The Importance of Species: Perspectives on Expendability and Triage

By Peter Kareiva and Simon A. Levin

A great many species are threatened by the expanding human population. Though the public generally favors environmental protection, conservation does not come without sacrifice and cost. Many decision makers wonder if every species is worth the trouble. Of what consequence would the extinction of, say, spotted owls or snail darters be? Are some species expendable?


Given the reality of limited money for conservation efforts, there is a compelling need for scientists to help conservation practitioners set priorities and identify species most in need of urgent attention. Ecology should be capable of providing guidance that goes beyond the obvious impulse to protect economically valuable species (salmon) or aesthetically appealing ones (snow leopards). Although some recent books have considered the ecosystem services provided by biodiversity as an aggregate property, this is the first to focus on the value of particular species. It provides the scientific approaches and analyses available for asking what we can expect from losing (or gaining) species.


The contributors are outstanding ecologists, theoreticians, and evolutionary biologists who gathered for a symposium honoring Robert T. Paine, the community ecologist who experimentally demonstrated that a single predator species can act as a keystone species whose removal dramatically alters entire ecosystem communities. They build on Paine's work here by exploring whether we can identify species that play key roles in ecosystems before they are lost forever. These are some of our finest ecologists asking some of our hardest questions.


They are, in addition to the editors, S.E.B. Abella, G. C. Chang, D. Doak, A. L. Downing, W. T. Edmondson, A. S. Flecker, M. J. Ford, C.D.G. Harley, E. G. Leigh Jr., S. Lubetkin, S. M. Louda, M. Marvier, P. McElhany, B. A. Menge, W. F. Morris, S. Naeem, S. R. Palumbi, A. G. Power, T. A. Rand, R. B. Root, M. Ruckelshaus, J. Ruesink, D. E. Schindler, T. W. Schoener, D. Simberloff, D. A. Spiller, M. J. Wonham, and J. T. Wootton.

• Language: English
• Publisher: Princeton University Press
• Release date: Jan 22, 2015
• ISBN: 9781400866779

Book preview

THE IMPORTANCE OF SPECIES

THE IMPORTANCE OF SPECIES

PERSPECTIVES ON EXPENDABILITY AND TRIAGE

EDITED BY

Peter Kareiva and Simon A. Levin

PRINCETON UNIVERSITY PRESS     Princeton and Oxford

Copyright © 2003 by Princeton University Press

Published by Princeton University Press, 41 William Street, Princeton, New Jersey 08540

In the United Kingdom: Princeton University Press, 3 Market Place, Woodstock, Oxfordshire OX20 1SY

All Rights Reserved

Library of Congress Cataloging-in-Publication Data

The importance of species : perspectives on expendability and triage / edited by Peter Kareiva and Simon A. Levin.

   p. cm.

Papers presented at a symposium held in honor of Robert Treat Paine, upon the occasion of his retirement from the University of Washington.

Includes bibliographical references.

ISBN 0-691-09004-1 (alk. paper) — ISBN 0-691-09005-X (pbk. : alk. paper)

1. Conservation biology—Congresses. 2. Species diversity—Congresses. 3. Endangered species—Congresses. 4. Biological diversity conservation—Congresses. I. Kareiva, Peter M., 1951–II. Levin, Simon A.

QH75 .I4 2003

333.95′16—dc21    2002025137

British Library Cataloging-in-Publication Data is available

This book has been composed in Palatino

Printed on acid-free paper. ∞

www.pupress.princeton.edu

Printed in the United States of America

10  9  8  7  6  5  4  3  2  1

Contents

Contributors

 ix

Preface

 xiii

Foreword

 xv

PART I

USING EXPERIMENTAL REMOVALS OF SPECIES TO REVEAL THE CONSEQUENCES OF BIODIVERSITY DEPLETION

P. Kareiva and S. A. Levin

1

1.  Native Thistles: Expendable or Integral to Ecosystem Resistance to Invasion?

S. M. Louda and T. A. Rand

5

2.  The Overriding Importance of Environmental Context in Determining the Outcome of Species-Deletion Experiments

B. A. Menge

16

3.  Species Importance and Context: Spatial and Temporal Variation in Species Interactions

C.D.G. Harley

44

4.  Effects of Removing a Vertebrate versus an Invertebrate Predator on a Food Web, and What Is Their Relative Importance?

