Why It Matters. . .

In some open woodlands in Central America, flocks of chestnut-headed oropendolas (Psarocolius wagleri), members of the blackbird family, build hanging nests in isolated trees (Figure 51.1). Female giant cowbirds (Molothrus oryzivorus) often bully their way into a colony, laying an egg or two in each oropendola nest. Cowbirds are brood parasites on oropendolas, tricking them into caring for cowbird young. The cowbird chicks grow faster than oropendola chicks, and they consume much of the food that the oropendolas bring to their own offspring. Because cowbird chicks take food away from their oropendola nest mates, we might expect adult oropendolas to eject cowbird eggs and chicks from their nests—but often they don't.

Why do some oropendolas care for offspring that are not their own? In an ingenious study conducted in the 1960s, Neal Smith of the Smithsonian Tropical Research Institute determined that cowbird chicks could actually increase the number of offspring that some oropendolas raise. Oropendola chicks are frequently parasitized by botfly larvae, which feed on their flesh. The aggressive cowbird chicks snap at adult botflies and pick fly larvae off their nest mates. Although cowbird chicks eat food meant for oropendola chicks, they also protect them from potentially lethal parasites; twice as many young oropendolas survive in nests with cowbird chicks as in nests without them.

In other areas of Central America, oropendolas build nests near the hives of bees or wasps. These oropendolas chase cowbirds from their colonies, and when a cowbird does manage to sneak an egg into one of their nests, the oropendolas frequently eject it. Why do oropendolas in these colonies reject cowbird eggs? Smith determined that the swarms of bees and wasps keep botflies away from the oropendola colonies. At these sites, twice as many oropendola chicks survive in nests without cowbirds as in those that include them. Thus the oropendolas derive no benefit from having cowbird chicks in their nests, and natural selection has favored discriminating behavior in oropendolas that nest near bees and wasps.

The story of the oropendolas, cowbirds, botflies, bees, and wasps provides an example of the population interactions that characterize life in an ecological community, an assemblage of species living in the same place. And as this story reveals, the presence or absence of certain species may alter the effects of such interactions in almost unimaginably complex ways. We begin this chapter with a description of some of the many ways that populations in a community interact. We then examine how population interactions and other factors, such as the kinds of species present and the relative numbers of each species, influence a community's characteristics.

Population Interactions

Population interactions usually provide benefits or cause harm to the organisms engaged in the interaction (Table 51.1). And because interactions with other species often affect the survival and reproduction of individuals, many of the relationships that we witness today are the products of long-term evolutionary modification. Before examining several general types of population interactions, we briefly consider how natural selection has shaped the relationships between interacting species.

Population interactions change constantly. New adaptations that evolve in one species exert selection pressure on another, which then evolves adaptations that exert selection pressure on the first. The evolution of genetically based, reciprocal adaptations in two or more interacting species is described as coevolution.

Some coevolutionary relationships are straightforward. For example, ecologists describe the coevolutionary interactions between some predators and their prey as a race in which each species evolves adaptations that temporarily allow it to outpace the other. When antelope populations suffer predation by cheetahs, natural selection fosters the evolution of faster speed in the antelopes. Cheetahs then experience selection for increased speed so that they can overtake and capture antelopes. Other coevolved interactions provide benefits to both partners. For example, the flower structures of different monkey-flower species have evolved characteristics that allow them to be visited by either bees or hummingbirds (see Figure 21.7).

Although one can hypothesize a coevolutionary relationship between any two interacting species, documenting the evolution of reciprocal adaptations is difficult. As our introductory story about oropendolas and their parasites illustrated, coevolutionary interactions often involve more than two species. Indeed, most organisms experience complex interactions with numerous other species in their communities, and the simple portrayal of coevolution as taking place between two species rarely does justice to the complexity of these relationships.

Predation and Herbivory Define Many Relationships in Ecological Communities

Because animals acquire nutrients and energy by consuming other organisms, predation (the interaction between predatory animals and the animal prey they consume) and herbivory (the interaction between herbivorous animals and the plants they eat) are often the most conspicuous relationships in ecological communities.

Adaptations for Feeding

Both predators and herbivores have evolved remarkable characteristics that allow them to feed effectively. Carnivores use sensory systems to locate animal prey and specialized behaviors and anatomical structures to capture and consume it. For example, a rattlesnake (genus Crotalus) uses heat sensors on its head (see Figure 39.22) and chemical sensors in the roof of its mouth to find rats or other endothermic prey. Its hollow fangs inject toxins that kill the prey and begin to digest its tissues even before the snake consumes it. And elastic ligaments connecting the bones of its jaws and skull allow a snake to swallow prey that is larger than its head. Herbivores have comparable adaptations for locating and processing their food plants. Insects use chemical sensors on their legs and heads to identify edible plants and sharp mandibles or sucking mouthparts to consume plant tissues or sap. Herbivorous mammals have specialized teeth to harvest and grind tough vegetation (see Section 45-5).

All animals must select their diets from a variety of potential food items. Some species, described as specialists, feed on one or just a few types of food. Among birds, for example, the Everglades kite (Rostrhamus sociabilis) consumes just one prey species, the apple snail (Pomacea paludosa). Other species, described as generalists, have broader tastes. Crows (genus Corvus) consume food ranging from grains to insects to carrion.

How does an animal select what type of food to eat? Some mathematical models, collectively described as optimal foraging theory, predict that an animal's diet is a compromise between the costs and benefits associated with different types of food. Assuming that animals try to maximize their energy intake in a given feeding time, their diets should be determined by the time and energy it takes to pursue, capture, and consume a particular kind of food compared with the energy that food provides. For example, a cougar (Puma concolor) will invest more time and energy hunting a mountain goat (Oreamnos americanus) than a jackrabbit (Lepus townsendii), but the payoff for the cougar is a bigger meal.

Food abundance also affects food choice. When prey are scarce, animals often take what they can get, settling for food that has a low benefit-to-cost ratio. But when food is abundant, they may specialize, selecting types that provide the largest energetic return. Bluegill sunfish (Lepomis macrochirus), for example, feed on Daphnia and other small crustaceans. When crustacean density is high, the fish hunt mostly large Daphnia, which provide more energy for their effort; but when prey density is low, bluegills feed on Daphnia of all sizes (Figure 51.2).

Defenses against Herbivory and Predation

Because herbivory and predation have a negative impact on the organisms being consumed, plants and animals have evolved mechanisms to avoid being eaten. Some plants use spines, thorns, and irritating hairs to protect themselves from herbivores. Many plant tissues also contain poisonous chemicals that deter herbivores from feeding. For example, plants in the milkweed family (Asclepiadaceae) exude a milky, irritating sap that contains cardiac glycosides, even small amounts of which are toxic to vertebrate heart muscle. Other compounds mimic the structure of insect hormones, disrupting the development of insects that consume them. Most of these poisonous compounds are volatile, giving plants their typical aromas; some herbivores have coevolved the ability to recognize these odors and avoid the toxic plants. Recent research indicates that some plants increase their production of toxic compounds in response to herbivore feeding. For example, potato and tomato plants that have been damaged by herbivores produce higher levels of protease-inhibiting chemicals; these compounds prevent herbivores from digesting proteins they have just consumed, reducing the food value of these plant tissues.

