Chapter 6: Principles of ecology
Edited by Peter Moyle & Douglas Kelt
By Mary A. Orland, July 2004
PRINCIPLES OF ECOLOGY
Conservation of wildlife requires an understanding of ecology, the science devoted to study of the interactions among organisms and their environment. Ecology is defined as “the study of the abundance and distribution of organisms” (Begon, Harper and Townsend 1996). Understanding why animals thrive where they do requires an intimate knowledge of both the organisms and they environment in which they live, including other organisms. Ecology is a subdiscipline of biology, the scientific study of life. Biology is a broad discipline that spans many levels of organization, from molecules to ecosystems. In this chapter we present the broad principles that are incorporated into ecological thinking, or, as Aldo Leopold so gracefully stated, into “thinking like a mountain.”
THE IMPORTANCE OF SCALE IN BIOLOGY
Traditionally, ecology focuses on the larger scales in biology, from the individual organism through populations, communities, ecosystems, and the biosphere (Box 6.1).
Box 6.1 Levels of organization in ecology: emergent properties.
- Populations are interbreeding groups of individuals of the same species, generally living in the same contiguous habitat.
- Communities are interacting populations of different species.
- Ecosystems are comprised of both the biotic (living) and abiotic (non-living) factors in a given area; they contain both the broad biological community and all the physical processes (such as weather, soil, hydrology, nutrients, energy flow etc.) that influence that community.
- The biosphere is global in scale, and includes all the biological and physical processes that allow for and influence life on Earth.
- Higher scales of organization contain smaller scales (e.g., a given community contains populations of various species), yet all scales also possess properties that are unique to that scale and that cannot be deduced from properties of included scales. For example, the nature and strength of interactions among two species of meadow mice cannot be fully explained even with the most sophisticated models of their population dynamics. Thus, many interactions at the community level are emergent properties of communities.
In contrast, molecular and cell biology studies scales smaller than the individual organism. The great difference in the scale of the subject matter of these sub-disciplines of biology has resulted in large differences in the study methods, ways of thinking, and disciplinary culture of these two sub-disciplines. In fact, these sub-disciplines are now essentially treated as different fields at most major universities, with different departments, undergraduate majors, and faculty who do not read the same literature or go to the same conferences. As a result a dichotomy has been growing within biology. There are also biologists who work at the individual organism scale, studying topics like anatomy, physiology, and behavior, but even many of these “organismal” biologists will tend toward either the broader-scale ecological way of thinking or the smaller-scale molecular/cellular approach.
This leads us to ask the following questions. Why does a divide seem to occur in biology at the scale of the individual organism? Why are the biological processes, and the ways of studying them, so radically different at larger, ecological scales vs. smaller, molecular and cellular scales? What happens at the scale of the individual organism that is so crucial to biology, and how might it explain the seemingly dichotomous nature of modern biology?
Natural selection is the driving force of adaptation and evolution, and it is such an important topic in biology that we have devoted an entire chapter to it (Chapter 5). Biologists generally agree that natural selection occurs predominantly on the scale of the individual organism (Williams 1966), although some argue that it can theoretically occur at other scales as well (Wilson 1980). Each individual organism acts to maximize its own survival and ability to produce offspring that are in turn able to survive and reproduce (called fitness by biologists), even at the expense of other organisms of the same species. Thus, natural selection means that those heritable traits that increase the fitness of an individual organism have a greater probability of being present in future generations within the population. The fact that selection is nearly always strongest on the scale of the individual organism has important ramifications for understanding ecology (Levin 2002). At scales smaller than the individual organisms, the units (cells, genes) are inextricably dependent upon each other for survival. As a result, cooperation among these units is high, and overall they work together for the “good” of the whole organism. However, it is not necessarily easy to keep these units working together, and just how the conflicts among genes and cell lineages were resolved so that multi-cellular organisms could evolve is a fascinating question in evolutionary biology. In fact, cancer is one instance where cells increase their own rate of replication at the expense of the other cells, even though that results in the demise of the entire organism. Our immune systems keep this from happening most of the time, removing those cells that are not working for the good of the whole, yet still 24% of Americans get cancer at some point in their lives (Masters 1996).1
In contrast, at scales larger than that of the individual organism, the units generally do not work together for the good of the whole. Each individual organism is out to maximize its own individual survival and reproduction, even if it does not benefit the population, community, or ecosystem. The amazing thing is that in spite of the fact that essentially all the sub-units of ecological systems are behaving in a selfish “cancerous” way, populations and communities continue to thrive and ecosystem processes continue to exist. The conditions for life are not regulated for the good of the whole ecosystem, but rather the persistence of life emerges from the interactions between organisms and the environment. This may sound brutal, and in many respects it is. In natural systems the vast majority of organisms die before reproducing, usually being eaten by other organisms or starving to death. Despite this, organisms do things to benefit each other, although they do so mainly if it also benefits them as individuals or does not harm them in any way. This is in direct contrast to many popular versions of ecological systems, which often describe communities and ecosystems as a “balanced” assemblage of organisms working together to maintain life. The old “balance of nature” paradigm basically assumed each ecosystem had some ideal state that it could maintain indefinitely, at least in the absence of humans. We now realize that even in the absence of human influence, ecosystems are constantly changing in response to changes in the environment (e.g., climate change caused by volcanic eruptions, big floods on rivers, fires in forests) and the organisms that live there (evolution). In fact, if there is a new paradigm for ecology it is that “the only constant is change.” Certainly, the fossil history of life provides good evidence of this!