T. W. Schoener and D. A. Spiller

69

5.  Understanding the Effects of Reduced Biodiversity: A Comparison of Two Approaches

J. T. Wootton and A. L. Downing

85

PART II

THE ANTHROPOGENIC PERSPECTIVE

P. Kareiva and S. A. Levin

105

6.  Models of Ecosystem Reliability and Their Implications for the Question of Expendability

S. Naeem

109

7.  Predicting the Effects of Species Loss on Community Stability

D. Doak and M. Marvier

140

8.  One Fish, Two Fish, Old Fish, New Fish: Which Invasions Matter?

J. L. Ruesink

161

9.  Ecological Gambling: Expendable Extinctions Versus Acceptable Invasions

M. J. Wonham

179

10.  Rarity and Functional Importance in a Phytoplankton Community

D. E. Schindler, G. C. Chang, S. Lubetkin, S.E.B. Abella, and W. T. Edmondson

206

11.  Community and Ecosystem Impacts of Single-Species Extinctions

D. Simberloff

221

PART III

LINKAGES AND EXTERNALITIES

P. Kareiva and S. A. Levin

235

12.  Social Conflict, Biological Ignorance, and Trying to Agree Which Species Are Expendable

E. G. Leigh Jr.

239

13.  Which Mutualists Are Most Essential? Buffering of Plant Reproduction against the Extinction of Pollinators

W. F. Morris

260

14.  The Expendability of Species: A Test Case Based on the Caterpillars on Goldenrods

R. B. Root

281

15.  An Evolutionary Perspective on the Importance of Species: Why Ecologists Care about Evolution

S. R. Palumbi

292

16.  Recovering Species of Conservation Concern—Are Populations Expendable?

M. Ruckelshaus, P. McElhany, and M. J. Ford

305

17.  Virus Specificity in Disease Systems: Are Species Redundant?

A. G. Power and A. S. Flecker

330

Conclusion

P. Kareiva and S. A. Levin

347

References

353

Index

415

Contributors

Sally E. B. Abella, Department of Zoology, University of Washington, Seattle, WA 98195-1800. Current Address: King County Water and Land Resources, 201 S. Jackson St., Suite 600, Seattle, WA 98104

Gary C. Chang, Department of Zoology, University of Washington, Seattle, WA 98195-1800. Current Address: Department of Plant, Soil, and Entomological Sciences, University of Idaho, Moscow, ID 83844

Dan Doak, Department of Biology, University of California, Santa Cruz, CA 95064

Amy L. Downing, Department of Ecology and Evolution, University of Chicago, 1101 E. 57th Street, Chicago, IL 60637-1573. Current Address: Department of Zoology, Ohio Wesleyan University, Delaware, OH 43015

W. T. Edmondson (deceased), Department of Zoology, University of Washington, Seattle, WA 98195-1800

Alexander S. Flecker, Department of Ecology and Evolutionary Biology, Cornell University, Corson Hall, Ithaca, NY 14853

Michael J. Ford, National Marine Fisheries Service, Northwest Fisheries Science Center, 2725 Montlake Boulevard East, Seattle, WA 98112-2097

Christopher D. G. Harley, Department of Zoology, University of Washington, Seattle, WA 98195-1800. Current Address: Hopkins Marine Station, Oceanview Boulevard, Pacific Grove, CA 93950

Egbert Giles Leigh, Jr., Smithsonian Tropical Research Institute, Apartado 2072, Balboa, Panama

Svaťa M. Louda, School of Biological Sciences, University of Nebraska, Lincoln, NE 68588-0118

Susan Lubetkin, Quantitative Ecology and Resource Management, University of Washington, 407 Bagley Hall, Seattle, WA 98195

Michelle Marvier, Department of Biology, Environmental Studies Institute, Santa Clara University, Santa Clara, CA 95053

Paul McElhany, National Marine Fisheries Service, Northwest Fisheries Science Center, 2725 Montlake Boulevard East, Seattle, WA 98112-2097

Bruce A. Menge, Department of Zoology, Oregon State University, Cordley Hall 3029, Corvallis, OR 97331-2914

William F. Morris, Department of Biology, Duke University, Durham, NC 27708-0338

Shahid Naeem, Department of Zoology, University of Washington, 24 Kincaid Hall, Seattle, WA 98195-1800

Stephen R. Palumbi, Department of Biological Sciences, Hopkins Marine Station, Stanford University, Oceanview Blvd., Pacific Grove, CA 93940

Alison G. Power, Department of Ecology and Evolutionary Biology, Cornell University, Corson Hall, Ithaca, NY 14853

Tatyana A. Rand, School of Biological Sciences, University of Nebraska, Lincoln, NE 68588-0118

Richard B. Root, Department of Ecology and Evolutionary Biology, Cornell University, Corson Hall, Ithaca, NY 14853

Mary Ruckelshaus, National Marine Fisheries Service, Northwest Fisheries Science Center, 2725 Montlake Boulevard East, Seattle, WA 98112-2097

Jennifer L. Ruesink, Department of Zoology, University of Washington, Seattle, WA 98195-1800

Daniel E. Schindler, Department of Zoology, University of Washington, Seattle, WA 98195-1800

Thomas W. Schoener, Section of Evolution and Ecology, University of California, 1 Shields Avenue, Davis, CA 95616-8755

Daniel Simberloff, Department of Ecology and Evolutionary Biology, University of Tennessee, Knoxville, TN 37996

David A. Spiller, Section of Evolution and Ecology, University of California, 1 Shields Avenue, Davis, CA 95616-8755

Marjorie J. Wonham, University of Washington, Department of Zoology, Box 351800, Seattle, WA 98915-1800. Current Address: University of Alberta, Centre for Mathematical Biology, CAB 632, Edmonton, AB, Canada T6G 2G1

J. Timothy Wootton, Department of Ecology and Evolution, University of Chicago, 1101 East 57th Street, Chicago, IL 60637-1573

Preface

A born naturalist, Robert Treat Paine has firmly established himself as one of the leading ecologists of this—or any—century. His ability to carry out unparalleled and innovative field experiments, informed by solid natural history and deep insights about the structure and functioning of ecological communities, has made his work among the most influential in the development of ecology over the past thirty years. He selected intertidal rocky shores as the ideal system for testing seminal questions about community dynamics, and his long-term commitment to elucidating ecological relationships in the intertidal has provided one of the benchmark ecological systems. In particular, he has illustrated the power and necessity of experimental approaches to community organization and has combined this approach with powerful theory to elucidate food web complexity, trophic dynamics, patch dynamics, and community energetics.

