Nature and Human Society: The Quest for a Sustainable World (1997)
Chapter: The Loss of Population Diversity and Why It Matters
Page 71
The Loss of Population Diversity and Why it Matters.
Biodiversity encompasses variation at all levels of biological organization, including individuals, populations, species, and ecosystems (Wilson 1988), yet much of the current scientific and public concern over the extinction crisis is focused on the loss of species. The rate of species extinction, however, reflects only one aspect of the loss of biodiversity and its consequences. What if no further species became extinct, but every nonhuman species suddenly were reduced to a single, minimal population? Although global species diversity would remain unchanged, the planet would be largely devoid of life, and civilization as we know it would collapse. This is because many of the benefits that biodiversity confers on humanity are delivered locally, through populations of species. This extreme scenario highlights the idea that species, although important, are not the only dimension of biodiversity that we should be concerned about losing.
In this paper, we examine the consequences of the gradual extinction of populations that is occurring today. First, we discuss the importance of populations to humanity. Then, we present estimates of population diversity, that is, the number of populations on Earth. Finally, we make a preliminary attempt to evaluate the rate of populations extinction.
What is a Population?
Populations are geographical entities within a species, usually distinguished ecologically or genetically (Ehrlich and Daily 1993). The ecological entity is a demographic unita group of individuals whose population dynamics are not
Page 72
influenced substantially by migration from nearby conspecific groups; that is, the fluctuations in the size of the population of one group are independent of those of other groups (Brown and Ehrlich 1980). The genetic entity is a Mendelian population (Sinnott and others 1950), defined here as a genetically distinguishable group of individuals that evolves independently of other groups. Demographic units may be Mendelian populations and vice versa, but the two are not necessarily congruent. As with species, both kinds of populations exist as parts of continua in space, rather than as clear, discrete units.
We adopted the Mendelian-population definition for our estimates of population diversity and its extinction rate for two reasons. First, the Mendelian definition directly includes the genetic variation between groups of individuals, and, as discussed below, this variation is of great importance to humanity. Second, we found that quantitative data on genetic-population structure was more comparable across species and investigations than were quantitative data on demographic-population structure.
The Importance of Populations
Why should one be concerned about the extinction of populations? Much has been written about the ethical and practical reasons for halting the species-extinction crisis that is driven by human activities (Ehrlich and Ehrlich 1981; Ehrlich and Wilson 1991; Myers 1979; Wilson 1992). Certainly, we agree with these reasons, yet simply arguing for saving species obscures an essential link between biodiversity and human welfare. Ultimately, most of the benefits that biodiversity confers on humanity are delivered through populations. These benefits include aesthetic enjoyment, discovery and improvement of pharmaceuticals and agricultural crops, species conservation, replenishment of stocks of economically valuable species, and, perhaps most important, delivery of ecosystem services.
Aesthetic Value
Natural ecosystems are composed of populations of organisms, their physical environments, and the interactions between them. As such systems are disrupted or destroyed, people's enjoyment of their ambience and the aesthetic values of their component populations (for example, birds, butterflies, reef fishes, flowering plants, and shade trees) is diminished. In addition, the total aesthetic value of individual species declines as their populations disappear, although the aesthetic value of “rarity” may partially, and somewhat paradoxically, compensate for this loss. For instance, although wild populations of grizzly bears and remnants of old-growth redwood forest exist in the United States, the total aesthetic benefit conferred on Americans by watching a grizzly cub play or by hiking in a cathedral-like, old-growth redwood forest is relatively small because few people can experience them firsthand.
Genetic Value
Much of the genetic diversity in species exists as genetic differences between populations. For an average animal species, 25–30% of its total ge-
Page 73
netic variability is due to differences between populations. In an average outcrossing plant species, 10–20% of its genetic variability occurs among populations, whereas a selfing plant species exhibits about 50% of its genetic variability among populations (Hammond 1995). One result of this differentiation is that populations of the same species may produce different types or quantities of defensive chemicals (Dolinger and others 1973; Goméz-Pompa and others 1972; Hwang and Lindroth 1997), compounds that may have medicinal value.