Many animals have evolved an appearance that provides a passive defense against predation (Figure 51.3). Caterpillars that look like bird droppings, for example, may not attract much attention from a hungry predator. And as you learned in Chapter 1 (see Figure 1.9), cryptic coloration helps some prey (as well as some predators) to blend in with their surroundings.

Once discovered by a predator, many animals first try to run away. When cornered, they may try to startle or intimidate the predator with a display that increases their apparent size or ferocity (Figure 51.4). Such a display might confuse the predator just long enough to allow the potential victim to escape. Other species seek shelter in protected sites. For example, flexible-shelled African pancake tortoises (Malacochersus tornieri) retreat into rocky crevices and puff themselves up with air, becoming so tightly wedged between rocks that predators cannot extract them.

Other animals defend themselves actively. North American porcupines (genus Erethizon) release hairs modified into sharp, barbed quills that stick in a predator's mouth, causing severe pain and swelling. Other species fight back by biting, charging, or kicking an attacking predator. Chemical defenses also provide effective protection. Skunks release a noxious spray when threatened, and some frogs and toads produce neurotoxic skin secretions that paralyze and kill mammals. Some insects even protect themselves with poisons acquired from plants. The caterpillars of monarch butterflies (Danaus plexippus) are immune to the cardiac glycosides in the milkweed leaves they eat. They store these chemicals at high concentration, even through metamorphosis, making adult monarchs poisonous to vertebrate predators.

Poisonous or repellant species often advertise their unpalatability with bright, contrasting patterns, called aposematic coloration (Figure 51.5). Although a predator might attack a black-and-white skunk, a yellow-banded wasp, or an orange monarch butterfly once, it quickly learns to associate the gaudy color pattern with pain, illness, or severe indigestion—and rarely attacks these easily recognized animals again.

Mimicry, in which one species evolves an appearance resembling that of another (Figure 51.6), is also a form of defense. In Batesian mimicry, named for English naturalist Henry W. Bates, a palatable or harmless species, the mimic, resembles an unpalatable or poisonous one, the model. Any predator that eats the poisonous model will subsequently avoid other organisms that resemble it. In Müllerian mimicry, named for German zoologist Fritz Müller, two or more unpalatable species share a similar appearance, which reinforces the lesson learned by a predator that attacks any species in the mimicry complex.Despite the effectiveness of many antipredator defenses, coevolution has often molded the responses of predators to overcome them. For example, when threatened by a predator, the pinacate beetle (Eleodes longicollis) raises its rear end and sprays a noxious chemical from a gland at the tip of its abdomen. Although this behavior deters many would-be predators, grasshopper mice (genus Onychomys) of the American southwest circumvent this defense: they grab the beetles and shove their abdomens into the ground, rendering the beetle's spray ineffective (Figure 51.7).

Interspecific Competition Occurs When Different Species Depend on the Same Limiting Resources

Populations of different species often use the same limiting resources, causing interspecific competition (competition between species). The competing populations may experience increased mortality and decreased reproduction, responses that are similar to the effects of intraspecific competition (see Section 50-4). Interspecific competition reduces the size and population growth rate of one or more of the competing populations.

Community ecologists identify two main forms of interspecific competition. In interference competition, individuals of one species harm individuals of another species directly. Animals may fight for access to resources, as when lions chase smaller scavengers like hyenas and jackals from their kills. Similarly, many plant species, including creosote bushes (see Figure 50.4), release toxic chemicals that prevent other plants from growing nearby. In exploitative competition, two or more populations use ("exploit") the same limiting resource. The presence of one species reduces resource availability for the others, even in the absence of snout-to-snout or root-to-root confrontations. For example, in the deserts of the American Southwest, many bird and ant species feed largely on seeds. Thus, each seed-eating species may deplete the food supply available to others.

Competitive Exclusion and the Niche Concept

In the 1920s, the Russian mathematician Alfred J. Lotka and the Italian biologist Vito Volterra independently proposed a model of interspecific competition, modifying the logistic equation (see Section 50-4) to describe the effects of competition between two species. In their model, an increase in the size of one population reduces the population growth rate (r) of the other.

A Russian biologist, G. F. Gause, tested the model experimentally in the 1930s. He grew cultures of two Paramecium species (ciliate protists) under constant laboratory conditions, regularly renewing food and removing wastes. Both species fed on bacteria suspended in the culture medium. When grown alone, each species exhibited logistic growth. When grown together in the same dish, however, Paramecium aurelia persisted at high density, but Paramecium caudatum was nearly eliminated (Figure 51.8). These results inspired Gause to define the competitive exclusion principle: populations of two or more species cannot coexist indefinitely if they rely on the same limiting resources and exploit them in the same way. One species is inevitably more successful, harvesting resources more efficiently and producing more offspring than the other.

Ecologists developed the concept of the ecological niche as a tool for visualizing resource use and the potential for interspecific competition in nature. We define a population's niche by the resources it uses and the environmental conditions it requires over its lifetime. In this context, the niche includes food, shelter, and nutrients as well as abiotic conditions, such as light intensity and temperature, which cannot be depleted. In theory, one could identify an almost infinite variety of conditions and resources that contribute to a population's niche. In practice, ecologists usually analyze a few critical resources for which populations might compete. Sunlight, soil moisture, and inorganic nutrients are important resources for plants. Food type, food size, and nesting sites are important for animals.

Ecologists distinguish the fundamental niche of a population, the range of conditions and resources that it can possibly tolerate and use, from its realized niche, the range of conditions and resources that it actually uses in nature. Realized niches are smaller than fundamental niches, partly because all tolerable conditions are not always present in a habitat, and partly because some resources are used by other species. We can visualize competition between two populations by plotting their fundamental and realized niches with respect to one or more resources (Figure 51.9). If the fundamental niches of two populations overlap, they might compete in nature.

Evaluating Competition in Nature

The observation that several populations use the same resource does not demonstrate that competition occurs. For example, all terrestrial animals consume oxygen, but they don't compete for oxygen because it is usually plentiful. Nevertheless, two general observations provide indirect evidence that interspecific competition may have important effects. The first is the extremely common observation of resource partitioning, the use of different resources or the use of resources in different ways, by species living in the same place. For example, weedy plants might compete for water and dissolved nutrients in abandoned fields. But they avoid competition by partitioning these resources, collecting them from different depths in the soil (Figure 51.10).

A second phenomenon that suggests the importance of competition is observed in comparisons of species that are sometimes sympatric (that is, living in the same place) and sometimes allopatric (that is, living in different places). In several studies of animals, researchers have documented character displacement: allopatric populations are morphologically similar and use similar resources, but sympatric populations are morphologically different and use different resources. The differences between the sympatric populations allow them to coexist without competing. Differences in bill size among sympatric finch species on the Galápagos Islands (see Sections 19-2 and 20-3) may be the product of character displacement (Figure 51.11).Data on resource partitioning and character displacement merely suggest the possible importance of interspecific competition in nature. To demonstrate conclusively that interspecific competition limits natural populations, one must show that the presence of one population reduces the population size or distribution of its presumed competitor. In a classic field experiment, Joseph Connell of the University of California, Santa Barbara, determined that competition between two barnacle species caused the realized niche of one species to be smaller than its fundamental niche (Figure 51.12).