R. T. Paine has been not only a great scientist but a model mentor. His students have been his primary concern, and he has inspired them (and others) through example and guided them to brilliant careers. Community ecology in general, and intertidal ecology in particular, have been deeply influenced by him. Robert Paine has been often and justly honored—as one of the few ecologists in the National Academy of Sciences and the American Academy of Arts and Sciences; by selection as the third Tansley Lecturer of the British Ecological Society; and through his receipt of the MacArthur Award of the Ecological Society of America, the Wright Award of the American Society of Naturalists, and the Ecology Institute Prize. His influence, always great, continues to grow.

Foreword

Not everyone is enthusiastic about protecting biodiversity. The practice of conserving species and biodiversity requires major societal sacrifices, and in many cases cannot be accomplished without substantial economic costs. Given these costs, it is not surprising that members of the public often ask conservationists whether every species must be protected. Of course, many of the hard decisions surrounding conservation have nothing to do with science—they represent a choice among values and are of a political, ethical, and philosophical nature. Still, there is a need for scientists to respond to questions about the consequences of losing particular species or segments of biodiversity. In this book, leading ecologists, evolutionary biologists, and conservation biologists explore questions about the value (or conversely the expendability) of species. The book provides an in-depth scientific exploration of what kinds of evidence and research can inform society about the choices it must make regarding species extinctions.

Contributions were initially presented as talks at a symposium held to honor Robert Treat Paine, on the occasion of his retirement as a Professor of Zoology at the University of Washington. Much of Bob’s research focus, and the focus of those inspired by him, has been on species interactions. Consequently, whereas several other recent books address the value of biodiversity from the perspective of ecosystem services (Daily 1997, Baskin 1997), this monograph emphasizes how the value of species is revealed through the particulars and generalities of species interactions. Although biodiversity is the rallying call of many conservation groups, the protection of biodiversity often comes down to individual species and their interactions. When the public at large questions conservation efforts, the challenge commonly involves queries about why we should care about a particular species such as the spotted owl or the snail darter. Although ecology and evolutionary biology are not in the business of offering reasons for caring, science can make clear what is likely to happen if different species are eliminated. Indeed, one of the dominant activities of ecologists, following the lead of Bob Paine, has been research that involves the experimental removal of species to assess the response of the surrounding community to the loss of particular organisms.

Challenging biologists to assess the expendability of species pushes them to the limits of their science. Discussions about species loss in the popular press often resort to metaphors. According to the rivet metaphor, for example, losing species is like popping rivets out of an airplane: at first losing a few rivets appears inconsequential, but eventually, if too many rivets are popped out, the plane is not one on which anyone would like to be a passenger (Ehrlich and Ehrlich 1981). Such metaphors provide powerful imagery and have played an important role in sensitizing people to the problem of species extinction. This monograph takes the next step, moving beyond metaphors to struggle substantively with the difficult question of what it can mean to lose a species from a community, however unimportant that species might appear on casual (or even careful) observation. Bob Paine has used a mix of experiments, theory, and natural history to illuminate ecology, and the contributors to this book use a similar blend to examine the importance of species.

THE IMPORTANCE OF SPECIES

PART I

USING EXPERIMENTAL REMOVALS OF SPECIES TO REVEAL THE CONSEQUENCES OF BIODIVERSITY DEPLETION

For decades, Bob Paine has exhorted ecologists to conduct field experiments in which species are removed and the community-wide responses to those removals noted. The lessons learned from these manipulations of natural communities have been impressive and have reshaped our understanding of ecological systems. Given the central role of experiments in community ecology, it is surprising that the large body of results from species-removal experiments has been so neglected in the debate surrounding the importance of biodiversity. After all, there is no more straightforward way of examining the importance of a species than to remove it. Thus, experimental investigations of species interactions should speak directly to the question of species expendability. In the following five contributions, we find compelling evidence that the loss of even a single species can have severe ramifications for ecosystem structure and functioning. But we also find clear limitations of the experimental approach as an indicator of species importance.

Louda and Rand begin by describing the interactions between native thistles and the insects that feed on those thistles. They make a persuasive case that insects can play a major role in limiting the reproductive success of thistles. Pointing out that the insects feeding on native thistles also feed on noxious invasive thistles, Louda and Rand argue that native thistles, by harboring populations of herbivores, provide resistance to invasion by exotic thistles. The argument is logical and is based on careful natural history and a detailed understanding of insect-plant interactions. Missing, however, is the definitive experiment that uses species removal to test whether weedy thistles have an easier time invading areas where native thistles have been removed.

Louda and Rand provide an excellent example of why we have come to value species-removal experiments as a research tool; without such experiments, even the best quantitative and natural history studies leave us with uncertainty about the role a species plays. Although we suspect that Louda and Rand are correct in their assessment regarding the importance of native thistles to community resistance, we cannot be convinced entirely until the critical removal experiment is performed.