An example of how genetic variation among populations is important to pharmaceuticals is the story behind the development of penicillin. The successful development of penicillin as a therapeutic drug did not occur until 15 years after Alexander Fleming's discovery of the compound in common bread mold. One reason for this delay was a worldwide search to find a strain (that is, a population) of the mold that produced greater quantities of penicillin than the original strain produced (Dowling 1977).
Population diversity among wild relatives of crops also supplies critical genetic material to agricultural strains. Genetically uniform strains of the world's three major crops (wheat, rice, and maize) are planted widely; as a result, large fractions of the harvest can be threatened at one time by a new disease or pest (Plucknett and others 1987). Thousands of strains, or populations, of wild relatives of crops may need to be tested until one is found that carries the desired genetic resistance that can be used to protect the crop. For example, when the grassy stunt virus emerged as a serious threat to the rice crop in Southeast Asia in the late 1960s and 1970s, an extensive search for resistant varieties of rice was conducted at the gene bank of the International Rice Research Institute. Five thousand accessions from populations all over the world and 1,000 breeding lines were screened. Only one accession of a wild rice collected in India was found to resist the virus (Plucknett and others 1987). Genetic variation in wild populations of crop species also will be crucial in providing genetic material to sustain yields with changing growing conditions, especially climate (Daily and Ehrlich 1990).
Species-Conservation Value
By definition, populations are essential to the conservation of species diversity, and the number and size of populations influence the probability of persistence of the entire species. Migrants between populations can prevent the local extinction of a species by contributing critical individuals when numbers are low (the rescue effect) (Brown and Kodric-Brown 1977) or by supplying the genetic variation needed to adapt to changing environmental conditions (Lande 1988). If local extinction does occur, individuals from other populations can recolonize the area. The threat of rapid global climatic change makes the safety net of population diversity for species even more important; a species that has many populations is more likely to include individuals that are genetically suited to new conditions than is a species that has only one or a few populations (Kareiva and others 1993).
Page 74
Direct Economic Value
Destruction of populations of an economically valuable species not only increases the probability that the species will become extinct in the near future but also may decrease the species' harvest level. In the short term, as populations are exterminated, fewer will remain to be harvested; in the longer term, when a species is composed of a metapopulation, the stock levels of the remaining populations also may decline (Pulliam 1988). The reduction of these economically important species often has direct consequences for local peoples. For instance, overharvesting of oceanic fish stocks and the resulting decline in yields lead to loss of income to fishermen and loss of an important source of protein for much of the human population (Kaufman and Dayton 1997; Peterson and Lubchenco 1997; Safina 1995).
Ecosystem-Service Value
Perhaps the most important benefit that populations confer on humanity is ecosystem services. Ecosystem services include natural processes, such as purification of air and water, detoxification and decomposition of waste, generation and maintenance of soil fertility, pollination of crops and natural vegetation, and control of pests (Daily 1997). These services are provided by populations, and population diversity (that is, the number of populations) at global, regional, and local levels affects the provisioning of ecosystem services. (The size and density of populations also influence the provisioning of ecosystem services. These dimensions will be discussed later. For now, we simply address numbers of populations.)
Greater global population diversity probably enhances the delivery of global ecosystem services, such as regulation of biogeochemical cycles and stabilization of climate (Alexander and others 1997). The larger the area that remains under natural tree cover in the Canadian taiga, the greater the amount of carbon stored there. Although deforestation in this region might not result in the extinction of any species, a large-scale loss of tree populations would influence the balance of greenhouse gases in the atmosphere worldwide (Woodwell and others 1983).
