population and community dynamics. Predator-prey interactions sometimes produce characteristic cycles in population abundances where predator abundance peaks one-quarter of a cycle behind the peak in prey abundance. For example, such cycles are seen in the historical records of lynx and hare abundances across northern North America. Modeling results show that when the prey and predators are allowed to rapidly evolve in response to each other, the peak abundances of both predator and prey occur farther apart in time and more asymmetrically in time relative to each other; these types of cycles cannot be generated without the evolutionary responses of predator and prey. Recent experiments in laboratory microcosms containing rotifer (Brachionus calyciflorus) predators and algal (Chlorella vulgaris) prey have demonstrated the validity of these model results. When only one or two genotypes of algae were present, and thus the algae could not evolve, predator and prey abundances rose and fell rapidly in cycles, and prey abundances peaked shortly before predator abundances in each cycle. However, when many genotypes of algae were present, and thus the algal population could evolve, population abundance peaks were significantly farther apart in time, and predator abundance peaked when prey abundance was at its nadir. Similar types of alterations to ecological dynamics caused by evolutionary responses of predators and prey have been demonstrated in laboratory experiments using bacterial prey (Escherichia coli) and predatory lytic bacteriophage viruses.

Studies of the evolutionary history of traits also provide windows into the evolution of communities through the changing abilities of species. For example, North American columbines (Aquilegia) have long spurs on their flowers for holding nectar. Pollinators must have long tongues to reach the nectar, and in the process of foraging for nectar they will inadvertently transfer pollen they have picked up from other flowers. Evolutionary reconstruction studies suggest that spur length evolves to be longer when a new pollinator with a longer tongue begins feeding on the species. In another compelling example, Dalechampia vine and scrub flower traits were shown to have evolved their pollinator reward system of resin secretions first as a defensive mechanism against herbivores; only later did these secretions become a pollinator reward. Conversely, others have hypothesized that some defenses against herbivores may have originated from traits that originally evolved to attract pollinators. These kinds of examples show why past evolutionary events are important for understanding present-day patterns in species interactions and community structure.

3. MACROEVOLUTIONARY CHANGE AND COMMUNITY EVOLUTION

While buildup of ecologically important genetic diversity within species and coevolutionary adaptation among species are certainly important in shaping community properties, the greatest changes to community structure must occur when new species are added to or lost from the community. The macroevolutionary processes of speciation and extinction can fundamentally change species richness and diversity as well as the functional types of species present on both local and regional scales.

Myriad processes can cause speciation (see chapters VI.3, VI.4, and VI.5), some modes of which will have profound effects on the ecological capabilities of the resulting species. Most important among these processes is ecological speciation, namely, the reproductive isolation between two or more lineages that results as a by-product of their ecological differentiation. Ecological population differentiation can occur along many different types of environmental axes and presumably occurs most frequently when lineages adapt to exploit unutilized ecological opportunities. Adaptive radiations are the most spectacular exemplars of ecological speciation (see chapter VI.10), but ecological speciation is a prevalent mechanism creating new species in most clades. Sticklebacks (Gasterosteus aculineatus) have been shown to speciate as a result of adapting to different environments (e.g., marine versus freshwater) and prey in different habitats (e.g., prey on the bottom of lakes versus in the water column). Rhagoletis flies and Timema walking sticks have each diversified to specialize in feeding on many different host plants. Anolis lizards have undergone repeated adaptive radiations to utilize different microhabitats of trees on different Caribbean islands. Enallagma and Lestes damselflies have diversified to live with different top predators (i.e., fish versus large dragonflies) in ponds and lakes across North America. Thus, ecological diversification appears to be a prevalent mode of speciation that is not restricted to any particular type of species interaction.

Most instances of ecological population differentiation probably do not result in new species. As one of the best examples of adaptive differentiation, Trinidadian guppy populations living with different predators show no signs of reproductive differentiation among the populations. However, in the case of ecological speciation, the prime consequence is the introduction of a new functional group to a community. For example, as marine sticklebacks colonized and adapted to living in freshwater lakes as the glaciers retreated 18,000 years ago, they established themselves as a new functional type of predator that fed on both zooplankton and benthic prey in the lakes. Subsequent speciation events via secondary invasions in a small subset of these lakes created two stickleback species: one specialized for feeding on zooplankton in the open water, and the other specialized for feeding on benthic invertebrates. In lakes with only one stickleback species, the dynamics of the benthic and zooplankton prey should be linked by the shared predation of the one stickleback species. Conversely, zooplankton and benthic prey dynamics might be largely decoupled in lakes with two stickleback species, because each has its own predator. These differences also propagate to other features of the ecosystem.

