In the past, avian ecologists documented simple patterns in communities and sought their explanations in the neat formulations of niche theory, Doing community ecology was exciting, fashionable, and fun. Increasingly, however, the complexity, variability, and ambiguity of nature have made community ecology more difficult and lessened its allure. In this and the previous volume, I have detailed a litany of problems, pitfalls, and perils that have plagued studies of avian communities. With the benefit of hindsight enriched by our increased awareness of the complexity of communities and the logical and methodological demands for doing science properly, it is easy to see where we may have gone astray. Such retrospection is useless, however, if we do not use it to redirect future activities in the discipline. We are faced with the prospect that many ecologists may become discouraged in their attempts to understand communities and will turn their attention to entirely different questions, while those who continue to study communities will become polarized into descriptive pattern-seekers or armchair theoreticians. In the meantime, the actual dynamics of communities may be left unattended because they are too fuzzy, too difficult to study, or not amenable to generalization or theory. We need to redirect our approach to community investigations in a way that will retain the excitement and satisfaction of the past yet not do violence to nature by oversimplification.

I attempt to provide some elements of this redirection in this chapter. Because theory is so central to our approach to community studies, I comment first on what we should expect of ecological theory. I then suggest some elements of future approaches to the study of bird communities.

The most frequently offered explanation of the patterns of communities is that they are products of interspecific competition, acting either contemporaneously or in the past. Writing in 1975, Ricklefs stated that ‘few ecologists doubt that competition is a potent ecological force or that it has guided the evolution of species relationship within communities’ (1975: 581), and, in his more recent review, Giller (1984) attributed most of the community patterns he discussed to competition. The challenges that have brought controversy to community ecology have centered largely on whether or not competition produces the patterns we see and how applicable a good deal of competition theory is to nature.

It is important to distinguish between the process of competition and its application as an explanation of community patterns. When Brown (1981) wrote that he found competition theory ‘a disappointment’, he was concerned with the theory and its application, not the process itself. Competition theory applies the process of competition under specified conditions to predict community patterns. If the predictions fail to match observations, it is not because the process is flawed but because the theory is incorrect or has been misapplied. The challenges to the competition paradigm have questioned the simplicity of the theory and its overly enthusiastic or illogical applications, not the reality of the process itself. There is nothing about the process of competition, as an interaction among individuals, that requires equilibrium or optimization or precludes opportunism, nonequilibrium, or stochastic influences. The notion that equilibrium and optimization are associated with competition derives from conventional applications of the theory that predict population-level consequences of the interactions, usually in the framework of the Lotka–Volterra equations.

Patterns in avian communities have usually been explained as the results of interspecific competition, and other processes or factors that might contribute to the patterns have been given only superficial consideration or ignored. Finding that individuals of an island population of a species differ in behavior from their mainland conspecifics, for example, ecologists often have not asked what factors might account for the differences but instead have explained the patterns solely in terms of the absence of presumed competitor species on the island. In the previous chapters, I have argued that this approach is overly simplistic and sometimes logically invalid. Even if competition is shown to occur, this does not mean that other processes or factors are absent or unimportant. Only if the patterns predicted on the basis of these other factors differ from those expected from competition can one justly conclude that competition is the primary cause of the patterns and that the contributions of other processes are unimportant. Testing such alternatives is a pivotal feature of any logical approach to scientific explanation (see Volume 1).

According to Schoener (1982), the challenges to the competition paradigm that developed during the late 1970s followed several avenues: (1) mathematical modification of the theory, (2) statistical re-evaluation of the patterns through the use of null models, (3) consideration of the influences of environmental variability, (4) emphasis of the role of predation, and (5) determination of the meaning of resource overlap. Of these areas of concern, only the fourth really emphasizes an alternative process; the first and fifth involve adjustments of competition theory, the second deals with the patterns, and the third explores the constancy or intermittency of the competition process.