Despite the ever-changing nature of ecological systems, all organisms and species within an ecosystem are dependent upon other life forms for their existence. It is clear that predators, such as wolves, are dependent upon their herbivore prey, such as elk and moose, for existence. It is also readily apparent that elk and moose are dependent upon the plant species they eat. Those plants are in turn dependent upon the microorganisms that form soil, and often on symbiotic fungi in their roots, called mycorrhizae, to obtain water and nutrients. Without the mycorrhizae there quite likely would be no wolves! The wolves themselves also have indirect effects that benefit other species. For instance, ecologists have observed that where wolves are present in the Greater Yellowstone Ecosystem of the northern Rocky Mountains, there is greater abundance and diversity of songbirds (Berger et al. 2001). This is because when wolves are present there are fewer moose present in the tree-lined riparian zones of streams. Moose are voracious herbivores, and they can lower plant diversity in the riparian zone by over-grazing preferred species of trees and shrubs. When wolves are present the moose densities go down, plant diversity goes up, and as a result the diversity of rare songbirds that are dependent upon the riparian habitat also increases. The interdependence of species in ecological communities is also illustrated by the example of beaver creating habitat for dozens of other species and increasing ecosystem productivity as described in Chapter 2. Even the oxygen in the atmosphere that all animals require accumulated from the photosynthetic action of algae in the oceans over the millennia.
Figure 6.1: Wildlife of the Greater Yellowstone Ecosystem exemplify the interdependence of species in ecological communities. Top predators such as the grey wolf (pics 1,2, starting top left) and grizzly bear (pic 3) reduce the grazing by large herbivores such as elk (pic 4,5) and moose (pic 6,7) in riparian zones. This in turn increases the abundance of bird species, including the calliope hummingbird (pic 8) and Wilson’s warbler (pic 9), by increasing the vegetation that provides their habitat. Sources: Pics 1 and 5; Yellowstone National Park website. Pic 6; Rocky Mountain National Park website. Pics 2,3,4,7,8; Gerald and Buff Corsi © California Academy of Sciences. Pics 9,10; Dr. Lloyd Glenn Ingles © California Academy of Sciences.
All of these ecological interdependencies are crucial to the persistence of wildlife, and all other forms of multicellular life on earth. However, the fact that life is dependent upon interactions in ecological systems in which the organisms are not acting for the benefit of the whole gives ecological systems some distinct, and disturbing (to us), characteristics. Thus, alteration of ecological systems is often irreversible (at least within the time frame of human lives), and in general the exact response of ecological systems to perturbations is difficult to predict. Humans accelerate the processes of ecosystem change and push them in unanticipated directions, particularly through alteration of physical processes. This change often has undesirable consequences for humans, which is why we need to take such care to protect ecosystems and biodiversity. Scientific investigation of ecosystem properties can be challenging, but ecologists are making significant progress in better understanding life at larger scales of organization.
1Actually, there are fairly good explanations for this as well. Perhaps most fundamentally, most cancers don’t affect a person until after they have passed breeding age. Thus, by the time a person has cancer, he/she has already transmitted his/her genes to a subsequent generation, and there is no selection against developing such cancer.