The four remaining contributions in this section report the results of actual removal experiments. The chapters by Menge and Harley both deliver a message of context dependency: that a single species can appear inconsequential in one experiment but of central importance in a different removal experiment. Harley makes this point by modifying a measure of interaction strength advocated by Paine (1992) and calculating this new metric of interaction strength for more than 150 experimentally perturbed pairs of species. It is a cliché to say that organisms, populations, and communities vary according to time and place. Harley goes beyond this cliché to quantify what this variation is likely to mean for a species’ importance as determined by measurements of interaction strength in one community or context. Specifically, Harley finds that the measure of interaction strength for a species pair in one context explains, on average, only 37% of the variation in the interaction strength observed in a different year or place. Menge remarks on the same phenomenon but goes on to suggest that the variation in interaction strengths observed in marine systems may well be predicted on the basis of large-scale variation in the physical environment. Menge predicts, for example, that the removal of predators will be especially important in highly productive upwelling systems, which tend to be characterized by strong top-down interactions.

Schoener and Spiller synthesize the results of several years of removal experiments involving a dominant lizard species and a dominant spider species in the Bahama Islands. The high point of their analysis is its anticipation that based on trophic position, removing lizards should have a much greater impact than the removal of spiders. Secondly, whereas much ado is made about ecosystem function when discussing the value of biodiversity, Schoener and Spiller add considerable meat to the concept. In particular, they show that the impact of lizard removal on increased herbivore density carries through to primary production, noting that leaf damage increases by threefold in areas where lizard densities have been reduced.

Finally, Wootton and Downing argue that the popular approach to assessing typical responses to reductions in biodiversity misses the clearest lesson of the last twenty years of species removal experiments: that the effect of a species-removal is highly idiosyncratic. As an alternative, they suggest a hybrid approach in which targeted species removals are combined with general diversity manipulations. Targeted removals are championed as invaluable, because they provide an experimental manipulation that can lead to predictions about the consequences of particular extinctions.

The chapters by Wootton and Downing and by Schoener and Spiller illuminate a path for future research that is likely to be especially valuable for those interested in the conservation of biodiversity. These authors redefine popular questions about biodiversity in a way that is amenable to ecological theory. Schoener and Spiller point out that expendability is largely a question for values but that conservation is often motivated by an urge to keep things the same. By understanding community dynamics, we can predict the role of species in keeping things the same—exactly as Schoener and Spiller were able to understand why lizard removal altered ecosystem function but spider removal did not. Wootton and Downing argue that the average effect of biodiversity is not as interesting as the effect of particular losses of biodiversity in particular settings. Thus, instead of asking whether every species is irreplaceable or whether every reduction of biodiversity represents environmental degradation, they seek a predictive theory that allows one to understand why the losses of some species have large impacts while the losses of others have negligible ecological impacts. Whereas many ecologists lose their way when discussing expendability, Schoener, Spiller, Wootton, and Downing derive clarity from their focus on species interactions and the consequences of targeted species removals.

At its most basic level, research involving species removal is significant simply because it allows ecologists to accumulate data regarding species interactions. Now that we have accumulated a large suite of results from targeted removal experiments, contributors to this book are able to synthesize the consequences of removal experiments in a manner that sheds light on the likely consequences of species extinctions.

CHAPTER 1

Native Thistles: Expendable or Integral to Ecosystem Resistance to Invasion?

Svaťa M. Louda and Tatyana A. Rand

One way of addressing the question of whether some species are expendable is to ask what role, if any, a minor species, even one that seems obnoxious, plays in the functioning of its community. Thistles (Cirsium spp.) are prickly plants native to North America that are numerically minor and are often considered unattractive or undesirable. So, thistles might be considered expendable. Yet, can we assume that such minor, seemingly undesirable species can be eliminated without disrupting important interdependencies or losing key ecological services? Our long-term studies of thistle-insect interactions are beginning to provide evidence that even such species may play important, unexpected roles in ecological dynamics and in economic welfare. These studies suggest that determining the cost of losing a species requires criteria other than relative abundance and general attractiveness (see Root this volume).

In this chapter, we summarize the natural history of the native Cirsium species, which we study in the upper Great Plains. Then, we briefly review experimental evidence that the native insects that feed on native thistles can restrict their abundance and weediness. Finally, we present new observational data suggesting that these native insects, moving over from a native species such as tall thistle (Cirsium altissimum (L.) Spreng.), are likely involved in limiting the invasion of bull thistle, Cirsium vulgare (Savi) Ten., in the tallgrass prairie region of Nebraska. Bull thistle is a Eurasian species that is highly invasive elsewhere (Austin et al. 1985; Randall 1991; Julien and Griffiths 1998; Olckers and Hill 1999) and often becomes expensive to control in agronomic regions around the world. Such invasions of nonindigenous species can present major economic and ecological threats to ecosystem structure and functioning (Mooney and Drake 1986; Drake et al. 1989; Simberloff et al. 1997). We hypothesize that our study represents a case of a numerically minor species that, acting as a reservoir of native insects, provides an ecologically and economically valuable ecosystem service: resistance to invasion by an alien weed. The case also suggests that more research on the potential biotic resistance provided by natural enemies may help us better understand those factors that influence whether an alien species becomes invasive after naturalization.

Based on our studies, we argue that there are practical as well as aesthetic and ethical reasons for working to maintain minor, even seemingly obnoxious, species and their interactions. In particular, this case suggests that we are not yet in a position to predict the cost associated with the decline and loss of a specific species, since its ecological function and economic value may not be obvious.