Viewed at any scale of resolution, from an individual's territory to a biogeographic region or a continent, environments are mosaics of patches. Individuals and populations repond to this spatial heterogeneity in various ways, some spending long periods of time within specific patches, some using certain patches for feeding and others for reproduction, still others moving over a mosaic in an apparently aimless fashion but distinguishing between different mosaics. Because the patch structure of an environment changes through time, the responses of individuals and populations to spatial patchiness are dynamic. The structure and dynamics of communities, in turn, are strongly influenced by this spatial variation.

Kareiva (1986) observed that ecologists have had difficulty making sense of insect communities because environmental heterogeneity has generally not been included in theories of species interactions and because spatial dimensions of variation have usually been ignored in field studies. The same could be said of many investigations of bird communities. Much of the theory that has guided community studies has, for simplicity, assumed nature to be spatially homogeneous. Often these theories fail to generate realistic predictions when spatial patchiness is introduced (Schluter 1981, Tilman 1982, Chesson 1986). Likewise, field studies have often been restricted to small sites selected on the basis of their apparent internal homogeneity. In recent years, however, efforts to incorporate the influences of heterogeneity into population or community models have increased (e.g. Roff 1974, Okubo 1980, Paine and Levin 1981, Nisbet and Gurney 1982, Chesson 1981, 1985, 1986, Levin et al. 1984), and the importance of ‘patch dynamics’ has become more widely appreciated in field studies (Pickett and White 1985).

Anyone who has watched birds for several years in a woodlot or other local habitat becomes aware of variations. Some species are present in some years, absent in others, and populations change in abundance between years. Other populations vary about a long-term average that seems to be relatively stable. The abundance of wintering Stonechats (Saxicola torquata) or tits in Europe, for example, changes from year to year, apparently in response to weather and food conditions during the previous winter that affect overwinter survival and the reproductive output of the birds during the summer. Long-term trends in abundance are not evident, however (Dhondt 1983, van Balen 1980, Klomp 1980). In other cases, variations are more episodic. The winter of 1962–63 was the harshest in southern England since 1740, and populations of the Wren (Troglodytes troglodytes) were reduced by 75%, the Song Thrush (Turdus philomelos) by 50% (Williamson 1975). A prolonged drought in the Sahel (sub–Saharan Africa) between 1968 and 1974 devastated populations of wintering palearctic migrants, leading to sharp reductions in their breeding numbers. Abundances of Whitethroats (Sylvia communis), for example, decreased by as much as 77% in some parts of England between 1968 and 1969, and similar reductions were recorded elsewhere in its breeding range (Winstanley et al. 1974, Berthold 1973). When the drought broke in 1975, Whitethroat abundances rebounded, increasing by 60% in 1976 (Batten and Marchant 1977).

Over a longer time period, populations may show systematic trends in abundance.

The study of bird communities begins with the search for patterns, but rarely does it end there. A pattern, by its very existence, begs for explanation. What processes have caused it to be the way it is? The search for patterns in bird communities has usually been conducted within the framework of the MacArthurian paradigm, which focuses on interspecific competition as the major (often the only) process determining these patterns. As a consequence, patterns consistent with that view have been emphasized, even when methodological flaws render them suspect or when they cannot be distinguished from patterns that might be generated by other processes. There is thus a bias to the sorts of patterns we document and discuss. The well-worn tenets of competition theory fit comfortably over those patterns, providing explanations that are satisfying to many ecologists. Processes are inferred from patterns and then treated as demonstrated facts rather than as hypotheses awaiting tests. Unfortunately, it is considerably easier to assert the operation of a process such as competition than to test that assertion, and this has reinforced the tendency to rely on inference and assertion in process explanations.

To break away from this doctrinaire approach to interpreting community patterns, it is necessary to recognize the folly of single-factor explanations and to evaluate alternative process hypotheses. In order to test these hypotheses, however, we must consider the criteria for documenting the operation of a process with a given degree of certainty and evaluate the evidence at hand. This is my objective in the following three chapters. Because of its historical prominence, I consider competition first and then examine various other processes that may affect community patterns.