For many ecosystem services, however, global numbers of populations are not as important as regional population diversity. In other words, for these services, it is not only necessary that many populations exist somewhere in the world but also that they exist within the region of interest. These services include, for instance, mitigation of floods and droughts by forests and purification of water by forests and wetlands (Ewel 1997; Myers 1997). Loss of these services occurs when forests and wetlands are destroyed in a region, regardless of the continued existence of their component species elsewhere. New York City provides an excellent example of the value of regional population diversity. The city was famed for its pure water, which came from the Catskill Mountains, 100 miles to the north. For most of the city's history, natural purification processes, which are carried out by populations of soil organisms and plants, were sufficient to cleanse the water, but in recent years, land development and associated human activities reduced the efficacy of these processes. In 1996, city water officials floated an environmental bond issue to purchase land, freeze development on other lands,
Page 75
and subsidize the improvement of septic tanks in the water-supply area. It is hoped that these actions will restore and safeguard the local populations that filter and purify the water. If so, an investment of $1 billion in natural purification services will have saved city taxpayers $6–$8 billion, the additional avoided cost (over 10 years) of building a water-treatment plant (Chichilnisky and Heal 1998).
Regional population diversity is also necessary for control of pests. The importance of populations that serve a pest-control function is illustrated dramatically when an organism is transplanted to a new environment that lacks populations of predators capable of keeping it from becoming a pest. The importation of the prickly pear (Cactus opuntia) into Australia by early settlers is a classic case. Apparently originally intended as an ornamental plant, in the absence of its normal predators the cactus spread over vast areas. It occupied some 25 million hectares in New South Wales and Queensland, and half the area was covered so densely that the land could not be used for farming or ranching. The costs of poisoning or removing the cactus were more than the land was worth. The problem was solved eventually by importing a moth that is a voracious cactus-eater from the South American homeland of the opuntia. Once regional populations of that moth, Cactoblastis cactorum, were established, the cactus was decimated and the problem was solved. Although the cactus still can be found in Australia, it occurs only in scattered clumps since natural pest control has been reestablished (Ehrlich 1986).
Pollinators are critical to agriculture, and the decline of regional populations of native pollinators, chiefly as a result of pesticides and destruction of habitat, has not gone unnoticed (Buchmann and Nabhan 1996). For more than 60 crops planted in the United States, farmers are forced to pay keepers of the European honeybee to transport their hives to the fields or orchards that require pollinating. Hiring beekeepers costs farmers more than $60 million a year and the federal government more than $80 million in subsidies, and these numbers are still increasing because of growing problems in the beekeeping industry (disease and hybridization with the aggressive Africanized honeybee) (Nabhan and Buchmann 1997).
Population diversity at a particular location (that is, local species diversity) also affects ecosystem functioning and thus the delivery of ecosystem services (Chapin and others 1997). In greenhouse and field experiments, plant productivity has been found to increase with species diversity (Naeem and others 1994). The stability of plant productivity also has been linked with greater richness of species. More diverse grassland plots seem to be more resistant to drought and grazing disturbances than less diverse plots (McNaughton 1977, Frank and McNaughton 1991; Tilman and Downing 1994). Thus, it appears that local population diversity is closely coupled to local ecosystem functioning.
Because regional and global services are performed by an aggregate of local ecosystems, the consequences of a reduction in local population diversity probably will extend beyond the local ecosystem. In other words, the loss of populations from one location, which alters the functioning of the local ecosystem, may in turn affect the delivery of larger-scale services. For example, the global carbon
Page 76
cycle may be influenced not only by the total number of tree populations on the planet but also by the diversity of populations at many locations.
One important question that remains to be resolved is the extent to which “weedy” species, spreading into and establishing populations in areas where native populations have been extirpated, can continue to supply ecosystem services. For such services as pest control, evidence is abundant that such compensation will be rare. The cotton disaster in the Cañete Valley in Peru is a classic example. Populations of natural enemies of potential cotton pests were destroyed by repeated, heavy applications of pesticides, and no weedy species moved in to assume the role of the natural predators. As a result, numerous obscure organisms became pests and destroyed the cotton crop (Barducci 1972).