Connell first observed the distributions of barnacles in undisturbed habitats. Chthamalus stellatus is generally found in shallow water on rocky coasts, where it is periodically exposed to air. Balanus balanoides typically lives in deeper water, where it is usually submerged.

Connell determined the fundamental niche of each species by removing either Chthamalus or Balanus from rocks and monitoring the distribution of each species in the absence of the other. When Connell removed Balanus from rocks in deep water, larval Chthamalus colonized the area and produced a flourishing population of adults. Connell had observed that Balanus physically displaced Chthamalus from these rocks. Thus, interference competition from Balanus prevents Chthamalus from occupying areas where it would otherwise live. By contrast, the removal of Chthamalus from rocks in shallow water did not result in colonization by Balanus. Balanus is apparently unable to live in habitats that are frequently exposed to air. Connell therefore concluded that competition from Chthamalus does not affect the distribution of Balanus. Thus, the competitive interaction between these two species is asymmetrical: Balanus has a substantial effect on Chthamalus, but Chthamalus has virtually no effect on Balanus.

In Symbiotic Associations, the Lives of Two or More Species Are Closely Intertwined

Some species have a physically close ecological association called symbiosis (sym = together, bio = life, sis = condition). Biologists define three types of symbiotic interactions—commensalism, mutualism, and parasitism—that differ in their effects.

Commensalism, in which one species benefits and the other is unaffected, is rare in nature, because few species are unaffected by their interactions with another. One possible example is the relationship between cattle egrets (Bubulcus ibis), birds in the heron family, and the large grazing mammals with which they associate (Figure 51.13). Cattle egrets feed on insects and other small animals that their commensal partners flush from grass. Feeding rates of egrets are higher when they associate with large grazers than when they do not. The birds clearly benefit from this interaction, but the presence of birds has no apparent positive or negative impact on the mammals.

Mutualism, in which both partners benefit, is extremely common. The coevolved relationships between flowering plants and animal pollinators are largely mutualistic (see Figure 27.32). Animals that feed on a plant's nectar or pollen carry its gametes from one flower to another (Figure 51.14). Similarly, animals that eat the fruits of flowering plants disperse the seeds, "planting" them in a pile of nutrient-rich feces. These mutualistic relationships between plants and animals do not require active cooperation. Each species simply exploits the other for its own benefit.

Some associations between either bacteria or fungi and plants are also mutualistic. For example, mycorrhizae are fungi that grow alongside the roots of many plant species. These fungi facilitate the plants' uptake of nitrogen and phosphorus from the soil, and the plants provide the fungi with carbohydrates in return. Another important mutualism is the close relationship between the nitrogen-fixing bacterium Rhizobium and leguminous plants, such as peas, beans, and clover (see Section 33-3).

Mutualistic relationships between animal species are also common. For example, some small marine fishes feed on parasites that attach to the mouths and gills of large predatory fishes (Figure 51.15). Parasitized fishes hover motionless while the "cleaners" scour their tissues. The relationship is mutualistic because the cleaner fishes get a meal, and the larger fishes are relieved of parasites.

The relationship between the bullhorn acacia tree (Acacia cornigera) of Central America and a small ant species (Pseudomyrmex ferruginea) is one of the most highly coevolved mutualisms known (Figure 51.16). Each acacia is inhabited by an ant colony that lives in the tree's swollen thorns. The ants swarm out of the thorns to sting—and sometimes kill—herbivores that touch the tree. The ants also clip any vegetation that grows nearby. Thus, acacia trees that are colonized by ants grow in a space free of herbivores and competitors, and occupied trees grow faster and produce more seeds than unoccupied trees. In return, the plants produce sugar-rich nectar consumed by adult ants and protein-rich structures that the ants feed to their larvae. Ecologists describe the coevolved mutualism between these species as obligatory, at least for the ants; they cannot subsist on any other food sources.

Parasitism is a type of interaction in which one species, the parasite, uses another, the host, in a way that is harmful to the host. Parasite–host relationships are like predator–prey relationships: one population of organisms feeds on another. But parasites rarely kill their hosts quickly because a dead host is useless as a continuing source of nourishment.

Tapeworms and other parasites that live within a host are endoparasites. Many endoparasites acquire their hosts passively, when a host accidentally ingests the parasite's eggs or larvae (see Focus on Applied Research, Chapter 29). Endoparasites generally complete their life cycle in one or two host individuals. By contrast, leeches, aphids, mosquitoes, and other parasites that feed on the exterior of a host are ectoparasites. Most animal ectoparasites have elaborate sensory and behavioral mechanisms that allow them to locate specific hosts, and they feed on numerous host individuals during their lifetimes. Some plants, such as mistletoes (genus Phoradendron), live as ectoparasites on the trunks and branches of trees; their roots penetrate the host's xylem and extract water and nutrients.

Not all parasites feed directly on a host's tissues. The giant cowbirds described earlier are brood parasites, as are other species of cowbirds and cuckoos. Although oropendolas sometimes benefit from the presence of cowbirds, most brood parasites have negative effects on their hosts. For example, brood parasitism by the brown-headed cowbird (Molothrus ater) has played a large role in the near-extinction of Kirtland's warbler (Dendroica kirtlandii).

The feeding habits of some insects, called parasitoids, fall somewhere between true parasitism and predation. A female parasitoid lays eggs in the larva or pupa of another insect species, and her young consume the tissues of the living host. Because the hosts chosen by most parasitoids are highly specific, agricultural ecologists often release parasitoids to control populations of insect pests.

Study Break 51–1

1. Why are some carnivores willing to spend more time and energy capturing large prey than small prey?
2. What are the differences between cryptic coloration, aposematic coloration, and mimicry? Can a mimic ever have aposematic coloration?
3. How can field experiments demonstrate conclusively that two species compete for limiting resources?

Think Outside the Book

Using the terms and concepts introduced in this chapter, describe the interactions that humans have with ten other species. Try to pick at least eight species that we do not eat.

The Nature of Ecological Communities

Ecologists have often debated the nature of ecological communities, asking if they have emergent properties that transcend the interactions among the populations they contain.

Most Ecological Communities Blend into Neighboring Communities

How do complex population interactions affect the organization and functioning of ecological communities? In the 1920s, ecologists in the United States developed two extreme hypotheses about the nature of ecological communities. Frederic Clements of the University of Minnesota championed an interactive view of communities. He described communities as "superorganisms," assemblages of species bound together by complex population interactions. According to this view, each species in a community requires interactions with a set of ecologically different species, just as every cell in an organism requires services that other types of cells provide. Clements believed that once a mature community was established, its species composition—the particular combination of species that occupy the site—was at equilibrium. If a fire or some other environmental factor disturbed the community, it would return to its predisturbance state.

Henry A. Gleason of the University of Michigan proposed an alternative, individualistic view of ecological communities. He believed that population interactions do not always determine species composition. Instead, a community is just an assemblage of species that are individually adapted to similar environmental conditions. According to Gleason's hypothesis, communities do not achieve equilibrium; rather, they constantly change in response to disturbance and environmental variation.