Many of the views of communities that developed during the 1960s and early 1970s portrayed a dream world of stability and homegeneity, in which spatial or temporal variations in environments were either nonexistent or were closely matched by responses of the community. To a large degree, these views were nutured by the mathematical constructs of community theory, which operated under the assumption that models reflecting a world in stable equilibrium and having no spatial dimension were reasonable representations of natural communities. Because natural communities were thought to tend strongly toward equilibrium, the effects of history disappeared, environmental perturbations had no lasting effects, spatial heterogeneity was of little consequence, and chance was limited to a small role in affecting the dispersal of species to a locality (Chesson and Case 1986).

But of course environments and communities do vary in time and space. They do so with varying amplitudes, periodicities, and degrees of stochasticity, on a wide range of scales. Communities vary in the presence and absence of species, in the density levels of these species, and in their densities relative to one another. Because the stability and predictability of other aspects of ecological relationships among the species hinge on the relative constancy of these patterns, such variability hinders attempts to detect and explain community patterns. This temporal and spatial variation of communities, although perhaps not entirely ‘music to the ecologist’ (Simberloff 1982a), at least can no longer be considered simply as uninteresting and unwanted ‘noise’, to be muffled by statistical analysis or simply disregarded altogether.

In the previous chapter, I described the difficulties of demonstrating interspecific competition and linking it with community patterns. Birds that feed chiefly upon floral nectar, however, are especially well-suited to investigations of competitive interactions, and it is therefore appropriate to consider their ecology in some detail.

In most parts of the world, a relatively small number of taxonomically related species comprise the chief members of the nectar-feeding guild: hummingbirds (Trochilidae) in the New World, honeyeaters (Meliphagidae) in Australasia, sunbirds (Nectariniidae) in Africa and parts of Asia, and honeycreepers (Drepanididae) in Hawaii. The species are often nectar specialists, and the resource of primary importance to them is thus clearly defined. Moreover, the abundance and availability of the nectar resource can be quantified with precision and related directly to the energetics of the birds. Many plants are adapted to pollination by nectarivorous birds (Stiles 1978, Feinsinger 1978, Feinsinger et al. 1982, Kodric-Brown and Brown 1979, Brown and Kodric-Brown 1979, Grant and Grant 1968), and in some cases the linkage between birds and plants suggests a tightly coevolved system. Because the flowers are adapted to make their nectar available to certain bird pollinators (Murray et al. 1987), measurements of nectar standing crop may accurately reflect resource availability as it is viewed by the foragers (Carpenter 1978). For these reasons, nectar-feeding birds and their resources have been popular subjects for investigations of time-energy budgeting (e.g. Wolf et al. 1975, Gill and Wolf 1977, Hixon et al. 1983), optimal foraging behavior (e.g. Pyke 1978), ecophysiology (e.g. Calder and Booser 1973), and species interactions.

This book and its companion volume represent a personal statement about the ecology of bird communities – what they are, what we know about them, and what we need to know. They are also about how avian community ecology has been practiced as a science – how we have gone about gaining our knowledge of bird comunities and how logical and methodological considerations affect the certainty we can attach to that knowledge. Because studies of birds have contributed a good deal to the foundation of contemporary community ecology and because concerns about logic, methodology, and epistemology are central to any science, I believe that the themes and viewpoints I develop are relevant to community ecology well beyond the somewhat artificial boundaries dictated by my focus on birds. My topic is really community ecology as it has been practiced on birds rather than bird communities per se.

Avian community ecology is a complex, multifaceted discipline that is enriched by controversy. I have written these volumes partly in an attempt to examine the complexity of communities and to probe the dimensions of the controversies, but partly also out of a simple enjoyment of the subject. I have directed my comments particularly toward advanced undergraduates and graduate students with interests in avian community ecology or, more broadly, in birds or in ecology, for I feel that they are in the best position to put my comments into practice or to challenge my views. I hope that my colleagues – practicing ecologists – will also find much to interest (or outrage!) them.