For other services, such as flood control and soil retention, the potential for substitution by weeds, at least in the short term, sometimes may be high. In many cases, however, we are largely ignorant of the ability of weeds to maintain services over the long run. Furthermore, the capacity for large-scale technological substitution of ecosystem services appears limited (Ehrlich and Mooney 1983). The Biosphere 2 project, a materially closed, human-made ecosystem, is a case in point. Despite hundreds of millions of dollars invested in development and operating costs, scientists failed to engineer a system that could support eight people with food, air, and water for 2 years (Cohen and Tilman 1996; see also Daily 2000). That venture dramatically illustrated that we do not know yet how to replicate the life-support services that the mix of populations in natural ecosystems provides for free.
The Extent of Population Diversity
Given the numerous reasons to be concerned about the fate of population diversity, we recently attempted to quantify the extent of that diversity and the rate of its loss. In this section and the next, we give an overview of these calculations (for further details, see Hughes and others 1997). Again, we define population diversity as the number of populations on the planet; another aspect of population diversity is the degree of divergence among populations, but we do not consider that aspect here.
Many of the difficulties that plague attempts to estimate species diversity also hinder an estimation of population diversity. The debate over definitions of species has persisted for decades (for example, Coyne and others 1988; Dobzhansky 1935; Ehrlich 1961; Masters and Spencer 1989; Mayr 1940 and 1969), and defining a population is no simpler. Also, the small fraction of species cataloged so far (approximately 1.75 million species of 10 million or more [Hammond 1995]) represents a regionally and taxonomically biased view of the planet's biodiversity. These problems are inherent in estimates of species diversity and are inevitably present in estimates of population diversity as well. For instance, as with most estimates of species, our population estimate is restricted to eukaryotes, because information on the diversity of bacteria and viruses is almost nonexistent, although the diversity is probably enormous. Nonetheless, just as approximations of species diversity have been made despite these difficulties, enough information
Page 77
exists to allow us to make a preliminary evaluation of biodiversity at the level of the population.
Our method of estimating global population diversity involved three steps. First, we reviewed the literature on population differentiation for a broad range of taxa and estimated the average number of populations per unit area for a series of species. Then we calculated the average size of the range of a species with a sample of available species range maps. The product of the resulting two numbers yielded an approximation of the average number of populations per species. Finally, we multiplied that number by the total number of species to arrive at the number of populations on Earth.
We searched 15 journals published from 1980 to 1995 for genetic studies on population differentiation, reading more than 400 articles and finding 81 that provided appropriate data for our calculations. We were able to estimate the number of populations per unit area for 82 species. Most of the species were vertebrates (n = 35), followed by plants (n = 23), arthropods (n = 19), mollusks (n = 4), and one flatworm (platyhelminth).
To quantify the number of populations of a species per unit area, we determined whether the sampling locations described in the articles were in separate populations or were within a single population. If statistically significant differentiation between localities was reported in the paper, we considered all the localities to be separate populations. We then calculated the number of populations per unit area as the number of sampling locations divided by the extent of the entire sampling area. If the researchers did not find significant differentiation between the localities, we assumed that they had sampled from within one population and that the size of a population was the size of the sampling area. Many studies found an intermediate amount of differentiation. For instance, in some studies, a significant difference was found only between two clusters of sites. In these cases, we assumed that there were two populations within the sampling area. This procedure yielded a conservative estimate of one population per 10,000 km2 for an average species.
What are some problems with this evaluation of populations per unit area? First is the taxonomic bias mentioned above. Arthropods make up about 65% of the planet's species, and birds account for probably less than 0.01% (Hammond 1995). In our data on population structure, however, arthropods accounted for only 20% of the species, whereas birds accounted for more than 11%. Second, the evaluation of population differentiation for an average species is limited by the sampling intensity of each study. In other words, the estimate is probably conservative, since in many cases additional sampling in the study area may have revealed further differentiation. Finally, the molecular markers chosen may not always reveal notable differences between groups (for example, Legge and others 1996), again making the estimate on the conservative side.
