Smith, T. M., & Smith, R. L. (2015). Elements of Ecology (9th ed.). Boston, MA: Pearson.
12.1 Species Interactions Can Be Classified Based on Their Reciprocal Effects
If we designate the positive effect of one species on another as +, a detrimental effect as −, and no effect as 0, we can use this qualitative description of the different ways in which populations of two species interact to develop a classification of possible interactions between two co-occurring species (Table 12.1). When neither of the two populations affects the other, the relationship is (00), or neutral. If the two populations mutually benefit, the interaction is (++), or positive, and the relationship is called mutualism (Chapter 15). When one species maintains or provides a condition that is necessary for the welfare of another but does not affect its own well-being, the relationship (+0) is called commensalism . For example, the trunk or limb of a tree provides the substrate on which an epiphytic orchid grows (Figure 12.1). The arrangement benefits the orchid, which gets nutrients from the air and moisture from aerial roots, whereas the tree is unaffected.
When the relationship is detrimental to the populations of both species (−−), the interaction is termed competition (Chapter 13). In some situations, the interaction is (−0). One species reduces or adversely affects the population of another, but the affected species has no influence in return. This relationship is amensalism . It is considered by many ecologists as a form of asymmetric competition, such as when taller plant species shade species of smaller stature.
Relationships in which one species benefits at the expense of the other (+−) include predation, parasitism, and parasitoidism (see Chapter 14 for more information on predation and Chapter 15 for more information on parasitism and parasitoidism). Predation is the process of one organism feeding on another, typically killing the prey. Predation always has a negative effect on the individual prey. In parasitism , one organism feeds on the other but rarely kills it outright. The parasite and host live together for some time. The host typically survives, although its fitness is reduced. Parasitoidism , like predation, kills the host eventually. Parasitoids, which include certain wasps and flies, lay eggs in or on the body of the host. When the eggs hatch, the larvae feed on it. By the time the larvae reach the pupal stage, the host has succumbed.
12.2 Species Interactions Influence Population Dynamics
The varieties of species interactions outlined in the previous section typically involve the interaction of individual organisms. A predator captures a prey or a bacterium infects a host organism. Yet through their beneficial or detrimental effects on the individuals involved, these interactions influence the collective properties of birth and death at the population level, and in doing so, influence the dynamics of the respective populations. For example, by capturing and killing individual prey, predators function as agents of mortality. We might therefore expect that as the number of predators (Npredator) in an area increases, the number of prey captured and killed will likewise increase. If we assume the simplest case of a linear relationship, we can represent the influence of changes in the predator population (Npredator) on the death rate of the prey population (dprey) as shown in Figure 12.2a. As the number of predators in the population (Npredator) increases, the probability of an individual in the prey population (Nprey) being captured and killed increases. Subsequently, the death rate of the prey population increases. The net effect is a decline in the growth rate of the prey population. Note the similarity in the functional relationship presented in Figure 12.2a with the example of density-dependent population control presented earlier (Chapter 11, Figure 11.1). Previously, we examined how an increase in population size can function as a negative feedback on population growth by increasing the mortality rate or decreasing the birthrate (density-dependent population regulation; Section 11.2 and Figure 11.4). The relationship shown in Figure 12.2a expands the concept of density-dependent population regulation to include the interaction between species. As the population of predators increases, there is a subsequent decline in the population of prey as a direct result of the prey’s increased rate of mortality.
A similar approach can be taken to evaluate the positive effects of species interactions. In the example of predation, whereas the net effect of predation on the prey is negative, the predator benefits from the capture and consumption of prey. Prey provides basic food resources to the predator and directly influences its ability to survive and reproduce. If we assume that the ability of a predator to capture and kill prey increases as the number of potential prey increase (Nprey), and that the reproductive fitness of a predator is directly related to its consumption of prey, then we would expect the birthrate of the predator population (bpredator) to increase as the size of the prey population increases (Figure 12.2b). The result is a direct link between the availability of prey (size of the prey population, Nprey) and the growth rate of the predator population (dNpredator/dt).
In Chapter 11, we developed a logistic model of population growth. It is a model of intraspecific competition and density-dependent population regulation using the concept of carrying capacity, K. The carrying capacity represents the maximum sustainable population size that can be supported by the available resources. The carrying capacity functions to regulate population growth in that as the population size approaches K, the population growth rate approaches zero (dN/dt = 0).