In the 1960s, Robert Whittaker of Cornell University suggested that ecologists could determine which hypothesis was correct by analyzing communities along environmental gradients, such as temperature or moisture (Figure 51.17). According to Clements' interactive hypothesis, species that typically occupy the same communities should always occur together. Thus, their distributions along the gradient would be clustered in discrete groups with sharp boundaries between groups (see Figure 51.17A). According to Gleason's individualistic hypothesis, each species is distributed over the section of an environmental gradient to which it is adapted. Different species would have unique distributions, and species composition would change continuously along the gradient. In other words, communities would not be separated by sharp boundaries (see Figure 51.17B).

Most gradient analyses support Gleason's individualistic view of ecological communities. Environmental conditions vary continuously in space, and most plant distributions match these patterns (see Figure 51.17C, D). Species occur together in assemblages because they are adapted to similar conditions, and the species compositions of the assemblages change gradually across environmental gradients.

Nevertheless, the individualistic view does not fully explain all patterns observed in nature. Ecologists recognize certain assemblages of species as distinctive communities and name them accordingly—redwood forests and coral reefs are good examples. But the borders between adjacent communities are often wide transition zones, called ecotones. Ecotones are generally rich with species because they include plants and animals from both neighboring communities as well as some species that thrive only under transitional conditions. In some places, however, a discontinuity in a critical resource or some important abiotic factor produces a sharp community boundary. For example, chemical differences between soils derived from serpentine rock and sandstone establish sharp boundaries between communities of native California wildflowers and introduced European grasses (Figure 51.18).

Study Break 51–2

1. Which view of communities suggests that they are just chance assemblages of species that happen to be adapted to similar abiotic environmental conditions?
2. Why would you often find more species living in an ecotone than you would in the communities on either side of it?

   Community Characteristics

   Although the species composition of an ecological community may vary somewhat over geographical gradients, every community has certain characteristics that define its overall appearance and structure.
   

The Growth Forms of Plants Establish a Community's Overall Appearance

The growth forms of plants—their sizes and shapes—vary markedly in different environments. Warm, moist environments support complex vegetation with multiple vertical layers. For example, tropical forests include a canopy, formed by the tallest trees; an understory of shorter trees and shrubs; an herb layer under openings in the canopy; vinelike lianas; and epiphytes, which grow on the trunks and branches of trees (Figure 51.19). By contrast, physically harsh environments are occupied by low vegetation with simple structure. For example, trees on mountaintops buffeted by cold winds are short, and the plants below them cling to rocks and soil. Other environments support growth forms between these extremes (see Chapter 49).

Foundation Species Moderate the Abiotic Environment within a Community

In many ecological communities, one common species can function as a foundation species, defining the nature of a community by creating locally stable environmental conditions. For example, trees are the foundation species in forested ecosystems, because their form defines the physical structure of the community, and their leaves and branches moderate short-term fluctuations in abiotic environmental factors like temperature, runoff from rainfall, and wind speed.

Saltmarsh cordgrass (Spartina alterniflora) is a foundation species in the wetlands surrounding Narragansett Bay, Rhode Island, because patches of this meter-high grass slow the velocity of the incoming tide and stabilize the stony beach habitat along the shore. In the absence of Spartina, tidal surges move the stones on the beach, disrupting the germination and growth of several small, herbaceous plant species. John F. Bruno of Brown University surveyed the plants growing adjacent to more than 350 Spartina patches of varying size. His research revealed that large patches of Spartina are more effective than small patches in moderating tidal effects. Not surprisingly, the percentage of a stony beach occupied by herbaceous plants was directly proportional to the length of the Spartina patch that bordered the beach (Figure 51.20).

Communities Differ in Species Richness and the Relative Abundance of Species

Communities differ greatly in their species richness, the number of species that live within them. For example, the harsh environment on a low desert island may support just a few species of microorganisms, fungi, algae (photosynthetic protists), plants, and arthropods. By contrast, tropical forests, which grow under milder physical conditions, include many thousands of species. Ecologists have studied global patterns of species richness (described below in Section 51-7) for decades. Today, as human disturbance of natural communities has already reached a tipping point, conservation biologists focus on such studies to determine which regions of Earth are most in need of preservation (see Chapter 53).

Within every community, populations differ in their commonness or the relative abundance of individuals. Some communities have just one or two dominant species, which represent a majority of the individuals present, as well as a number of rare species, each represented by just a few individuals. In other communities, species are represented by more equal numbers of individuals. For example, in a temperate deciduous forest in West Virginia, tulip poplar (Liriodendron tulipifera) and sassafras (Sassafras albidum) are dominant, accounting for nearly 85% of the trees. By contrast, a tropical forest in Costa Rica may include more than 200 tree species, each making up only a small percentage of the total.

Species richness and relative abundance together contribute to a community characteristic that ecologists call species diversity. To demonstrate the concept of species diversity, we will compare three hypothetical forest communities (Figure 51.21). Two of the communities include 50 trees distributed among 10 species. In Forest A, the dominant species is represented by 39 individuals, two species by two individuals each, and seven species by one individual each. In Forest B, each of the 10 species is represented by five individuals. Although both communities have the same species richness (10 species), Forest A is less diverse than Forest B, because more than three-quarters of its trees are of the same species. The third forest has only two tree species (Forest C in Figure 51.21); it is less diverse than either of the others.

Feeding Relationships within a Community Determine Its Trophic Structure

All ecological communities, regardless of their species richness, also have a trophic structure (trophe = nourishment) that comprises all of the plant–herbivore, predator–prey, host–parasite, and potential competitive interactions (Figure 51.22).

Trophic Levels

We can visualize the trophic structure of a community as a hierarchy of trophic levels, defined by the feeding relationships among its species (see Figure 51.22A). Photosynthetic organisms are the primary producers, the first trophic level. Primary producers are often described as autotrophs (auto = self) because they capture sunlight and convert it into chemical energy, using simple inorganic molecules acquired from the environment to build larger organic molecules that other organisms can use. Plants are the dominant primary producers in terrestrial communities. Multicellular algae (macroalgae) and plants are the major primary producers in shallow freshwater and marine environments. Photosynthetic or chemosynthetic bacteria, cyanobacteria, and "protists" are the primary producers in deep, open water.

Animals, by contrast, are consumers. Herbivores, which feed directly on producers, form the second trophic level, the primary consumers. Carnivores that feed on herbivores are the third trophic level, or secondary consumers, and carnivores that feed on other carnivores form the fourth trophic level, the tertiary consumers. For example, songbirds feeding on herbivorous insects are secondary consumers, and falcons feeding on songbirds are tertiary consumers. Some organisms, like humans and some bears, are omnivores, feeding at several trophic levels simultaneously.

A separate and distinct trophic level includes organisms that extract energy from the organic detritus (refuse) produced at other trophic levels. Scavengers, or detritivores, are animals such as earthworms and vultures that ingest dead organisms, digestive wastes, and cast-off body parts such as leaves and exoskeletons. Decomposers are small organisms, such as bacteria and fungi, that feed on dead or dying organic material. As described in Chapter 52, detritivores and decomposers serve a critical ecological function because their activity reduces organic material to small inorganic molecules that producers can assimilate.