Natural History Background

The thistle genus Cirsium (L.), indigenous in Eurasia, North America, and North and East Africa, contains about 250 species (Bremer 1994). The North American contingent of this genus is represented by at least 96 indigenous, endemic taxa (USDA, NRCS 1999). We have extensive quantitative data on four native species that are characteristic of the prairie grasslands of the upper Great Plains. All typically occur singly or in small stands, and none are considered major weeds (McCarty et al. 1967; Louda et al. 1990; Stubbendieck et al. 1994). The data presented here are for the native tall thistle (Cirsium altissimum (L.) Spreng.), a late-flowering monocarpic species in the tallgrass region of Nebraska (McCarty et al. 1967), and the naturalized Eurasian thistle species bull (spear) thistle (Cirsium vulgare (Savi) Tenore), also a late-flowering monocarpic species that occurs as small stands in disturbed roadside or overgrazed grassland in Nebraska. Although several alien thistles are listed as noxious weeds in Nebraska, bull thistle is not. Our studies of tall and bull thistles were conducted in Lancaster County, east of the City of Lincoln.

As fugitive species, the performance and density of thistles are related to the availability of seed, the level and spacing of disturbance, and the vigor of grass competition (Hamrick 1983; Hamrick and Lee 1987; Louda et al. 1990, 1992; Popay and Medd 1990; Louda and McEachern 1995; Louda and Potvin 1995; Bevill and Louda 1999). Seed availability usually limits local thistle seedling density in open grasslands (see de Jong and Klinkhamer 1986; Louda and McEachern 1995; Louda and Potvin 1995).

Thistles have a suite of adapted insects (Zwölfer 1965, 1988; Lamp and McCarty 1979, 1981, 1982a, b; Redfern 1983; Zwölfer and Romstöck-Völkl 1991). The most common insects specializing on Cirsium species in Nebraska are picture-winged flies (Tephritidae), weevils (Curculionidae), moths (Pyralidae, Pterophoridae), butterflies (Nymphalidae, Hesperiidae), lacebugs (Tingidae), aphids (Aphididae), and sucking bugs (Hemiptera: Cicadellidae, Membracidae, Miridae, Pentatomidae). Several studies demonstrate that these native insects significantly affect key components of individual plant fitness (Louda et al. 1990, 1992; Louda and Potvin 1995; Guretzky and Louda 1997; Stanforth et al. 1997; Bevill 1998; Jackson 1998; Bevill et al. 1999).

There is also evidence that thistle-feeding insects often adopt similar hosts in alternate or novel environments. For example, several thistle-feeding insects have been imported from Eurasia and released in the United States by the U.S. Department of Agriculture as biological control agents for exotic thistles (Julien and Griffiths 1998). At least two of the weevils being distributed for the control of alien thistles have also adopted native species as hosts; these include Rhinocyllus conicus Fröl. (Louda et al. 1997; Louda and Arnett 2000) and Larinus planus (F.) (Louda and O’Brien 2002).

In sum, since insects can reduce plant performance of thistles and since host range expansion in thistle-feeding insects occurs, the potential clearly exists for native insects to adopt potentially invasive thistles as hosts. If this is the case, these insects likely play a role in limiting the reproduction and spread of nonindigenous thistles that are closely related, or ecologically similar, to native species.

Native Insect Herbivores Limit Densities of Indigenous Thistles

Multiple studies of the role of coevolved insects in the population dynamics of native thistles in prairie grasslands have demonstrated clearly that native insects significantly decrease the survival, growth, reproduction, lifetime fitness, and density of native thistles under field conditions. These experiments document significant reductions by insect herbivores of (1) juvenile survivorship and growth (Guretzky and Louda 1997; Stanforth et al. 1997; Bevill et al. 1999), especially in the context of competition with grasses (Louda et al. 1990, 1992); (2) subsequent flowering effort of surviving plants (Bevill 1998; Bevill et al. 1999); (3) successful seed maturation (Louda et al. 1990, 1992; Louda and McEachern 1995; Louda and Potvin 1995; Jackson 1998; Louda 1999a, b; Maron et al. unpub. data); (4) lifetime fitness (Louda and Potvin 1995); and (5) seedling, juvenile, and adult densities (Louda and Potvin 1995). Populations of the monocarpic thistles have been shown to be seed-limited in undamaged prairie grassland (Louda and Potvin 1995). Platte thistle density, for example, increased 100–600% when floral insect herbivores were reduced, especially in disturbances but also in ungrazed prairie (Louda et al. 1990, 1992; Louda and Potvin 1995). Individually, these studies show that coevolved thistle-feeding insects significantly reduce key parameters of individual plant performance. Collectively, the studies suggest that the chronic pressure exerted by a diverse, dependent assemblage of adapted natural enemies often limits population density and patch regeneration of native thistles in grasslands under indigenous conditions.

Native Thistle Harbors Insects that Attack an Exotic Thistle

Ecosystem resistance to invasion has been listed as an important property of intact ecosystems (e.g., Daily 1997). One mechanism that produces such resistance to exotic plant invasion is competition from native plants (see Drake et al. 1989; McKnight 1993; Mack 1996). Herbivory by native insects could be another mechanism that reduces the invasiveness of some nonindigenous plant species (Louda 1999b). In this case, insects from a native thistle contribute significant resistance to the potential for invasive population growth and spread by bull (spear) thistle, Cirsium vulgare (Savi) Ten., in eastern Nebraska.