When individuals of two different species share a common limiting resource that defines the carrying capacity, there is potential for competition between individuals of the two species (interspecific competition). For example, let’s define a population of a grazing antelope inhabiting a grassland as N1, and the carrying capacity of the grassland to support that population as K1 (the subscript 1 refers to species 1). The logistic model of population growth (see Section 11.1) would then be:
dN1/dt = r1N1(1 − N1/K1)dN1/dt = r1N1(1 − N1/K1)
Now let’s assume that a second species of antelope inhabits the same grassland, and to simplify the example, we assume that individuals of the second species—whose population we define as N2—have the same body size and exactly the same rate of food consumption (grazing of grass) as do individuals of the first species. As a result, when we evaluate the role of density-dependent regulation on the population of species 1 (N1), we must now also consider the number of individuals of species 2 (N2) because individuals of both species feed on the grass that defines the carrying capacity of species 1 (K1). The new logistic model for species 1, will be:
dN1/dt = r1N1(1 − (N1 + N2)/K1)dN1/dt = r1N1(1 − (N1 + N2)/K1)
For example, if the carrying capacity of the grassland for species 1 is 1000 individuals (K1 = 1000)—because species 2 draws on the exact same resource in exactly the same manner—the combined carrying capacity of the grassland is also 1000. If there are 250 individuals of species 2 (N2 = 250) living on the grassland, it effectively reduces the carrying capacity for species 1 from 1000 to 750 (Figure 12.3a). The population growth rate of species 1 now depends on the population sizes of both species 1 and 2 relative to the carrying capacity (Figure 12.3b). Although we have defined the two antelope species as being identical in their use of the limiting resource that defines the carrying capacity, this is not always the case. In reality, it is necessary to evaluate the overlap in resource use and quantify the equivalency of one species to another (see Quantifying Ecology 12.1).
In all cases in which individuals of two species interact, the nature of the interaction can be classified qualitatively as neutral, positive, or negative, and the influence of the specific interaction can be evaluated in terms of its impact on the survival or reproduction of individuals within the populations. In the discussion that follows, we develop quantitative models to examine how the diversity of species interactions outlined in Table 12.1 influence the combined population dynamics of the species involved (Chapters 13, 14, and 15). In all cases, these models involve quantifying the per capita effect of interacting individuals on the birthrates and death rates of the respective populations.
Quantifying Ecology 12.1 Incorporating Competitive Interactions in Models of Population Growth
When individuals of two different species (represented as populations N1 and N2) share a common limiting resource that defines the carrying capacity for each population (K1 and K2), there is potential for competition between individuals of the two species (interspecific competition). Thus, the population density of both species must be considered when evaluating the role of density-dependent regulation on each population. In Section 12.2, we gave the example of two species of antelope that share the common limiting food resource of grass. We assumed that individuals of the two species were identical in their food selection and the rate at which they feed, therefore, with respect to the carrying capacity of the grassland, individuals of the two species are equivalent to each other; that is, in resource consumption one individual of species 1 is equivalent to one individual of species 2. As a result, when evaluating the growth rate of species 1 using the logistic model of population growth, it is necessary to include the population sizes of both species relative to the carrying capacity (see Figure 12.4):
dN1/dt = r1N1(1 − (N1 + N2)/K1)dN1/dt = r1N1(1 − (N1 + N2)/K1)
However, two species, even closely related species, are unlikely to be identical in their use of resources. So it is necessary to define a conversion factor that can equate individuals of species 2 to individuals of species 1 as related to the consumption of the shared limited resource. This is accomplished by using a competition coefficient, defined as a, that quantifies individuals of species 2 in terms of individuals of species 1 as related to the consumption of the shared resource. Using the example of two antelope species, let us now assume that both species still feed on the same resource (grass), however, individuals of species 2 have on average only half the body mass of individuals of species 1 and therefore consume grass at only half the rate of species 1. Now an individual of species 2 is only equivalent to one-half an individual of species 1 with respect to the use of resources. In this case, a = 0.5, and we can rewrite the logistic model for species 1 shown previously as:
dN1/dt = r1N1(1 − (N1 + αN2)/K1)dN1/dt = r1N1(1 − (N1 + αN2)/K1)
Because in Section 12.2 we defined the carrying capacity of the grassland for species 1 as K1 = 1000, we can substitute the values of a and K1 in the preceding equation:
dN1/dt = r1N1(1 − (N1 + 0.5N2)/1000)dN1/dt = r1N1(1 − (N1 + 0.5N2)/1000)
Now the growth rate of species 1 (dN1/dt) approaches zero as the combined populations of species 1 and 2, represented as N1 + 0.5N2, approach a value of 1000 (the value of K1).