All of the consumers in a community—the animals, fungi, and diverse microorganisms—are described as heterotrophs (hetero = other) because they acquire energy and nutrients by eating other organisms or their remains.

Food Chains and Webs

Ecologists depict the trophic structure of a community in a food chain, a portrait of who eats whom. Each link in a food chain is represented by an arrow pointing from the food to the consumer. Simple, straight-line food chains are rare in nature because most consumers feed on more than one type of food, and because most organisms are eaten by more than one type of consumer. These complex relationships are portrayed as a food web, a set of interconnected food chains with multiple links.

In the food web for the waters off the coast of Antarctica (see Figure 51.22B), most organisms at the bottom of the food web are tiny, and they occur in vast numbers. Huge pastures of phytoplankton (microscopic algae and diatoms) are responsible for most photosynthesis. They are consumed by herbivorous zooplankton (some protists, copepods, and shrimplike krill), which are in turn eaten by larger species, such as carnivorous zooplankton, squids, fishes, and suspension-feeding baleen whales. Some of these secondary consumers are themselves eaten by birds and mammals at higher trophic levels. The top carnivore in this ecosystem, the orca (Orcinus orca), feeds on carnivorous birds and mammals. As you can see in Figure 51.22B, ecological relationships within a food web are complex because many species feed at more than one trophic level.

Food-Web Analysis

In the late 1950s, Robert MacArthur of Princeton University pioneered the analysis of food webs to determine how the many links between trophic levels may contribute to a community's stability—its ability to maintain its species composition and relative abundances when environmental disturbances eliminate some species from the community. MacArthur hypothesized that in species-rich communities, where animals feed on many food sources, the absence of one or two species would have only minor effects on the structure and stability of the community as a whole. He therefore proposed a connection between species diversity, food-web complexity, and community stability.

Recent research has confirmed MacArthur's reasoning. For example, the average number of links per species generally increases with increasing species richness. Comparative food-web analyses also reveal that the relative proportions of species at the highest, middle, and lowest trophic levels are reasonably constant across communities. When researchers compared the number of prey species to the number of predator species in food webs from 92 communities of freshwater invertebrates, they discovered that, regardless of species richness, a community includes between two and three prey species for every predator species.

Interactions among species in a food web are often complex, indirect, and hard to unravel. In desert communities of the American Southwest, for example, rodents and ants potentially compete for seeds, their main food source. And the plants that produce the seeds compete for water, nutrients, and space. Rodents generally prefer to eat large seeds, but ants prefer small seeds. Thus, feeding by rodents reduces the potential population sizes of plants that produce large seeds. As a result, the population sizes of plants that produce small seeds may increase, ultimately providing more food for ants.

Some analyses of food webs focus on interactions in which predators or prey have a significant influence on the growth rates and sizes of other populations in the community; these strong interactions can affect overall community structure. In the next section we provide examples of strong interactions when we describe how consumers influence the competitive interactions among populations of their prey.

Study Break 51–3

1. What plant growth forms are common in tropical forests?
2. What is the difference between species richness and relative abundance?
3. Peregrine falcons are predatory birds that have been introduced into many North American cities, where they feed primarily on pigeons. The pigeons eat mostly vegetable matter. To what trophic level do pigeons and peregrine falcons belong?

Effects of Population Interactions on Community Characteristics

Numerous studies have shown that interspecific competition and predation can influence a community's species composition.

Interspecific Competition Can Reduce Species Richness within Communities

Interspecific competition can cause the local extinction of species or prevent new species from becoming established in a community, thus reducing its species richness. During the 1960s and early 1970s, ecologists emphasized competition as the primary factor structuring communities. Observations of resource partitioning and character displacement suggested that some process had fostered differences in resource use among coexisting species, and competition provided the most straightforward explanation of these patterns.

Seeking to uncover direct evidence of competition, ecologists undertook many field experiments on competition in natural populations. The experiment on barnacles depicted in Figure 51.12 is typical of this approach, in which researchers determine whether adding or removing a species changes the distribution or population size of its presumed competitors. In the early 1980s, two independent reviews of the literature on these field experiments, one by Joseph Connell and the other by Thomas W. Schoener of the University of California, Davis, suggested that competition is sometimes a potent force. Connell's survey, which included 527 published experiments on 215 species, identified competition in roughly 40% of the experiments and more than 50% of the species. Schoener's review, which used different criteria to evaluate 164 experiments on approximately 400 species, found that competition affected more than 75% of the species.

Although these reviews confirm the importance of competition, the ecological literature upon which they were based probably contains several significant biases. First, ecologists who set out to study competition are more likely to study interactions in which they think competition occurs, and they are more likely to publish research that documents its importance. Accordingly, the literature includes more studies of competition in K-selected species than in r-selected species (review Section 50-5). Recall that populations of r-selected species, such as herbivorous insects, rarely reach carrying capacity, and competition may not limit their population sizes. Thus, the Connell and Schoener surveys may overestimate the importance of competition. Another bias, which Connell called "the ghost of competition past," underestimates the importance of competition. If, as many ecologists believe, resource partitioning and character displacement are the results of past competition, we are unlikely to witness much competition today, even though it was once important in structuring those population interactions.

Ecologists have still not reached consensus about whether interspecific competition strongly influences the species composition and structure of most communities. Plant ecologists and vertebrate ecologists, who often study K-selected species, generally believe that competition has a profound effect on species distributions and resource use. Insect ecologists and marine ecologists, who often study r-selected species, argue that competition is not the major force governing community structure, pointing instead to predation or parasitism and physical disturbance.

Predators Can Boost Species Richness by Stabilizing Competitive Interactions among Their Prey

Predators can influence the species richness and structure of communities by reducing the population sizes of their prey. On the rocky coast of the American Northwest, for example, algae and sessile invertebrates compete for attachment sites on rocks, a requirement for life on a wave-swept shore. California mussels (Mytilus californianus) are the strongest competitors for space, eliminating other species from the community. But at some sites, predatory sea stars (Pisaster ochraceus) preferentially feed on mussels, reducing their numbers and creating space for other species to grow. Because the interaction between Pisaster and Mytilus affects other species as well, it qualifies as a strong interaction.

In the 1960s, Robert Paine of the University of Washington conducted removal experiments to evaluate the effects of Pisaster predation (Figure 51.23). In predator-free experimental plots, mussels outcompeted barnacles, chitons, limpets, and other invertebrate herbivores, reducing species richness from 18 species to two or three. In control plots that contained predators, however, all 18 species persisted. Ecologists describe predators like Pisaster as keystone species, those that have a greater effect on community structure than their numbers might suggest.