Bull thistle, a Eurasian species, is an aggressive weed in many areas of the world. Its invasive potential is clear from published descriptions, studies, and control efforts in grazed grasslands of Australia, New Zealand, and South Africa (e.g., Austin et al. 1985; Julien and Griffiths 1998; Olckers and Hill 1999) and in natural areas of California (Randall 1991). However, although bull thistle has been present in the tallgrass prairie region of eastern Nebraska for at least 35 years, its numbers remain relatively low (C. P. Andersen and S. M. Louda, unpub. data). Contemporary agricultural practices control weeds, including bull thistle, within crop fields, but numbers have also remained low in roadsides and perennial pastures despite disturbance and grazing. Bull thistle is not common enough to be classified as a major weed throughout eastern Nebraska.

Native insects from the indigenous tall thistle have included the potentially invasive Eurasian thistle in their diet. When native insect herbivores fed on developing buds and flowering heads, the seed production of bull thistle in Nebraska was severely reduced. In Lancaster County, for example, insects destroyed 88% of the potential seed production by bull thistle plants sampled in 1997, 79% in 1998, and 71% in 1999, significantly reducing the reproductive success of bull thistle (fig. 1.1). At least four insects were often found feeding on or in the reproductive shoots and developing flower heads of bull thistle. These insects were Platyptilia carduidactyla (Riley) (Pterophoridae), Baris subsimilis Casey (Curculionidae), Papaipema mitela Guen. (Noctuidae), and Paracantha culta Wiedeman (Tephritidae). All these species typically feed on or in the reproductive shoots and developing flower heads of the native, tall thistle (Louda, pers. obs.). Feeding by these insects severely reduced the number of flower heads that matured and the number of viable seeds that were produced by the native tall thistle from 1994 to 1995 (Jackson 1998) and from 1997 to 1999 (fig. 1.2), significantly reducing reproductive success. The levels of use of bull thistle were also high, though not as high as those observed on the native tall thistle (see fig. 1.1 vs. fig. 1.2; Jackson 1998; Louda, unpub. data). Thus, the adapted, dependent insects of native thistles are exerting tremendous pest resistance pressure on this exotic, potentially invasive weed under its newly adopted conditions in eastern Nebraska.

Our previous experiments, in addition, have shown that herbivory by native insects on tall thistle further limits the survival and growth of young plants (Guretzky and Louda 1997) and the flowering success of older individuals (Jackson 1998). Subsequent competition with grasses for nutrients and moisture, known to restrict the success of established thistles (Austin et al. 1985; Hamrick 1983; Hamrick and Lee 1987; Popay and Medd 1990), likely reduces plant density further by limiting the survival and growth of the seedlings that do establish. If the consequences of insect feeding on bull thistle are similar to those for tall thistle—and this is currently being tested (L. M. Young, unpub. data)—then population growth and the development of high densities of bull thistle are opposed by the pressure exerted on this exotic thistle by native insects, insects—that are adapted to and maintained by a thistle native to this region. Disruption of these interactions, through a loss of the native thistle and its reservoir of native insects, would be expected to increase the probability of a full-blown invasion by bull thistle, as has occurred elsewhere. Thus, loss of the native thistle and its insects could create a more noxious, economic weed out of a currently innocuous exotic plant.

Figure 1.1 Average number (X, SE) of potential seeds as florets initiated; florets and seeds destroyed by feeding of native insects; and viable seeds released per plant by the Eurasian bull (spear) thistle, Cirsium vulgare (Savi) Ten. in Lancaster County, Nebraska (N = 10, 5, and 10, in 1997, 1998, and 1999, respectively). Native insects, transferring from the co-occurring, late-flowering native thistle (Cirsium altissimum, tall thistle), reduced seed production by this exotic, potentially invasive thistle by 88%, 79%, and 71% in 1997, 1998, and 1999, respectively. * = p < 0.05 in orthongonal contrasts.

Discussion

The reasons for preserving species are scientific, functional, and practical. First, studies of thistle dynamics and interactions have contributed basic ecological insights into how biological interactions can structure, limit, and influence the numerical abundance and distribution of native plants in grasslands (e.g., Louda et al. 1990, 1992; Louda and Potvin 1995; Guretzky and Louda 1997; Jackson 1998; Bevill et al. 1999). Furthermore, parallel studies have added to an understanding of the role of interactions in plant rarity (Louda and McEachern 1995; Stanforth et al. 1997; Bevill and Louda 1999; Bevill et al. 1999).

Figure 1.2 Average number (X, SE) of potential seeds as florets initiated; florets and seeds destroyed by feeding of native insects; and viable seeds released per plant by the native thistle, Cirsium altissimum (L.) Spreng. (tall thistle) in Lancaster County, Nebraska (N = 5 and 10, in 1998 and 1999, respectively). The coevolved, inflorescence-feeding insects reduced seed production by this native thistle species 99.3% and 94.8% in 1998 and 1999, respectively. * = p < 0.05 in orthongonal contrasts.