To estimate the average range of a species, we digitized more than 2,400 species range maps from guidebooks for birds, mammals, fishes, and butterflies from a number of geographical regions. Equally weighting the four taxonomic groups, the mean size of the range of a species is 2.6 million square kilometers. Averaging the range size estimates of the largest group, the arthropods (here just butter-
Page 78
flies), led to a range of 2.2 million square kilometers per species. These numbers are quite similar, so we conservatively used the lower number, 2.2 million square kilometers, as our estimate of the average size of the range of a species.
This evaluation of the average size of the range of a species is the most probable source of inflation in our estimate of population diversity. The shaded areas on distribution maps virtually always encompass unsuitable habitats, where populations do not occur (Gaston 1994). Also, the majority of sources we used were limited to temperate regions, even though it is estimated that two-thirds of species diversity exists in the tropics (Raven 1983). This misrepresentation also may inflate the population estimates because, in some taxa, the sizes of species ranges tend to increase toward the poles (Pagel and others 1991; Rapoport 1982).
One aspect of our method may compensate somewhat for these biases, however. The sources we used restricted their species range maps to one continent, so the full range of intercontinental species was not taken into account. Therefore, we may have underestimated considerably the size of the range of some species, such as birds that have Holarctic ranges.
The product of the estimates of the average populations of a species per unit area and the average size of the range of a species was an average of 220 populations per species. Using three published calculations of global numbers of species (5, 14, and 30 million from, respectively, Hammond 1995, Raven 1985, and Erwin 1982), we arrived at three estimates of the total number of populations: 1.1, 3.1, and 6.6 billion populations.
Population Extinction
In presenting the methods of our estimation of the current rate of population extinction, it is useful to begin with a summary of how species extinction rates usually are assessed. Estimates are derived largely from species-area relationships and from the rate habitat loss due to deforestation (Lawton and May 1995; Wilson 1992). The most commonly used species-area model is S = cAz, in which S is the number of species, c and z are constants estimated from empirical studies, and A is the area where the species are found (Rosenzweig 1995; see also Pimm and Brooks this volume). This relationship between area (size of the habitat) and number of species is illustrated in figure 1. The graph reveals a convenient rule of thumb: a 90% decrease in area of habitat should result in roughly a 50% decrease in species diversity.
By applying estimates of rates of tropical deforestation to this model, one can approximate the rate of species extinction in tropical forests. With a very conservative estimate of tropical deforestation of 0.8% per year, the rate of extinction of tropical forest species is predicted to lie between 0.1% and 0.3% each year, depending on the value of z used in the species-area model. If we assume that 14 million species exist globally and that two-thirds of all species exist in tropical forests, species diversity in tropical forests is declining by roughly 9,000–26,000 species per year, or 1–3 species per hour (this last calculation was reported incorrectly in Hughes and others 1997).
Page 79
No comparable work relates numbers of populations to area of habitat. Although a wide range of relationships could be justified, depending on the spatial and time scales considered, in the absence of information we used the simplest and most intuitive, namely, that changes in population numbers and area correspond in a roughly one-to-one fashion in ecological time. That is, when 90% of an area is destroyed, about 90% of the populations in the original area are exterminated (figure 1). The basis of the difference between the population-area relationship and the species-area relationship is the size of the unit. When a population is destroyed, other populations of the species still may exist elsewhere. Thus, initially the population-loss curve in figure 1 is steeper than the species-loss curve. Eventually, however, when the last populations are destroyed, all the species become extinct as well, and the curves converge.
Figure 1
Predicted species-area and populations-area relationships. The species curve
(- • -) is S = cAz, where S is the number of species, A is the area size, and
c and z are constants. Here, z = 0.30, so a 90% decrease in area corresponds
to a 50% decrease in species diversity. In contrast, the population curve is linear, so a 90%
decrease in area corresponds to a 90% decrease in population diversity.