Herbivores May Counteract or Reinforce Competition among Their Food Plants

Herbivores also exert complex effects on communities. In the 1970s, Jane Lubchenco, then of Harvard University, studied herbivory in a periwinkle snail (Littorina littorea), a keystone species on rocky shores in Massachusetts (Figure 51.24). Periwinkles preferentially graze on the tender green alga Enteromorpha. In tidepools, which are usually submerged, Enteromorpha outcompetes other algae. Moderate feeding by periwinkles, however, eliminates some Enteromorpha, allowing less competitive algal species to grow. Moderate herbivory by periwinkles therefore increases algal species richness in tidepools. But on high rocks, which are exposed to air during low tide, the dehydration-resistant red alga Chondrus is competitively dominant. Periwinkles don't eat the tough Chondrus, however, feeding instead on the less abundant and competitively inferior Enteromorpha. Thus, on exposed rocks, feeding by the snails reduces algal species richness.

Study Break 51–4

1. How is the scientific literature on interspecific competition biased?
2. What are keystone species, and how do they influence species richness in communities?

   Effects of Disturbance on Community Characteristics

   Recent research tends to support the individualistic view that many communities are not in equilibrium and that their species composition changes frequently. Environmental disturbances—storms, landslides, fires, floods, and cold spells—often eliminate some species, providing opportunities for others to become established.
   

   Frequent Disturbances Keep Some Communities in a Constant State of Flux

   Physical disturbances are common in some environments. For example, lightning-induced fires commonly sweep through grasslands, powerful hurricanes routinely demolish patches of forest, and waves wash over communities that live at the edge of the sea.

   Joseph Connell and his colleagues conducted an ambitious long-term study of the effects of disturbance on coral reefs, shallow tropical marine habitats that are among the most species-rich communities on Earth. In some parts of the world, reefs are routinely battered by violent storms, which wash corals off the substrate, creating bare patches in the reef. The scouring action of storms creates opportunities for coral larvae to settle on bare substrates and start new colonies; ecologists use the word recruitment to describe the process in which young individuals join a population.

   From 1963 to 1992, Connell and his colleagues tracked the fate of the Heron Island Reef at the south end of Australia's Great Barrier Reef (Figure 51.25). The inner flat and protected crests of the reef are sheltered from severe wave action during storms, whereas some pools and crests are routinely exposed to physical disturbance. Because corals live in colonies of variable size, the researchers monitored coral abundance by measuring the percentage of the substrate (that is, the seafloor) that colonies covered. They revisited marked study plots at intervals, photographing and identifying individual coral colonies.

   Five major cyclones crossed the reef during the 30-year study period. Coral communities in the exposed areas of the reef were in a nearly continual state of flux. In exposed pools, four of the five cyclones reduced the percentage of cover, often drastically. On exposed crests, the cyclone of 1972 eliminated virtually all of the corals, and subsequent storms slowed the recovery of these areas for more than 20 years. By contrast, corals in sheltered areas suffered much less storm damage. Nevertheless, their coverage also declined steadily during the study as a natural consequence of the corals' growth. As colonies grew taller and closer to the ocean's surface, their increased exposure to air resulted in substantial mortality.

   Connell and his colleagues also documented recruitment, the growth of new colonies from settling larvae, in their study plots. They discovered that the rate at which new colonies developed was almost always higher in sheltered areas than in exposed areas. However, recruitment rates were extremely variable, depending in part on the amount of space that storms or coral growth had made available.

   This long-term study of coral reefs illustrates that frequent disturbances prevent some communities from reaching an equilibrium determined by interspecific interactions. Changes in the coral reef community at Heron Island result from the combined effects of external disturbances that remove coral colonies from the reef and internal processes (growth and recruitment) that either eliminate colonies or establish new ones. In this community, growth and recruitment are slow processes, and disturbances are frequent. Thus, the community never attains equilibrium.

   Moderate Levels of Disturbance May Foster High Species Richness

   According to the intermediate disturbance hypothesis, proposed by Connell in 1978, species richness is greatest in communities that experience fairly frequent disturbances of moderate intensity. Moderate disturbances create some openings for r-selected species to arrive and join the community, but they allow K-selected species to survive. Thus, communities that experience intermediate levels of disturbance contain a rich mixture of species. Where disturbances are severe and frequent, communities include only r-selected species that complete their life cycles between catastrophes. Where disturbances are mild and rare, communities are dominated by long-lived K-selected species that competitively exclude other species from the community.

   Several studies in diverse habitats have confirmed the predictions of the intermediate disturbance hypothesis. For example, Colin R. Townsend and his colleagues at the University of Otago studied the effects of disturbance at 54 stream sites in the Taieri River system in New Zealand. Disturbance occurs in these communities when water flow from heavy rains moves the rocks, soil, and sand in the streambed, disrupting the habitats where animals live. Townsend and his colleagues measured how much of the substrate moved in different streambeds to index the intensity of the disturbance. Their results indicate that species richness is highest in areas that experience intermediate levels of disturbance (Figure 51.26).

   Some ecologists have also suggested that species-rich communities recover from disturbances more readily than do less diverse communities. For example, David Tilman and his colleagues at the University of Minnesota conducted large-scale experiments in midwestern grasslands on the relationship between species number and the ability of communities to recover from disturbance. Their results demonstrate that grassland plots with high species richness recover from drought faster than plots with fewer species.

   Study Break 51–5

   1. How might disturbances from storms allow coral reefs to be rejuvenated by the recruitment of young individuals?
   2. How do moderately severe and moderately frequent disturbances influence a community's species richness?

      Ecological Succession: Responses to Disturbance

      In response to disturbance, communities undergo ecological succession, a somewhat predictable series of changes in species composition over time.

      Succession Begins after Disturbance Alters a Landscape or Changes the Species Composition of an Existing Community

      Primary succession begins when organisms first colonize terrestrial habitats without soil, such as those created by erupting volcanoes and retreating glaciers (Figure 51.27). Lichens (see Section 28-3), which derive nutrients from rain and bare rock, are usually the first visible colonizers of such inhospitable habitats. They secrete mild acids that erode rock surfaces, initiating the slow development of soil, which is enriched by the organic material lichens produce. After lichens modify a site, mosses (see Section 27-2) colonize patches of soil and grow quickly.

      As soil accumulates, hardy r-selected plants—grasses, ferns, and broad-leaved herbs—colonize the site from surrounding areas. Their roots break up rock, and as they die, their decaying remains enrich the soil. Detritivores and decomposers facilitate these processes. As the soil gets deeper and richer, increased moisture and nutrients support bushes and, eventually, trees. Late successional stages are often dominated by K-selected species with woody trunks and branches that position leaves in sunlight and large root systems that acquire water and nutrients from soil.

      In the classical view of ecological succession, long-lived species eventually dominate a community, and new species join it only rarely. This relatively stable, late successional stage is called a climax community because the dominant vegetation replaces itself and persists until an environmental disturbance eliminates it, allowing other species to invade. Local climate and soil conditions, the surrounding communities where colonizing species originate, and chance events determine the species composition of climax communities. However, recent research suggests that even "climax communities" change slowly in response to environmental fluctuations, as described below.

      Secondary succession occurs after existing vegetation is destroyed or disrupted by an environmental disturbance, such as a fire, a storm, or human activity. The presence of soil makes the disturbed sites ripe for colonization. Moreover, the soil may contain numerous seeds that germinate after the disturbance. The early stages of secondary succession proceed rapidly, but later stages parallel those of primary succession.