We have considered how to incorporate the effects of competition from species 2 into the population dynamics of species 1 using the competition coefficient a, but what about the effects of species 1 on species 2? The competition for food resources (grass) will also function to reduce the availability of resources to species 2. We can take the same approach and define a conversion factor that can equate individuals of species 1 to individuals of species 2, defined as b. Because individuals of species 1 consume twice as much resource (grass) as individuals of species 2, it follows that an individual of species 1 is equivalent to 2 individuals of species 2; that is, b = 2.0. It also follows that if individuals of species 2 require only half the food resources as individuals of species 1, then the carrying capacity of the grassland for species 2 should be twice that for species 1; that is, K2 = 2000. The logistic growth equation for species 2 is now:
dN2/dt = r2N2(1 − (N2 + βN1)/K2)dN2/dt = r2N2(1 − (N2 + βN1)/K2)
or, substituting the values for b and K2
dN2/dt = r2N2(1 − (N2 + 2.0N1)/2000)dN2/dt = r2N2(1 − (N2 + 2.0N1)/2000)
We now have a set of equations that can be used to calculate the growth of the two competing species that considers their interaction for the limiting food resource. We explore this approach in more detail in the following chapter (Chapter 13).
In the example of the two hypothetical antelope species presented previously, the estimation of the competition coefficients (a and b) appear simple and straightforward. Both species are identical in their diet and differ only in the rate at which they consume the resource (which is defined as a simple function of their relative body masses). In reality, even closely related species drawing on a common resource (such as grazing herbivores) differ in their selection (preferring one group of grasses of herbaceous plants over another), foraging behavior, timing of foraging, and other factors that influence the nature of their relative competitive effects on each other. As such, quantifying species interactions, such as resource competition, can be a difficult task, as we shall see in the following chapter (Chapter 13, Interspecific Competition).
12.3 Species Interactions Can Function as Agents of Natural Selection
For a number of reasons, the interaction between two species will not influence all individuals within the respective populations equally. First, interactions among species involve a diverse array of physiological processes and behavioral activities that are influenced by phenotypic characteristics (physiological, morphological, and behavioral characteristics of the individuals). Secondly, these phenotypic characteristics vary among individuals within the populations (see Chapter 5). Therefore, the variations among individuals within the populations will result in differences in the nature and degree of interactions that occur. For example, imagine a species of seed-eating bird that feeds on the seeds of a single plant species. Individuals of the plant species exhibit a wide degree of variation in the size of seeds that they produce. Some individuals produce smaller seeds, whereas others produce larger seeds (Figure 12.4a), and seed size is a heritable characteristic (genetically determined). Seed size is important to the birds because the larger the seed, the thicker the seed coat, and the more difficult it is for a bird to crush the seed with its bill. If the seed coat is not broken, the seed passes through the digestive system undigested and provides no food value to the bird. As a result, birds actively select smaller seeds in their diet (Figure 12.4c). In doing so, the birds are decreasing the reproductive success of individual plants that produce small seeds while increasing the relative fitness of those individuals that produce larger seeds. The net effect is a shift in the distribution of phenotypes in the plant population to individuals that produce larger, harder seeds (Figure 12.4d). In this situation, the bird population (and pattern of seed predation) is functioning as an agent of natural selection, increasing the relative fitness of one phenotype over another (see Section 5.6). Over time, the result represents a directional change in the genetic structure of the population (gene frequencies), that is, the process of evolution (Chapter 5).
In this example, the predator functions as an agent of natural selection, decreasing the reproduction for certain phenotypes (small seed-producing individuals) within the plant population and increasing the relative fitness of other phenotypes (large seed-producing individuals). But the shifting distribution of phenotypes within the plant population and the resulting change in the distribution of food resources will in turn have a potential influence on the predator population (Figure 12.4b). The directional selection for increased seed size within the plant population decreases the relative abundance of smaller seeds, effectively decreasing the availability of food resources for birds with smaller bill sizes. If the birds with smaller bills are unable to crack the larger seeds, these individuals will experience a decreased probability of survival and reproduction, which increases the relative fitness of individuals with larger bill size. The shift in the distribution of phenotypes in the plant population, itself a function of selective pressures imposed by the bird population, now functions as an agent of natural selection in the predator (bird) population. The result is a shift in the distribution of phenotypes and associated gene frequencies within the bird population toward larger bill size (Figure 12.4e). This process in which two species undergo reciprocal evolutionary change through natural selection is called coevolution .