Page 80
If, indeed, a one-to-one population-area relationship exists, the rate of population extinction in tropical forests is estimated at 0.8% per year, directly proportional to the rate of habitat loss. Using our mid-range estimate of global population diversity (3.1 billion populations) and assuming that two-thirds of all populations exist in tropical forests (simply because species are distributed in this way), we estimate that 16 million populations per year, or roughly 1,800 per hour, are being exterminated in tropical forests alone. This is an absolute rate of 3 orders of magnitude higher and a percentage rate 3–8 times higher than conservative estimates of species extinction.
Biodiversity at Different Levels.
An investigation of population diversity does not complete the picture of biodiversity. Much remains to be explored at other levels, such as genetic, individual, and ecosystem levels, all of which are tightly interrelated. Little is known about how these different levels of biodiversity relate to ecosystem functioning. For example, for any given population, the number of individuals, the genetic variation between individuals, and the area occupied may affect the delivery of ecosystem services and other benefits provided by that population. The number of blue spruce trees may be important for global services, whereas the density of the trees may be critical for regional flood control. Similarly, although cougars exist in the San Francisco Bay area, the number of individuals is so low that numbers of local deer remain largely unchecked by these natural predators.
The effect of humans on natural areas is so extensive that every level of organization of biodiversity is threatened, even ecosystem diversity. In North America, for instance, the World Wildlife Fund estimates that 32 of a total of 116 ecoregions (that is, ecosystem types) in North America are critically threatened (Ricketts and others 1999). The consequences of the extinction of entire ecosystem types are not known, but the effect could be far-reaching if the particular assemblages of species are important for the delivery of some ecosystem services. In other words, the destruction of ecosystem types not only may result in the loss of the populations and species contained within them, but also may result in the loss of unique processes that are generated by certain combinations of species.
Conclusions
The crisis of biodiversity is more severe than species extinction rates alone would suggest: Population extinction is occurring at a rate that is 3 orders of magnitude higher than the rate of species extinction. The rapid loss of population diversity means the loss of the benefits described above and, in particular, the loss of the life-support systems on which humanity relies. Thus, the destruction and degradation of habitat and the decline of populations are of great concern even when they do not endanger species globally.
This conclusion has direct implications for both conservation biologists and policy-makers. Biologists must emphasize to the public and policy-makers the importance to humanity of all levels of biodiversity, instead of simply species diver-
Page 81
sity. This shift will require that biologists stress the functional benefits of biodiversity rather than relying only on the charismatic appeal of individual species.
The most important message for policy-makers is one that ecologists long have recognized: Preservation of habitat is crucial for the preservation of biodiversity and the life-support systems that maintain human civilization. The current legislative focus on species conservation neglects crucial dimensions of biodiversity. To protect the benefits that humanity derives from biodiversity, an Endangered Biodiversity Act would be more appropriate than an Endangered Species Act.
Policy-makers also should be putting major effort into developing the field of restoration ecology (Ehrlich and Daily 1993). Unlike species, populations often can be re-established in a relatively short time, and they sometimes evolve with substantial genetic differences from the source populations (Johnston and Selander 1971). Thus, it may be possible to alleviate some of the effects of population extinction, but funds are needed to encourage this line of research.
Finally, the development of an economic-accounting system that internalizes the values of ecosystem goods and services (Costanza and Folke 1997; Goulder and Kennedy 1997) seems critical for the implementation of these policies.
Acknowledgments
We thank Carol Boggs, Anne Ehrlich, Jessica Hellman, Claire Kremen, John-O Niles, and Taylor Ricketts for many constructive comments. This work has been supported by Peter and Helen Bing, the Pew Charitable Trusts, the Winslow Foundation, and the late LuEsther Mertz.
References and Recommended Reading
Alexander SE, Schneider SH, Lagerquist K. 1997. The interaction of climate and life. In: Daily GC (ed). Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. p 71–92.
Barducci TB. 1972. Ecological consequences of pesticides used for the control of cotton insects in Cañete Valley, Peru. In: Farvar MT, Milton JP (eds). The careless technology: ecology and international development.Garden City NY: Natural History Pr. p 423–38.