      Secondary succession in the north temperate zone is well studied in abandoned farms, called "old fields," where forests were cleared centuries ago. Because the transformation from old field back to forest takes at least a hundred years, ecologists use historical records to find the age of different stands of vegetation and reconstruct the successional sequence by comparing stands of different ages. In the Piedmont region of southeastern North America, an abandoned field is covered by crabgrass (genus Digitaria), an annual plant, during the first growing season. The following year, crabgrass is replaced by horseweed (Conyza canadensis), which cannot persist because it secretes substances that inhibit the germination of its own seeds. Ragweed (Ambrosia artemisiifolia), another annual, dominates during the third year, but it is gradually replaced by perennial asters (genus Erigeron) and broomsedges (genus Andropogon), which are, in turn, replaced by shrubs. Ten to fifteen years after the field was abandoned, pine (genus Pinus) seedlings germinate. Growing pines cast substantial shade and their fallen needles acidify the soil, making the site unsuitable for the plants from earlier successional stages. Because pines are intolerant of shade, pine seedlings don't flourish under mature pine trees. Thus, after 50 to 100 years, pines are replaced by a taller mixed hardwood forest of oaks (genus Quercus) and hickories (genus Carya), the seedlings of which are more shade tolerant than pines. The hardwood forest forms the climax community in the thick, moist soil after more than a century of successional change.

      Similar climax communities sometimes arise from alternative successional sequences. For example, hardwood forests also develop in sites that were once ponds. During aquatic succession, debris from rivers and runoff accumulates in a body of water, causing it to fill in at its margins. The pond is transformed into a swamp, inhabited by plants adapted to a semisolid substrate. As larger plants get established, their high transpiration rates dry the soil, allowing other plant species to colonize. Given enough time, the site may become a meadow or forest, where an area of moist, lowlying ground is the only remnant of the original pond.

      Community Characteristics Change during Succession

      Several characteristics undergo directional change as succession proceeds. First, because r-selected species are short-lived and K-selected species long-lived, species composition changes rapidly in the early stages, but slowly in the late stages of succession. Second, species richness increases rapidly during the early stages because new species join the community faster than resident species become extinct; as succession proceeds, however, species richness stabilizes or may even decline. Third, in terrestrial communities that receive sufficient rainfall, the maximum height and total mass of the vegetation increase steadily as large species replace small ones, creating the complex structure of the climax.

      Because plants influence the physical environment below them, the community itself increasingly moderates the microclimate. The shade cast by a forest canopy retains soil moisture and reduces temperature fluctuations. The trunks and canopy also reduce wind speed. By contrast, the short vegetation in an early successional stage does not effectively shelter the space below it.

      Although ecologists usually describe succession in terms of vegetation, animals undergo succession, too. As the vegetation shifts, new resources become available, and animal species replace each other over time. Herbivorous insects, which often have strict food preferences, undergo succession along with their food plants. And as the herbivores change, so do their predators, parasites, and parasitoids. In old-field succession in eastern North America, different successional stages harbor a changing assortment of bird species (Figure 51.28).

      Several Hypotheses Help to Explain the Processes Underlying Succession

      Differences in dispersal abilities, maturation rates, and life spans among species are at least partly responsible for ecological succession. Early successional stages harbor many r-selected species because they produce numerous small seeds that colonize open habitats and grow quickly. Mature successional stages are dominated by K-selected species because they are long-lived. Nevertheless, coexisting populations inevitably affect one another. Although the role of population interactions in succession is generally acknowledged, ecologists debate the relative importance of processes that either facilitate or inhibit the turnover of species in a community.

      The facilitation hypothesis suggests that species modify the local environment in ways that make it less suitable for themselves but more suitable for colonization by species typical of the next successional stage. For example, when lichens first colonize bare rock, they produce a small quantity of soil, which is required by mosses and grasses that grow there later. According to this hypothesis, changes in species composition are both orderly and predictable because the presence of each stage facilitates the success of the next. Facilitation is very important in primary succession, but it may not be the best model of interactions that influence secondary succession.

      The inhibition hypothesis suggests that new species are prevented from occupying a community by whatever species are already present. According to this hypothesis, succession is neither orderly nor predictable because each stage is dominated by whichever species happen to colonize the site first. Species replacements occur only when individuals of the dominant species die of old age or when an environmental disturbance reduces their numbers. Eventually, long-lived species replace short-lived species, but the precise species composition of a mature community is up for grabs.

      Inhibition appears to play a role in some secondary successions that follow environmental disturbances. For example, rocky intertidal communities on sheltered shores in the Gulf of Maine include some habitat patches that are dominated by an alga (Ascophyllum nodosum) and other patches dominated by a mussel (Mytilus edulis). Do these patches represent different alternative states for the mature community in this habitat? In winter, small patches in this habitat are sometimes scoured clean by sea-borne ice, after which the patch undergoes succession. In 2009, Peter Petraitis of the University of Pennsylvania and colleagues from several other institutions reported on the fate of habitat patches in which they had experimentally simulated "ice scour." If mussels colonized the cleared site first, they grew faster than the algae and eventually dominated the community. But if the alga colonized first, it provided cover for sea stars and other predators, which consumed mussels that subsequently grew there. Thus, the species composition of the mature community depended on which species arrived at the site first.

      The tolerance hypothesis asserts that succession proceeds because competitively superior species replace competitively inferior ones. According to this model, early-stage species neither facilitate nor inhibit the growth of later-stage species. Instead, as more species arrive at a site and resources become limiting, competition eliminates species that cannot harvest scarce resources successfully. In the Piedmont region of North America, for example, hardwood trees are more tolerant of shade than pine trees are, and hardwoods gradually replace pines during succession. Thus, the climax community includes only strong competitors. Tolerance may explain the species composition of many transitional and mature communities.

      At most sites, succession probably results from a combination of facilitation, inhibition, and tolerance, coupled with interspecific differences in dispersal, growth, and maturation rates. Moreover, within a community, the patchiness of abiotic factors also strongly influences plant distributions and species composition. In the deciduous forests of eastern North America, maples (genus Acer) predominate on wet, low-lying ground, but oaks (genus Quercus) are more abundant at higher and drier sites. Thus, a mature deciduous forest is more often a mosaic of species than a uniform stand of trees.

      Disturbance and density-independent factors also play important roles, in some cases speeding successional change. In northern forests, for example, moose prefer to feed on deciduous shrubs, accelerating the rate at which conifers replace them. In other cases, disturbance inhibits successional change, establishing a disturbance climax or disclimax community. In many grassland communities (see Section 49-4), grazing by large mammals and periodic fires kill the seedlings of trees that would otherwise become established. Thus, disturbance prevents the succession from grassland to forest, and grassland persists as a disclimax community.

      On a local scale, disturbances often destroy small patches of vegetation, returning them to an earlier successional stage. A hurricane may knock over a few trees in a forest, creating small, sunny patches of open ground. Locally occurring r-selected species take advantage of the resources that are suddenly available and quickly colonize the openings. These local patches then undergo succession that is out of step with the immediately surrounding forest. Thus, moderate disturbance, accompanied by succession in local patches, can increase species richness in many communities.