Unlike adaptation to the physical environment, adaptation in response to the interaction with another species can produce reciprocal evolutionary responses that either thwart (counter) these adaptive changes, as in the previous example, or in mutually beneficial interactions, magnify (reinforce) their effect. An example of the latter can be found in the relationship between flowering plants and their animal pollinators. Many species of flowering plants require the transfer of pollen from one individual to another for successful fertilization and reproduction (outcrossing; Figure 12.5). In some plant species, this is accomplished through passive transport by the wind, but many plants depend on animals to transport pollen between flowers. By attracting animals, such as insects or birds, to the flower, pollen is spread. When the animal comes into contact with the flower, pollen is deposited on its body, which is then transferred to another individual as the animal travels from flower to flower. This process requires the plant species to possess some mechanism to attract the animal to the flower. A wide variety of characteristics has evolved in flowering plants that function to entice animals through either signal or reward. Signals can involve brightly colored flowers or scents. The most common reward to potential pollinators is nectar, a sugar-rich liquid produced by plants, which serves no purpose for the individual plant other than to attract potential pollinators. Nectar is produced in glands called nectaries, which are most often located at the base of the floral tube (see Figure 12.5).
The relationship between nectar-producing flowers and nectar-feeding birds provides an excellent example of the magnification of reciprocal evolutionary responses—coevolution—resulting from a mutually beneficial interaction. The elongated bill of hummingbirds distinguishes them from other birds and is uniquely adapted to the extraction of nectar (Figure 12.6). Their extremely long tongues are indispensable in gaining nectar from long tubular flowers. Let us assume a species of hummingbird feeds on a variety of flowering plants within a tropical forest but prefers the flowers of one plant species in particular because it produces larger quantities of nectar. Thus, the reward to the hummingbird for visiting this species is greater than that of other plant species in the forest. Now assume that flower size (an inherited characteristic) varies among individuals within the plant population and that an increase in nectar production is associated with elongation of the floral tube (larger flower size). Individual plants with larger flowers and greater nectar production would have an increased visitation rate by hummingbirds. If this increase in visitation rate results in an increase in pollination and reproduction, the net effect is an increase in the relative fitness of individuals that produce larger flowers, shifting the distribution of phenotypes within the plant population. The larger flower size and longer floral tube, however, make it more difficult to gain access to the nectar. Individual hummingbirds with longer bills are more efficient at gaining access, and bill size varies among individuals within the population. With increased access to nectar resources, the relative fitness of longer-billed individuals increases at the expense of individuals with shorter bills. In addition, any gene mutation that results in increasing bill length with be selected for because it will increase the fitness of the individual and its offspring (assuming that they exhibit the phenotype). The genetic changes that are occurring in each population are reinforced and magnified by the mutually beneficial interaction between the two species. The plant characteristic of nectar production is reinforced and magnified by natural selection in the form of improved pollination success by the plant and reproductive success by the hummingbird. In turn, the increased flower size and associated nectar production functions as a further agent of natural selection in the bird population, resulting in an increase in average bill size (length). One consequence of this type of coevolutionary process is specialization, wherein changes in phenotypic characteristics of the species involved function to limit the ability of the species to carry out the same or similar interactions with other species. For example, the increase in bill size in the hummingbird population will function to limit its ability to efficiently forage on plant species that produce smaller flowers, restricting its feeding to the subset of flowering plants within the tropical forest that produces large flowers with long floral tubes (see Figure 12.6). In the extreme case, the interaction can become obligate, where the degree of specialization in phenotypic characteristics results in the two species being dependent on each other for survival and successful reproduction. We will examine the evolution of obligate species interactions in detail later (Chapter 15).
Unlike the case of mutually beneficial interactions in which natural selection functions to magnify the intensity of the interaction, interactions that are mutually negative to the species involved can lead to the divergence in phenotypic characteristics that function to reduce the intensity of interaction. Such is the case when the interaction involves competition for essential resources. Consider the case wherein two species of seed-eating birds co-occur on an island. The two populations differ in average body and bill size, yet the two populations overlap extensively in the range of these phenotypic characteristics (Figure 12.7a) and subsequently in the range of seed sizes on which they forage (Figure 12.7b). The selection of seeds by individual birds is related to body and bill size. Smaller individuals are limited to feeding on the smaller, softer seeds, whereas only larger individuals are capable of cracking the larger, harder seeds. Although larger birds are able to feed on smaller seeds, it is energetically inefficient; therefore, their foraging is restricted to relatively larger seeds (see Section 5.8 for an example).