CONTROL OF FIRE DISRUPTS NATURAL PROCESSES
The policy of suppressing forest fires in the western United States, which started around the beginning of the 20th century, illustrates how ecological systems can have unexpected and irreversible responses to human alterations of physical processes. The trees of many western forests are fully adapted to experiencing regular fires in their environment (Barbour et al. 1993). The large older trees have thick bark that makes them resistant to all but the largest fires, and many species cannot even reproduce without fire because the cones will not open unless they are heated to high temperatures. The montane forests of the Sierra Nevada, including the Giant Sequoia forests, are examples of these fire-adapted ecosystems. Under pre-European conditions there were frequent small fires in these forests, and the forests consisted largely of large, mature trees and an open, grassy understory. It was easy to walk through these forests, and both lightning and Native Americans often set small fires that helped keep the forests in this state.
Early European-American forest managers did not understand the role of fire in these ecosystems, and saw it as detrimental to both the forests and the wildlife. They instituted the policy of fire suppression, and as a result the open under-stories began to fill in with smaller trees that would not have been able to survive small fires. This diminished the diversity of wildlife because the open grassy understory was not available to deer and other species that need such habitat. After a few decades of fire suppression there were so many small trees in the forests that there was a high fuel load. In addition, these smaller trees provided a staircase by which fire could get up high enough to reach the crown of the large, mature trees in the forests, which were otherwise immune to smaller fires. In essence, the policy of fire suppression set the stage for enormous raging fires that can destroy every tree in the forest, and surrounding human developments as well. The late 20th and early 21st centuries have seen enormous forest fires throughout the western US that cost extraordinary sums to suppress. It may be quite difficult to return unburned forests to their previous fire regimes of small frequent fires because of the fuel loads that have built up. Indeed, ecological dynamics in these forests have been shifted towards less frequent, high intensity large fires instead of more frequent, low intensity small fires by decades of fires suppression. It addition, the diversity of tree species has declined as the forests shift towards increased dominance by shade-tolerant species such as white fir.
Clearing out the small understory trees might help return these forests to their previous state, but this is expensive because small trees have little economic value. The large, old trees are economically valuable, however, so unfortunately fire risk is increasingly used as an excuse to log large trees in the name of fuel load reduction on both public and private land. Logging the large trees results not only in a major loss of wildlife habitat, but it may actually make the forests less fire resistant in the long run because it is the most fire-resistant trees that are removed. Greater intensity and frequency of forest fires also is predicted to occur with global climate change. Due to changing fire regimes it is unlikely that the forests and wildlife of the western US will ever be what they were before intervention by European-Americans, even in protected areas like national parks.
Figure 6.2: Top: Elk in the Bitterroot River observe a raging fire in Montana in 2000. Source: Alaskan Type I Incident Management Team. Photographer: John McColgan Bottom: The largest trees survive a smaller ground fire in Bitterroot National Forest. Source: National Interagency Fire Center www.nifc.gov/gallery
EMERGENT PROPERTIES AND ECOLOGICAL PROCESSES
The concept of emergent properties leads us to some important insights into the scientific investigation of ecological systems. It is readily apparent that cells are composed of molecules, organisms are composed of cells, populations are composed of organisms, communities are composed of populations of species, ecosystems encompass many interacting communities, and the biosphere is composed of all the ecosystems on Earth. Hence each scale entirely includes the units of the previous scale. But are the properties at each scale predictable from an understanding of dynamics at lower scales? The concept of emergent principles emphasizes that while each higher level may be composed entirely of units from the next lower scale, it possesses distinct properties that are unique to that level, and that cannot be entirely explained by processes at the lower level.
Populations are composed of individual organisms. Part of the dynamics of a population may be explained by understanding the physiology of individual organisms, the next scale down. For example, a long cold winter with lots of snow in the Sierra Nevada can keep lakes covered with ice and snow much longer than usual, preventing light from penetrating and allowing the growth of algae. As a result the trout that live in these lakes may use up all the oxygen in the water and die. However, many of the dynamics commonly observed in populations, such as irruptions, crashes, cycles, and constant densities in a highly variable environment, can only be explained from the interactions between organisms and their larger environment. For example, territorial behavior will often stabilize wildlife populations because it spreads animals, at least in theory, more evenly across the landscape. Many birds and mammals are territorial, particularly carnivorous species. The marten is one such territorial carnivore, an arboreal (tree-climbing) member of the weasel family that lives in the coniferous forests of Canada and the mountainous western United States, where it specializes in hunting squirrels and other small mammals. A marten will aggressively attack another marten that enters its territory, especially when prey densities are low. Individual martens that do not have territories have much lower survival and reproduction rates, because they are much more likely to starve or get attacked by predators. The marten populations of Algonquin Park, Ontario were observed to be surprising constant through time and this stability is probably in large part due to the territoriality of this species (Fryxell el al. 1999). The fact that these population dynamics emerged from population level processes illustrates the idea of emergent properties.