Second, thistles contribute to the support of a broad array of animal species. In the Great Plains prairies, for example, at least 35 other species use native North American thistles (Louda, unpub. data). These species range from microscopic plant parasites (that harbor potentially useful secondary compounds) to macroscopic animals, including charismatic ones (e.g., as the American Goldfinch) and their predators (e.g., raptors) (Louda et al. 1998). Would such species decline if the native thistle were eliminated? No good answer exists yet.

Third, our data suggest that a native thistle can support a set of herbivorous insects that contribute to a major ecosystem service: the limitation of a potential weed. Native insects limit the seed production and density of populations of native thistles in native and disturbed grasslands (Louda and Potvin 1995; Guretzky and Louda 1997; Bevill et al. 1999). In addition, these insects are contributing to the suppression of the potentially invasive, exotic bull thistle, Cirsium vulgare.

What important role, then—if any—can minor, seemingly obnoxious species play? In this case, the data suggest that a prickly, inconspicuous native plant, tall thistle, supports insect herbivores that dramatically reduce the seed production of an incipient invasive species, likely constraining the density and the rate of spread of a potentially serious economic and environmental weed. Thus, we hypothesize that elimination of the native tall thistle would likely have at least one negative ecosystem response with economic implications. Its reduction and loss would be expected to cause a reduction of temporally synchronized adapted insects, decreasing resistance to invasion by the exotic species. Bull thistle would likely become a much greater problem—reaching a status similar to that in other rangeland regions—without indigenous thistles to harbor adapted, thistle-feeding insects.

Following Bob Paine’s example, several key experiments are now underway to test the hypothesis of significant ecosystem resistance provided by tall thistle. These experiments will quantify bull thistle response to (1) increased seed input, to address the question of whether population density is proportional to seed availability; (2) the exclusion of native insect herbivores, to address the question of whether potential seed production is limited by resources or factors other than intensive herbivory; and (3) the removal of neighboring native thistles, to address the question of whether proximity of native thistles influences the native insect use of bull thistle directly, as suggested by observational data (C. P. Andersen and S. M. Louda, unpub. data).

We propose that our studies, along with the knowledge that communities are often structured by trophic interactions (Elton 1927; Pimm 1982; Polis and Winemiller 1996), argue that even a naturally sparse, potentially noxious, native species like a thistle cannot necessarily be assumed to be expendable. Interestingly, the potential functional significance of tall thistle and its dependent insect herbivore guild became evident only in the context of an external challenge: the establishment and potential invasion by bull thistle. Thus, this case illustrates the difficulty of identifying and quantifying an ecosystem service until it is needed. Similarly, Root (this volume) has pointed out that the role a species plays may be altered, or may only become apparent, as a result of changing conditions, such as the colonization of an ecosystem by an exotic species. The novel interactions that may arise as a consequence of such changes make it difficult to predict the role that a species may play in the future.

It is also worth noting that even thistles are not interchangeable in their ecosystem functions. In this case, saving an early flowering thistle (e.g., Platte thistle) and its earlier-feeding insects would not be equivalent to saving the late-flowering thistle (tall thistle) whose insects are providing some resistance to invasion by bull thistle. Unlike biotic resistance provided by plant competitors, in which there is likely to be a substantial amount of functional redundancy among plant species, the function provided by specific plants that serve as reservoirs of coevolved natural enemies is likely to be sensitive to the elimination of a specific plant species.

Few studies have quantified the ecosystem resistance to invasion that is provided by indigenous insect herbivores in other systems. In fact, detection could be impossible in many cases, for example if resistance is sufficient to prevent initial colonization and naturalization. Mack (1996) presents floristic data suggesting that naturalization is enhanced for alien species that lack native relatives. He argues that this pattern may be due to the decreased likelihood of host extensions, or shifts, by native natural enemies that have coevolved with native plants onto exotics that are more distantly related to the native flora. Our data are consistent with this suggestion. Clearly, though, the ecosystem service provided by native plants when naturalization is prevented will be exceptionally difficult to detect and document.

Some indirect evidence suggests, however, that feeding by native insects on non-native plant species may be more common and potentially more important than generally thought. For example, host range expansion of native insects onto non-native plant species has been documented for a diverse group of plants, including other exotic thistles such as milk thistle (Silybum marianum Gaertn.) and Italian thistle (Carduus pycnocephalus L.) in southern California (Goeden 1971, 1974), a variety of other nonindigenous herbs (Wheeler 1974; Chew 1977; Berenbaum 1981; Thomas et al. 1987; Evans et al. 1994); and even introduced trees (Fraser and Lawton 1994). Such crossover effects are also common in agricultural systems (e.g., Strong 1974; Strong et al. 1977; Tabashnik 1983) where native host plants serve as reservoirs of crop pests (e.g., Herzog and Funderburk 1986). Furthermore, native insects have been manipulated successfully for biological control of exotic weeds (Sheldon and Creed 1995; Newman et al. 1998).

A prominent theory used to explain the invasiveness of exotic plants within novel habitats is that they have been released from their adapted natural enemies, such as insect herbivores, which are assumed to keep them in check within their native habitats (e.g., Andres and Goeden 1971; Crawley 1987; Blossey and Nötzold 1995; Mack 1996). The indirect evidence showing that native insect herbivores often expand their host range onto introduced plants leads to the hypothesis that such herbivores could also play a role in limiting exotic plant success in the new environment. This result could keep exotics below invasive levels or even prevent naturalization all together, especially when related native plants support an herbivore guild within the habitat.