Likewise, ecological speciation in association with recent glacial cycles created three new Enallagma damselfly species as they colonized and adapted to ponds and lakes in which large dragonflies were the top predators (but fish predators were lacking). These speciation events introduced a new functional group to the dragonfly-lake community that was much better at avoiding dragonfly predators but poorer at competing for food than other damselfly genera already present. In effect, these ecological speciation events changed this component of the community from a linear food chain to a diamond-shaped food web (i.e., dragonflies as the top predator feeding on two intermediate-level consumer functional groups [the new Enallagma species, as well as the damselflies that were already present in the community], and the two consumer functional groups feeding on prey below them in the food web). Because energy and materials flow very differently through a linear as opposed to a diamond-shaped food web, the introduction of the Enallagma functional group to the community may have fundamentally altered material flows and the dynamics of community response to perturbations.

One of the central concepts of community ecology is species coexistence, namely, ecological differences among species that promote their long-term persistence together. Long-term stable coexistence requires that species be ecologically differentiated from one another such that each has a greater demographic effect on members of its own species than it does on other species; that is, the two species have “different niches.” The process of ecological speciation should typically produce species that immediately coexist on either local or regional scales, because the process of ecological differentiation that drives ecological speciation produces species that fill new ecological roles in the community.

Because ecological speciation fills unutilized ecological opportunities (i.e., empty niches), the rate of ecological speciation must also diminish as new species representing new functional groups are added, and so the rate of ecological speciation must depend on the number of species/functional groups already present in a community (see chapter VI.11). Extinction rates may be at some background level when species richness is low but then rapidly increase after all the possible functional groups are present or nearly so, and ecological interactions among species should drive more poorly adapted species extinct via competitive exclusion. As a community is assembled via ecological speciation, ecologically unique species should be added to the community at a diminishing rate until speciation and extinction rates balance, and this balance should exist at a point where nearly all the available niche space is filled. At this macroevolutionary equilibrium, species turnover will continue as species are replaced at existing functional positions in the community, but total species richness should change very little.

In contrast with ecological speciation, other modes of speciation (e.g., chromosomal rearrangements, changes in mate recognition, sexual selection, or sexual conflict) may result in reproductive isolation but little or no ecological differentiation among sister species. For example, evolutionary changes in the traits that males and females use to discriminate conspecific mates from heterospecifics can rapidly generate reproductive isolation among many different lineages simultaneously. However, changes in these traits (e.g., breeding coloration, mate calls, genitalia shapes, biochemical signals for gamete recognition, and compatibility) may have little or no consequences for how these species obtain resources, avoid predators, combat parasites, or interact with mutualists. These speciation modes do not introduce new functional groups into communities but rather add nearly equivalent species to preexisting functional groups of communities where their ranges overlap. Thus, taxa in which these types of speciation modes are most prevalent should display “neutral” community dynamics, in which species relative abundances change slowly and at random over time. Moreover, species richness will greatly exceed the number of available niches in the community. As a result, most species will be on a long, slow sojourn to extinction.

One example of this type of speciation is the recent radiations of the Enallagma damselflies. While ecological speciation has played an important role in Enallagma diversification to colonize different lake types, the majority of speciation events in the genus appear to have involved changes in the shapes of secondary sexual structures used by males and females to identify potential mates to species, with little to no change in ecologically important traits These speciation events add species to a given lake type (e.g., many new species added to lakes with fish) but do not change the way they interact with other species. As a result, 8 to 12 Enallagma species can be found living together in lakes across much of North America. Recent field experiments have shown that these co-occurring species are in fact ecologically indistinguishable. In addition, Enallagma relative abundances do not correlate with any of the major environmental gradients (e.g., predator or prey abundances, productivity abiotic conditions) among lakes and appear to vary randomly among lakes: these patterns are also the expected results for ecologically equivalent species.

Different macroevolutionary and macroecological dynamics are thus expected for communities containing taxa like Enallagma in which such nonecological modes of speciation dominate. Here, the per lineage speciation rate should be independent of the number of species already present. In addition, extinction rate may be substantially depressed, because the time to competitive exclusion increases nonlinearly as species become ecologically more similar (i.e., extinction rate is inversely proportional to the time to extinction). As a result, the macroevolutionary equilibrium species richness can be quite high. Recall, too, that because these species are ecologically nearly identical with one another, they would represent only one functional group embedded in a larger community; and within this functional group, the species relative abundances should vary randomly through time and change at a rate that is inversely proportional to the total number of individuals in all the equivalent species.