The concept of emergent properties means that many important properties at higher scales of biological organization cannot be entirely explained by understanding lower scales; in other words, “the whole is more than the sum of its parts.” Of course, often times much can be explained by looking to the next lower scale of biological organization. The concept of emergent properties may be especially important in ecology because ecological phenomena occur on the scale larger than the predominant scale of natural selection, and as such interactions between organisms and emergent processes may play a larger role. The challenge to biologists is to look for causality at both lower levels and in properties unique to the scale of observation, recognizing emergent properties that occur in the nested hierarchy of biological systems.
Figure 6.3: The territorial marten (also known as the pine marten or American sable) in its natural habitat in Montana (left) and the Northwestern Territory of Canada (right). Sources: (left) Gerald and Buff Corsi © California Academy of Sciences, (right) www.nwtwildlife.rwed.gov.nt.ca.
REDUCTIONIST VS. HOLISTIC SCIENCE
The concept of emergent properties integrates two different approaches to science. Reductionism seeks to understand phenomena by “reducing” them to their parts, essentially looking for explanation at the lowest scales of organization. This is the traditional approach of Western science, and it has lead to some breathtakingly impressive explanations for numerous phenomena. Physics and chemistry are largely reductionist sciences, and reductionism is generally the main approach in molecular and cell biology. The alternative scientific approach has traditionally been called holistic science, a term which suggests the idea that “the whole is larger than the sum of its parts.” Holistic science looks for explanation at the same or larger scale than the phenomenon in question. Unfortunately the term “holistic” is also used by many by non-scientists to indicate all kinds of “fuzzy” thinking and pursuits that have very little to do with the highly technical scientific search for principles of causality from higher organizational scales. Scientific understanding of holistic causality and emergent phenomena is a new approach in contrast to reductionism, and as such is not as well-developed as a method. The cutting edge of theoretical research in this area can be quite complex and quantitative, involving teams of mathematicians and scientists and large computing facilities, very remote from the birds singing in mountain forests.
Holistic scientific explanations are particularly important to include in ecology because ecological systems provide the larger scale context in which many biological processes occur, and hence serve as the basis for the holistic explanations of many phenomena at organismal and lower scales. In addition emergent phenomena, which arise from interactions among organisms, likely are quite important within ecological systems. Thus the structure observed in discrete assemblage of animals is the result of interactions (predation, competition, disease, symbiosis) among them on both long-term (evolutionary) and short-term (ecological) scales. The emergent property often is an apparently stable (persistent through time) community in which each member has a distinct niche with little overlap with other community members (see Chapter 7). In California streams, for example, two small fishes that live in fast water, speckled dace and riffle sculpin, live in different parts of the stream and feed on different invertebrates. When sculpins are removed, however, the dace take over the places where sculpin lived previously and develop a broader diet. They are constrained by the aggressive behavior of sculpin who drive them away from the prime habitat. The pattern of segregation, however, is predominant in most streams.2
It is important to emphasize again that both reductionist and holistic explanations are important to biology and ecology because the causes of scientific phenomena can occur at both smaller and larger scales. The degree to which either holism or reductionism is sufficient to explain something may be dependent upon the system in question, and for some questions reductionism may be all that is needed. However, the nature of ecological systems often demand looking for explanation at both larger and smaller scales, as illustrated in the kangaroo rat example on the next page.
2See chapter 7 for the distinction between the fundamental niche and the realized niche, which this illustrates; in this case, the fundamental niche of the dace is constrained by interaction with sculpin.
ULTIMATE AND PROXIMATE CAUSALITY, HOW VERSUS WHY
So far we have discussed how the cause underlying biological phenomena can be rooted both in events occurring at the next lower scale of organization, as well as interactions with factors at the same or higher scales of organization. These two different forms of causality result from principle of emergent properties in biological organization, and represent the fundamental difference between reductionist and holistic philosophies of scientific inquiry. Another way that these two approaches are described is by the terms ultimate and proximate causality. The proximate cause is the mechanism that allows for something to happen, and the ultimate cause is the drive that makes it happen. These can be summarized as how (proximate cause) and why (ultimate cause). Within in the context of biology, the proximate cause is nearly always the reductionist explanation from the next lower scale. The ultimate cause is the conditions in the environment that place certain constraints on the organisms, and the selection based drive for the organisms to survive in the context of that environment and their interactions with other organisms. The following is an illustration of now ultimate and proximate causality work together to explain the biology of some desert rodents of North America kangaroo rats and pocket mice.