      Study Break 51–6

      1. What is the difference between primary succession and secondary succession?
      2. How does a climax community differ from early successional stages?
      3. How do the three hypotheses about the causes of ecological succession view the role of population interactions in the successional process?

         Variations in Species Richness among Communities

         Species richness often varies among communities according to a recognizable pattern. Two large-scale patterns of species richness—latitudinal trends and island patterns—have captured the attention of ecologists for more than a century.

         Many Types of Organisms Exhibit Latitudinal Gradients in Species Richness

         Ever since Darwin and Wallace traveled the globe (see Section 19-2), ecologists have recognized broad latitudinal trends in species richness. For many, but not all, plant and animal groups, species richness follows a latitudinal gradient, with the most species in the tropics and a steady decline in numbers toward the poles (Figure 51.29). Several general hypotheses may explain these striking patterns.

         Some hypotheses propose historical explanations for the origin of high species richness in the tropics. The benign climate in tropical regions allows some tropical organisms to have more generations per year than their temperate counterparts. And, given the small seasonal changes in temperature, tropical species may be less likely than temperate species to migrate from one habitat to another, thus reducing gene flow between geographically isolated populations (see Section 21-3). These factors may have fostered higher speciation rates in the tropics, accelerating the accumulation of species. Tropical communities may also have experienced severe disturbance less often than communities at higher latitudes, where periodic glaciations have caused repeated extinctions. Thus, new species may have accumulated in the tropics over longer periods of time.

         Other hypotheses focus on ecological explanations for the maintenance of high species richness in the tropics. Some resources are more abundant, predictable, and diverse in tropical communities. Tropical regions experience more intense sunlight, warmer temperatures in most months, and higher annual rainfall than temperate and polar regions (see Chapter 49). These factors provide a long and predictable growing season for the lush tropical vegetation, which supports a rich assemblage of herbivores, and through them many carnivores and parasites. Furthermore, the abundance, predictability, and year round availability of resources allow some tropical animals to have specialized diets. For example, tropical forests support many species of fruit-eating bats and birds, which could not survive in temperate forests where fruits are not available year-round.

         Species richness may therefore be a self-reinforcing phenomenon in tropical communities. Complex webs of population interactions and interdependency have coevolved in relatively stable and predictable tropical climates. Predator–prey, competitive, and symbiotic interactions may prevent individual species from dominating communities and reducing species richness.

         The Theory of Island Biogeography Explains Variations in Species Richness

         Although the species richness of communities may be stable over time, species composition is often in flux as new species join a community and others drop out. In the 1960s, Robert MacArthur of Princeton University and Edward O. Wilson of Harvard University addressed the question of why communities vary in species richness, using islands as model systems. Islands provide natural laboratories for studying ecological phenomena, just as they do for evolution (see Focus on Basic Research in Chapter 21). Island communities are often small, have well-defined boundaries, and are isolated from surrounding communities.

         In developing the equilibrium theory of island biogeography, MacArthur and Wilson sought to explain variations in species richness on islands of different size and different levels of isolation from other landmasses (Figure 51.30). They hypothesized that the number of species on any island was governed by a give and take between two processes: the immigration of new species to the island and the extinction of species already there (Figure 51.30A).
         

         According to the MacArthur–Wilson model, the mainland harbors a species pool from which species immigrate to offshore islands. Seeds and small arthropods are carried by wind or floating debris; some animals, such as birds, arrive under their own power. When few species are already on an island, the rate at which new species immigrate to the island is high. But as more species inhabit the island over time, the immigration rate declines because there are fewer species left in the mainland pool that can still arrive on the island as new colonizers.

         Once a species immigrates to an island, its population grows and persists for some time. But as the number of species on the island increases, the rate at which those species go extinct also rises. The extinction rate increases through time partly because there are more species that can go extinct there. In addition, as the number of species on the island increases, competition and predator–prey interactions can reduce the population sizes of some species and drive them to extinction.

         According to MacArthur and Wilson's theory, an equilibrium between immigration and extinction determines the number of species that ultimately occupy an island. In other words, once equilibrium is reached, the number of species remains relatively constant because one species already on the island goes extinct in about the same time it takes a new species to immigrate to the island. The model does not specify which species immigrate to the island or which ones already on the island go extinct. It simply predicts that the number of species on the island is in equilibrium, although species composition is not. The ongoing processes of immigration and extinction establish a constant turnover in the roster of species that live on any island.

         The MacArthur–Wilson model explains why some islands harbor more species than others. Large islands have higher immigration rates than small islands do because they present a larger target for dispersing organisms. Moreover, large islands have lower extinction rates because they can support larger populations and provide a greater range of habitats and resources. Thus, at equilibrium, large islands have more species than small islands (Figure 51.30B). Similarly, islands near the mainland have higher immigration rates than distant islands do, because dispersing organisms are more likely to locate islands that are close to their point of departure. Distance does not affect extinction rates. Thus, at equilibrium, islands that lie closer to a mainland source have more species than more distant islands (Figure 51.30C).

         The equilibrium theory's predictions about the effects of area and distance are generally supported by observational data on plants and animals (Figure 51.31). Daniel Simberloff, one of Wilson's graduate students at Harvard University, was the first person to test the theory's predictions experimentally; he monitored the immigration of arthropods to, and extinction of arthropods on, individual red mangrove trees in the Florida Keys (Figure 51.32). The trees, with canopies that spread from 11 to 18 m in diameter, grow in shallow water and are isolated from their neighbors; thus, each tree is an island that harbors an arthropod community. The species pool on the Florida mainland includes about 1,000 arthropod species, but each mangrove island contains no more than 40 species at one time.
         

         After cataloging the species on each island, Simberloff and Wilson hired an extermination company to eliminate all arthropods on them (Figure 51.32A). Simberloff then monitored both the immigration of arthropods to the islands and the extinction of species that became established on them. He surveyed six islands regularly for two years and at intervals thereafter.

         The results of this experiment confirm several predictions of MacArthur and Wilson's theory (Figure 51.32B). Arthropods recolonized the islands rapidly, and within eight or nine months the number of species living on each island had reached an equilibrium that was near the original species number. In addition, the island nearest to the mainland had more species than the most distant island. However, immigration and extinction were incredibly rapid, and Simberloffand Wilson suspected that some species went extinct even before they had noted their presence. The researchers also discovered that three years after the experimental treatments, the species composition of the islands was still changing constantly and did not remotely resemble the species composition in the islands before they were defaunated.

         As described in Insights from the Molecular Revolution, the equilibrium view of species richness also applies to mainland communities, which exist as islands in a metaphorical sea of dissimilar habitat. Lakes are "islands" in a "sea" of dry land, and mountaintops are habitat "islands" in a "sea" of low terrain. Species richness in these communities is partly governed by the immigration of new species from distant sources and the extinction of species already present. As human activities disrupt environments across the globe, undisturbed sites function as island-like refuges for threatened and endangered species. Conservation biologists now apply the general lessons of MacArthur and Wilson's theory to the design of nature preserves (see Chapter 53).
         

         Study Break 51–7

         1. What factors may foster the maintenance of high species richness in tropical communities?
         2. According to the equilibrium theory of island biogeography, what are the effects of an island's size and its distance from the mainland on the number of species that can occupy it?