Still other modes of speciation (e.g., hybridization and polyploidization) may produce new species that may be ecologically quite different from their progenitors, but these ecological differences may not coincide with the distribution of available niches in the community, as they do with ecological speciation. For example, hybridization between two plant species will produce a new species that can be phenotypically, and thus ecologically, quite different from both parents, but the phenotype of the new species is produced without respect to the ecological opportunities available in the community. Most new species produced by these mechanisms are probably quickly driven extinct because they have ecologically inferior phenotypes compared with those of their progenitors. However, some will stumble onto superior phenotypes that may allow them to coexist with their progenitors or to invade new ecological conditions or even to replace one of the progenitors.

These speciation modes may produce a third type of macroevolutionary and macroecological dynamic: because species must interact in hybridization and polyploidization modes of speciation, the per lineage speciation rate may actually increase with species richness. But since species will be ecologically fairly different, extinction rate is still expected to increase with species richness. However, species richness is then expected to come to equilibrium at some value above the number of species that can coexist with one another, but the rate of species turnover in the system may be quite high, because speciation and extinction rates equilibrate at high values of each.

4. GEOGRAPHY OF SPECIATION AND EXTINCTION

The geography of speciation and extinction also influences the evolution of community structure (see Section III: Natural Selection and Adaptation). In particular, in addition to the mechanisms that generate reproductive isolation, speciation events are typically classified according to the geographic structure of the differentiating lineages: allopatric, parapatric, peripatric, and sympatric. The geographies of speciation and extinction are important because they define the spatial scale at which new species are added to communities.

Because the differentiating lineages are spatially segregated, allopatric, parapatric, and peripatric speciation events do not add new species to local communities. Each lineage must already be embedded in a local community, and speciation results when these geographically distinct sets of populations differentiate from one another. As with the effects of guppy evolution on different predation regimes, the local evolutionary forces that drive differentiation may consequently alter the local properties of communities. The regional species richness pool will increase, and regional functional diversity will also increase if speciation is driven by ecological differentiation. If one species range (or both) subsequently expands into the other’s range, local richness and functional diversity may increase if the two species can live together in the same local community.

Conversely, sympatric speciation events do increase the species richness of a local community. The shift of Rhagoletis flies from hawthorn trees to feed and develop on apple trees has increased the number and types of herbivore species in forest and orchard communities of North America in the last 150 years. Similar shifts have presumably occurred to create the diversity of Rhagoletis species that are found feeding on trees and shrubs with fleshy fruits across eastern North America. Speciation via hybridization and polyploidy similarly must introduce new species locally, since the daughter species of the process are offspring of parental species individuals.

However, many speciation events do not fit neatly along the classic allopatric to sympatric continuum, particularly those involving ecological speciation caused by habitat shifts. Habitat shifts imply the invasion of a local community by a new functional species type, and so local species richness and functional diversity increases. However, habitats are typically distributed across the landscape in a variegated pattern, and so a speciation event via a habitat shift may occur within the geographic range of the parental species. For example, lakes with fish and lakes with dragonflies as top predators are interspersed across the landscape. In eastern North America, these two lake types have unique assemblages of Enallagma species, with each lake constituting a local community. The ancestral lake type for all Enallagma was a fish-lake species. Thus, the ecological speciation events within this genus must have been initiated by females of a fish-lake species laying eggs to create a founder population in one or more fishless lakes where large dragonflies were the top predator. Because all these species have broadly overlapping ranges today, these speciation events must have created new species within the ranges of the progenitors (sympatric speciation on a larger spatial scale), but because of their habitat differences, the resulting species are allopatric on a local scale.

Extinction can similarly have a spatial dimension. The presence of one species may drive others extinct on a local scale but have little effect on the much broader regional scales. For example, when fish are introduced to a previously fishless lake, the entire collection of Enallagma species in the lake are driven extinct (along with many other invertebrate taxa) and are eventually replaced by the collection of Enallagma species that can persist with fish. Such local extinction events occur routinely and are presumably the basis for metapopulation dynamics. Extinction of an entire species then requires that it become locally extinct in all the places where it could formerly support a population. It is a simple mathematical fact that species will become more susceptible to extinction as they become less abundant locally and as they are able to support populations in fewer local communities. Species extinctions are thus much more likely to occur in rare species and to be caused by some factor that influences an entire region.

As the fossil record shows, species have continually been added and lost from biological systems over the history of life on earth (see chapter VI.13). This reality implies an evolutionary dynamism for biological communities that is typically not contemplated. Change in the ecosystem is a result not only of change in the abiotic world but also because of the addition of new species and the loss or alteration of properties of existing species. The evolution of species and higher taxa is a major force for change in ecosystems.