Brown L, Ehrlich PR. 1980. Population biology of the checkerspot butterfly Euphydryas chalcedona: structure of the Jasper Ridge colony. Oecologia 47:239–51.
Brown JH, Kodric-Brown A. 1977. Turnover rates in insular biogeography: effect of immigration on extinction. Ecology 58:445–9.
Buchmann SL, Nabhan GP. 1996. The forgotten pollinators. Washington DC: Island Pr. Chapin III FS, Walker BH, Hobbs RJ, Hooper DU, Lawton JH, Sala OE, Tilman D. 1977. Biotic control over the functioning of ecosystems. Science 277:500–4.
Chichilnisky G, Heal G. 1998. Economic returns from the biosphere. Nature 391:629–630.
Cohen JE, Tilman D. 1966. Biosphere 2 and biodiversity: the lessons so far. Science 274:1150–1.
Costanza R, Folke C. 1997. Valuing ecosystem services with efficiency, fairness, and sustainability as goals. In: Daily GC (ed). Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. p 49–68.
Coyne JA, Orr HA, Futuyma DJ. 1988. Do we need a new species concept? Syst Zool 37:190–200.
Daily, GC. 2000. Countryside biogeography and the provision of ecosystem services. In: Raven PH, Williams T (eds). Nature and human society: the quest for a sustainable world. Washington DC: National Academy Press, p 104–13.
Daily GC (ed). 1997. Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. 392 p.
Page 82
Daily GC, Ehrlich PR. 1990. An exploratory model of the impact of rapid climate change on the world food situation. Proc Royal Soc London 241:232–44.
Dobzhansky T. 1935. A critique of the species concept in biology. Philos Sci 2:344–355. Dolinger PM, Ehrlich PR, Fitch WL, Breedlove DE. 1973. Alkaloid and predation patterns in Colorado lupine populations. Oecologia 13:191–204.
Dowling HF. 1977. Fighting infection. Cambridge MA: Harvard Univ Pr. 339 p.
Ehrlich PR, Daily GC. 1993. Population extinction and saving biodiversity. Ambio 22:64–8.
Ehrlich PR, Ehrlich AH. 1981. Extinction. New York NY: Ballantine Books.
Ehrlich PR, Mooney HA. 1983. Extinction, substitution, and ecosystem services. BioScience 33:248–54.
Ehrlich PR, Wilson EO. 1991. Biodiversity studies and science policy. Science 253:758–62.
Ehrlich PR. 1961. Has the biological species concept outlived its usefulness? Syst Zool 10:167–76.
Ehrlich PR. 1986. The Machinery of Nature. New York NY: Simon and Schuster.
Erwin TL. 1982. Tropical forests: their richness in Coleoptera and other arthropod species. Coleopt Bull 36:74–82.
Ewel KC. 1997. Water quality improvement by wetlands. In: Daily GC (ed). Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. p 329–344.
Frank DA, McNaughton SJ. 1991. Stability increases with diversity in plant communities: empirical evidence from the 1988 Yellowstone drought. Oikos 62:660–662.
Gaston KJ. 1994. Rarity. London UK: Chapman and Hall.
Global biodiversity assessment. 1995. Cambridge UK: Cambridge Univ Pr. p 113–38.
Goméz-Pompa A, Vásquez-Yanes C, Guevara S. 1972. The tropical rainforest: a nonrenewable resource. Science 177:762–5.
Goulder LH, Kennedy D. 1997. Valuing ecosystem services: philosophical bases and empirical methods. In: Daily GC (ed). Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. p 23–47.
Hammond PM. 1995. The current magnitude of biodiversity. In: Heywood VH (ed). Global biodiversity assessment. Cambridge UK: Cambridge Univ Pr. p 113–38.
Hughes JB, Daily GC, Ehrlich PR. 1997. Population diversity: its extent and extinction. Science 278: 689–692.