Why don’t kangaroo rats drink water? This will be the question around which we demonstrate the concepts of ultimate and proximate causality in biology; note that this question occurs at the scale of an individual organism. Kangaroo rats are small rodents adapted to life in the desert. Like many desert creatures they live in burrows underground during the day to escape the desert heat (and predators), and they are active at night. They have both longer life spans and slower reproductive rates than typical rodents of their size, which is probably an adaptation to the unpredictable food supply and harsh conditions of the desert. These animals eat predominantly seeds, which they gather in fur-filled pouches on either side of their mouth. They store these seeds in caches, from which they can later feed for many months. This seed caching behavior is also an adaptation to the low productivity and unpredictability of desert ecosystems. These animals are predominantly found in the very arid areas of North America, including parts of the desert that get less than 5 inches of rain on average per year, with dry years that can be as low as 1 inch per year (or less!). For the most part, therefore, there is little standing water available to them because even most streams are dry in most of the year. When these animals are kept in captivity in research labs they commonly do not drink water that is given to them and have even been known to bury their water dishes with sand. It should be noted, however, that the dozen or so species of kangaroo rat do vary somewhat in their ability to live without water.
Figure 6.3 Merriam’s kangaroo rat (right) is highly tolerant of dry desert conditions, and is found in the sandy creosote bush habitat of the Colorado Desert, shown here at the University of California Boyd Deep Canyon Desert Research Center. Kangaroo rat photo source: Dr. Lloyd Glenn Ingles © California Academy of Sciences
Explanations for why kangaroo rats do not have to drink water can be found at all the scales of biological organization, and all of them are correct. First we will look to lower scales of biological organization for explanation. Kangaroo rats have unusually large kidneys that allow them to remove nearly all of the water from their urine before excretion (Schmidt-Nielsen 1997). This is the result of an elongated loop of Henle within the kidneys, which in turn are the result of genes that code for proteins that form those structures. Kangaroo rats are also very efficient at utilizing water that they literally produce as a byproduct of carbohydrate metabolism from other cellular processes, and they preferentially eat seeds that have higher carbohydrate levels and lower protein levels. These explanations on the physiological, cellular, and genetic scales all qualify as proximate causes for the fact that kangaroo rats do not need to drink water. This is also a reductionist approach to explaining a biological phenomenon on the organismal scale.
What about explanations at and above the scale of the organism itself? These start with the classic natural selection argument that the reason that kangaroo rats do not need to drink water is that those animals that are less dependent on standing water have a higher probability of surviving and reproducing in desert conditions. Through time natural selection should favor those individuals that were more drought tolerant, and hence eventually led to the evolution of species that do not need to drink standing water at all. This process occurs because of interactions between individuals of the same species, so evolution by natural selection is a population scale explanation, one scale higher. We can also look for explanations on yet higher scales. Given that the desert is a stressful environment, why are kangaroo rats even living there at all? Why are they not living in a place with more rainfall, so that they would not have to have all these special adaptations that allow them to live without drinking water? The answer to this question is on the community scale of biological organization.
The reason kangaroo rats live in an environmentally extreme habitat is competition with other species of rodents. Kangaroo rats presumably could survive in many more productive habitats that have more food and water. However, there are other species in those habitats that are better competitors under those less stressful conditions, probably in large part because they are faster reproducers. Trade-offs exist between different species traits, so the very traits that allow a kangaroo rats to thrive in the desert may not make them good competitors under less arid conditions. The long life spans and slow reproductive rates that are adaptive in the harsh, unpredictable desert environment would result in lower reproduction and therefore lower competitive ability is less harsh environments. Furthermore, the energy invested in the physiological structures, such as enormous kidneys, that allow for kangaroo rats to live without standing water is energy that will not be invested in reproduction. Therefore, one reason why kangaroo rats do not need to drink water is lack of interspecific competition (competition between species), which is a community scale explanation.