Seed resources on the island are limited relative to the populations of the two species, hence, competition is often intense for the intermediate-sized seeds for which both species forage. If competition for intermediate-sized seeds functions to reduce the fitness of individual birds that depend on these resources, the result would be reduced survival and reproductive rates for larger individuals of the smaller species and smaller individuals of the larger species (Figure 12.7c). This result represents a divergence in the average body and bill size for the two populations that functions to reduce the potential for competition between the two species (Figure 12.7d).
12.4 The Nature of Species Interactions Can Vary across Geographic Landscapes
We have examined how natural selection can result in genetic differentiation, that is, genetic differences among local populations. Species with wide geographic distributions generally encounter a broader range of physical environmental conditions than species whose distribution is more restricted. The variation in physical environmental conditions often gives rise to a corresponding variation in phenotypic characteristics. As a result, significant genetic differences can occur among local populations inhabiting different regions (see Section 5.8 for examples). In a similar manner, species with wide geographic distributions are more likely to encounter a broader range of biotic interactions. For example, a bird species such as the warbling vireo (Vireo gilvus) that has an extensive geographic range in North America, extending from northern Canada to Texas and from coast to coast, is more likely to encounter a greater diversity of potential competitors, predators, and pathogens than will the cerulean warbler (Dendroica cerula), whose distribution is restricted to a smaller geographic region of the eastern United States (see Figure 17.2 for distribution maps). Changes in the nature of biotic interactions across a species geographic range can result in different selective pressures and adaptations to the local biotic environment. Ultimately, differences in the types of species interactions encountered by different local populations can result in genetic differentiation and the evolution of local ecotypes similar to those that arise from geographic variations in the physical environment (see Section 5.8 for examples of the latter). The work of Edmund Brodie Jr. of Utah State University presents an excellent example.
Brodie and colleagues examined geographic variation among western North American populations of the garter snake (Thamnophis sirtalis) in their resistance to the neurotoxin tetrodotoxin (TTX). The neurotoxin TTX is contained in the skin of newts of the genus Taricha on which the garter snakes feed (Figure 12.8a). These newts are lethal to a wide range of potential predators, yet to garter snakes having the TTX-resistant phenotype, the neurotoxin is not fatal. Both the toxicity of newts (TTX concentration in their skin) and the TTX resistance of garter snakes vary geographically (Figure 12.8b). Previous studies have established that TTX resistance in the garter snake is highly heritable (passed from parents to offspring), so if TTX resistance in snakes has co-evolved in response to toxicity of the newt populations on which they feed, it is possible that levels of TTX resistance exhibited by local populations of garter snakes will vary as a function of the toxicity of newts on which they feed. The strength of selection for resistance would vary as a function of differences in selective pressure (the toxicity of the newts).
To test this hypothesis, the researchers examined TTX resistance in more than 2900 garter snakes from 40 local populations throughout western North America, as well as the toxicity of newts at each of the locations. The researchers found that the level of TTX resistance in local populations varies with the presence of toxic newts. Where newts are absent or nontoxic (as is the case on Vancouver Island, British Columbia), snakes are minimally resistant to TTX. In contrast, levels of TTX resistance increased more than a thousand-fold with increasing toxicity of newts (see Figure 12.8b). Brodie and his colleagues found that for local populations, the level of resistance to TTX varies as a direct function of the levels of TTX in the newt population on which they prey (Figure 12.8c). The resistance and toxicity levels match almost perfectly over a wide geographic range, reflecting the changing nature of natural selection across the landscape.
In some cases, even the qualitative nature of some species interactions can be altered when the background environment is changed. For example, mycorrhizal fungi are associated with a wide variety of plant species (see Chapter 15, Section 15.11). The fungi infect the plant root system and act as an extension of the root system. The fungi aid the plant in the uptake of nutrients and water, and in return, the plant provides the fungi with a source of carbon. In environments in which soil nutrients are low, this relationship is extremely beneficial to the plant because the plant’s nutrient uptake and growth increase. (Figure 12.9a). Under these conditions, the interaction between plant and fungi is mutually beneficial. In environments in which soil nutrients are abundant, however, plants are able to meet nutrient demand through direct uptake of nutrients through their root system. Under these conditions, the fungi are of little if any benefit to the plant; however, the fungi continue to represent an energetic cost to the plant, reducing its overall net carbon balance and growth (Figure 12.9b). Across the landscape, the interaction between plant and fungi changes from mutually beneficial (++) to parasitic (+−) with increasing soil nutrient availability.