Hwang S-Y, Lindroth RL. 1997. Clonal variation in foliar chemistry of aspen: effects on gypsy moths and forest tent caterpillars. Oecologia 111:99–108.
Johnston RF, Selander RK. 1971. Evolution in the house sparrow. 11. Adaptive differentiation in North American populations. Evolution 25:1–28.
Kareiva PM, Kingsolver JG, Huey RB (eds). 1993. Biotic interactions and global change. Sunderland MA: Sinauer. 559 p.
Kaufman L, Dayton P. 1997. Impacts of marine resource extraction on ecosystem services. In: Daily GC (ed). Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. p 275–93.
Lande R. 1988. Genetics and demography in biological conservation. Science 241:1455–60.
Lawton JH, May RM (eds). 1995. Extinction rates. Oxford UK: Oxford Univ Pr. 233 p.
Legge JT, Roush R, Desalle R, Vogler AP, May B. 1996. Genetic criteria for establishing evolutionary significant units in Cryan's buckmoth. Cons Biol 10:85–98.
Masters JC, Spencer HG. 1989. Why we need a new genetic species concept. Syst Zool 30:270–9.
Mayr E. 1940. Systematics and the origin of species. New York NY: Columbia Univ Pr.
Mayr E. 1969. The biological meaning of species. Biol J Linnean Soc 1:311–20.
McNaughton SJ. 1977. Diversity and stability of ecological communities: a comment on the role of empiricism in ecology. Amer Natur 111: 515–25.
Myers N. 1997. Biodiversity's genetic library. In: Daily GC (ed). Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. p 255–73.
Myers N. 1979. The sinking ark. Oxford UK: Pergamon Pr.
Nabhan GP, Buchmann SL. 1997. Services provided by pollinators. In: Daily GC (ed). Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. p 133–50.
Naeem S, Thompson LJ, Lawler SP, Lawton JH, Woodfin RM. 1994. Declining biodiversity can alter the performance of ecosystems. Nature 368:734–7.
Pagel MD, May RM, Collie AR. 1991. Ecological aspects of the geographical distribution and diversity of mammalian species. Amer Natur 137:791–815.
Page 83
Peterson CH, Lubchenco J. 1997. Marine ecosystem services. In: Daily GC (ed). Nature's services: societal dependence on natural ecosystems. Washington DC: Island Pr. p 177–94.
Plucknett DL, Smith NJH, Williams JT, Anishetty NM. 1987. Gene banks and the world's food. Princeton NJ: Princeton Univ Pr.
Pulliam HR. 1988. Sources, sinks, and population regulation. Amer Natur 132:652–61.
Rapoport EH. 1982. Aerography: geographical strategies of species. Oxford UK: Pergamon.
Raven PH. 1985. Disappearing species: a global tragedy. Futurist 19:8–14.
Raven PH. 1983. The challenge of tropical biology. Bull Ecol Soc Amer Spring: 4–12.
Ricketts T, Dinerstein E, Olson D, Loucks C, Eichbaum W, Kavanagh K, Hedao P, Hurley P, Carney K, Abell R, Walters S. 1999. Terrestrial ecoregions of North America. A conservation assessment. Washington DC: Island Pr. 485 p.
Rosenzweig ML. 1995. Species diversity in space and time. Cambridge UK: Cambridge Univ Pr. 436 p.
Safina C. 1995. The world's imperiled fish. Sci Amer 273:46–53.
Sinnott EW, Dunn LK, Dobzhansky T. 1950. Principles of genetics. New York NY: McGraw-Hill.
Tilman D, Downing JA. 1994. Biodiversity and stability in grasslands. Nature 367:363–5.
Wilson EO (ed). 1988. Biodiversity. Washington DC: National Acad Pr. 521 p.
Wilson EO. 1992. Diversity of Life. Cambridge MA: Belknap.
Woodwell GM, Hobbie JE, Houghton RA, Melillo JM, Moore B, Peterson BJ, Shaver GR. 1983. Global deforestation: contribution to atmospheric carbon dioxide. Science 222:1081–6.