It is also possible to find an explanation for why kangaroo rats do not need to drink water at the global biosphere scale. If there were no deserts in North America to begin with, then there would be no need for kangaroo rats to be so drought tolerant. There are deserts in North America because of the nature of the Hadley cell global circulation patterns, which create deserts around the world at approximately 30° latitudes. This is discussed in much more detail in Chapter 3, but suffice it to say that if these global circulation patterns did not exist, the deserts of North America would be greatly reduced (recall that the Great Basin is a rainshadow desert, so it likely would still be present in this imaginary Hadley cell-free world), and at the very least we would expect fewer species of kangaroo rats, or perhaps none at all.
PROXIMATE AND ULTIMATE CAUSALITY IN YOUR LIFE
The concepts of ultimate and proximate causality also are relevant to situations other than biology. For example, imagine that you have an internship at a government agency in Sacramento (which, by the way, is a great opportunity available to UC Davis students), and that you are on the Yolo Causeway driving your car to work. If we were to ask the question “what is causing you to drive to Sacramento?” we could give different explanations based on proximate versus ultimate causes. The proximate cause behind your driving to Sacramento is the workings of your car engine, because the fact that it is burning gasoline and turning chemical energy into mechanical and kinetic energy is what is causing you to move toward Sacramento. In other words, this proximate cause explains how you are driving to Sacramento. The ultimate cause considers why you are driving to Sacramento. It starts by looking at the benefits to you as an individual for going to your job; you are driving to Sacramento because you need to make money, and because you are gaining valuable job experience to help your future career. Of course, there are larger-scale ultimate causes in addition to this, such as that you are fulfilling a role in the government agency, and that agency is in turn fulfilling a role in the governing of California that was deemed important by policy makers. If any of these ultimate or proximate causes did not exist, such as if the car engine did not work, if you did not benefit as an individual from the job, or if the agency had no need for your labor, then you would not be driving to Sacramento. Notice that the proximate causes focus on smaller scale explanations, such as the mechanics of your car, while the ultimate cause are larger scale, such as your individual economic needs and the existence of the job to begin with. The parallels with the ultimate and proximate causes in the kangaroo rat example should be readily apparent.
ALL YOU NEED TO KNOW TO BE A PRINCIPLED ECOLOGIST
In summary, some of the general principles of ecology presented in this essay are as follows:
- Ecology focuses on the biological scales of the individual organism as well as larger scales, including populations, communities, ecosystems, and the biosphere.
- The principle of emergent properties states that while each successively larger scale is composed of the units of the next smaller scale, it possesses properties unique to that scale.
- Causal explanations can occur at scales smaller than, the same as, or larger than the scale of the observed phenomenon.
- Causes at smaller scales are called reductionist or mechanistic explanations, and are examples of proximate causality.
- Causes at the same or larger scales are called holistic explanations or emergent phenomena, and are examples of ultimate causality.
- Essentially all species, including wildlife, in ecological systems are dependent upon other species for their existence.
- In spite of this interdependence, the organisms within ecological systems nearly always act to maximize their individual fitness, not to benefit the population, community or ecosystem.
- Constant change is a commonality of all levels of organization in ecology.
- Ecosystems are altered by human manipulations of the environment and these changes are often irreversible.
The following chapters will give you more detailed information of principles in population, community, and ecosystem ecology that will be quite useful for studying wildlife, and for understanding the consequences of human alterations of natural systems.
Barbour, M. B. Pavlik, F. Drysdale, and S. Lindstrom. 1993. California’s Changing Landscapes. California Native Plant Society, Sacramento, California.
Berger, J. P. Stacey. L. Bellis, and M. Johnson. 2001. A mammalian predator-prey imbalance: grizzly bear and wolf extinction affect avian neotropical migrants. Ecological Applications 11(4) 947-960.
Fryxell, J., J. Falls, E. Falls, R. Brooks, L. Dix, and M. Strickland. 1999. Density dependence, prey dependence, and population dynamics of martens in Ontario. Ecology 80(4): 1311-132.
Levin, S. 2002. Complex adaptive systems: exploring the known, the unknown, and the unknowable. Bulletin of the American Mathematical Society 40(1): 3-19.
Masters, Gilbert M. 1996. Introduction to Environmental Science and Engineering.
Schmidt-Nielsen, K. 1997. Animal Physiology: Adaptation and Environment, 5th Ed. Cambridge, UK: Cambridge University Press.
Williams, G. 1966. Adaptation and natural selection: a critique of some current evolutionary thought. Princeton University Press.
Wilson, D. S. 1980. The natural selection of populations and communities. Benjamin/Cummings Publishing.