During the past three decades, community ecology in general and avian community ecology in particular have undergone dramatic changes. The qualitative descriptions of community composition that characterized the 1950s gave way to the heady euphoria of the 1960s and early 1970s, when increasingly quantitative descriptions were linked with attractive conceptual or mathematical models of community structure and the role of competition in producing that structure. But then, during the mid-1970s, mutterings of dissatisfaction were heard here and there, and these increased to a sometimes chaotic clamor during the early 1980s. Investigations lost the unitary focus that competition-based community theory had provided and doubt in the validity of past studies or the wisdom of future studies of communities became widespread. Some investigators echoed the skepticism expressed in 1954 by Andrewartha and Birch, who had concluded that community studies were unlikely to contribute any understanding to the central questions of ecology.

Of necessity, much of my emphasis in this book and its companion volume has been on the problems of past studies and the inadequacies that have fueled this recent skepticism. In order to make progress in the difficult task of understanding how assemblages of organisms are put together and what processes act upon them, it has been necessary to examine where we have been and what we know versus what we only think we know. It would be easy to conclude that doing community ecology properly is an impossibly difficult task, that dealing with the logical problems and methodological pitfalls discussed in Volume 1 or the complicating effects of temporal and spatial variation detailed in Chapter 4 and 5 is an unattainable ideal.

All living organisms interact with their environments. They are influenced by a host of environmental factors and, to some extent, they modify their environment. The human species is no exception. We respond to patterns of temperature, rainfall, abundance of food and other resources and to the incidence of competitors, predators and disease. However, the relationship of mankind with the environment is unique. The extent and degree to which quite small populations can modify the environment is unparalleled, and this combines with the success of the species to make an overwhelming impact.

Although the environment has some intrinsic resilience to deleterious change, the rate of growth of the human population is such that its environmental effects are far outstripping the recovery potential of the earth. As the population continues to grow at an ever-increasing rate, so the capacity of the earth to support life is being eroded: reports of famine, deforestation, loss of plant and animal species, soil erosion and atmospheric pollution are becoming all too familiar.

However, for the first time in history we are in a position to begin to evaluate our many and complex relationships with our environment and to attempt to tackle the deleterious consequences of our activities. Major new initiatives among all disciplines concerned with the environment are beginning to get under way, with the aim of understanding the workings of the biosphere, and particularly man's effects upon it.

Many of the problems associated with overexploitation of the environment are multinational in origin and are consequently very difficult to resolve.

Introduction

Forests are arguably the most important vegetation zone on the face of the earth today. They play a far greater role in the well-being of the planetary ecosystem than they are often given credit for, and we may soon find that forests will effectively be called upon to make a still more critical contribution to planetary stability. Yet on every side, from the equator to the arctic, forests are being depleted or will shortly be depleted through human agency at a rate that could well reduce many of them to impoverished remnants by the end of the next century. Indeed, forests, which have been the predominant form of vegetation on our planet for hundreds of millions of years, may soon become a minority presence – whereupon we shall discover (by default) the full measure of their part in underpinning the ecological welfare of our biosphere.

What is the nature and scope of the role of forests in planetary workings? By ‘forests’ I mean tree-dominated communities with substantial canopies, Fig. 2.1 (by contrast with woodlands which feature much sparser tree cover) that are the repository of a greater abundance and diversity of terrestrial life forms than the rest of the earth put together. Tropical forests are specially rich in species and in the evolutionary capacity to generate new species. As tropical forests are cleared wholesale, there will be an impoverishing impact on the very course of evolution itself.

Moreover, forests help to regulate the hydrodynamics of great watersheds and river basins such as those of the Ganges and the Amazon.

The Darwin College Lecture Series was inaugurated in 1986 to provide a range of public lectures on topics of general interest. The first series was entitled Origins and it considered subjects such as the origin of the universe, the origin of man, and the origin of language. The present volume contains the second series which was delivered in Cambridge in 1987, under the original title ‘Man and the Environment’. We are grateful to Richard Grove, Christopher Viney and Jamie Whyte for their help in organising this second series.

The contributors to each series have been selected from a wide range of disciplines to allow a broad look at the topic. Although the contributors are acknowledged specialists in their fields, the lectures are aimed at nonspecialist readers.