Thus, the theory, the indirect evidence, and our data suggest that more investigations are merited. To clarify the role of adapted, indigenous herbivorous insects in providing resistance to plant invasion. More generally, an elucidation of important indirect linkages between plants that are mediated by mobile insects, herbivores, or pollinators will be critical to our ability to predict the community-level implications of the extirpation of individual plant species.

Few individuals are known to admire thistles (although the Scots are a notable exception). Most express a disregard for thistles, native or not, without knowing much about them or their ecological interactions and ecosystem functions. The impact on native North American thistles was not a major factor in the decision to release the biocontrol weevil Rhinocyllus conicus in 1969, in spite of evidence that it would use Cirsium species (Boldt 1997; Gassmann and Louda 2001). Nor is the potential impact of Larinus planus on thistle populations, exotic or native, a factor influencing its redistribution within North America. Distribution continues, in spite of evidence that each of these weevils accepts Cirsium species into its diet and that each can have a major nontarget effect on native species (Louda et al. 1997, 1998; Louda 2000; Louda and Arnett 2000; Louda and O’Brien 2002). Perhaps a disregard for noncharismatic species, especially prickly ones, is understandable. However, neither a lack of charisma nor relative rarity provide an adequate scientific basis for deciding whether a species has a significant ecological function or indirect economic value.

Conclusion

Our data suggest that thistles—minor species with a mixed or even unfavorable public image—may provide a vital ecosystem service: resistance to invasion by a putative economic and environmental weed. The conflict between our results and the general attitude toward thistles illustrates the need to develop scientific criteria for determining whether a species might be ecologically redundant and therefore potentially dispensable. Even given the motivation to develop such criteria, however, the question remains: could the interdependencies and numerical consequences of eliminating an indigenous thistle be anticipated and predicted in the context of evaluating its expendability? The task would be difficult. In our study of tall thistle, intensive studies of thistles and their dependent species were required to document the demographic interconnections and to discover the potential economic value of preserving the species. Furthermore, the ecosystem service provided would not have been detected without the challenge from a potentially invasive species. Thus, it seems clear that we are not in a position to define all species-specific ecological or practical roles. Rather than expendability, the critical issue seems to be this: how can we use and manage natural systems in a way that will minimize the probability that component, potentially important species will be lost?

Acknowledgments

We thank the many students, colleagues, and family members who have shared in and contributed to these studies of thistles. Bob Paine and his students, Bruce Menge and Paul Dayton, introduced Svata Louda to the challenge of understanding biological interactions in the field; and subsequent academic advisors, including Boyd Collier, Joe Connell, Tom Ebert, Bob Luck, Bill Murdoch, Jim Rodman, and Paul Zedler, continued that process. They all have her appreciation, but they also may bear some responsibility for the outcome! Our studies of thistle-insect interactions could not have been done without the funding provided by the Research Council of the University of Nebraska (1984–1997), the National Science Foundation (DEB92-21065, DEB96-15299), and the Nature Conservancy’s Rodney Johnson and Katharine Ordway Stewardship Endowments to S.M.L. and the David H. Smith Postdoctoral Fellowship to T.A.R.

CHAPTER 2

The Overriding Importance of Environmental Context in Determining the Outcome of Species-Deletion Experiments

Bruce A. Menge

Classic ecological experiments such as those of Paine (1966, 1974) leave little doubt that species loss can have profound consequences on a community. At the same time, there can be striking variation in these consequences. Thus, to understand the consequences of a loss of species from a community, one must learn the factors that are responsible for the variation observed in studies of removals, losses, introductions, or invasions of species. An ultimate goal in such efforts is prediction: can we forecast what will happen when a species is deleted from a community or ecosystem? It is still far from clear whether a meaningful prediction can be made of population or community consequences of species loss. The most that we can accomplish may be to understand the changes that have occurred and, from this understanding, make only general predictions of the outcomes of activities that are likely to lead to species loss.

Predicting the outcomes of species loss depends on discovering the rules that regulate species interactions and their aftereffects, and how these vary with environmental conditions (Belyea and Lancaster 1999). My goal in this contribution is to probe a small part of this issue. My primary focus is on environmental context, both abiotic and biotic, and the extent to which it dictates the aftermath of species loss in natural communities. Because a vast literature has developed on the consequences of species loss, here I restrict myself largely to examples that involve consumers in marine hard-bottom communities. Much of my treatment summarizes and synthesizes examples taken from my own research activities during a 30 +-year career that was launched with the guidance and insights of my dissertation advisor and mentor, Bob Paine. It is with the deepest respect that I dedicate this paper to Bob on the occasion of his retirement from the Department of Zoology at the University of Washington. The debt I owe him is large. My career-long focus on the factors that structure and regulate ecological communities was given its initial push—plus numerous prods along the way—by Bob.

Concepts of Species Impact

Species are far from equivalent in their impacts on communities. This fact was revealed clearly by Paine’s sea star manipulations in Washington, New Zealand, and Chile (Paine 1966, 1971, 1974; Paine et al. 1985). The keystone species concept that was fostered by this work (Paine 1969a) has become an integral feature of ecological theory and practical application, particularly in some schemes of ecosystem management (Mills et al. 1993). Despite recent controversy (Mills et al. 1993; Paine 1995; Power et