David Stasek Dissertation Final

MIAMI UNIVERSITY The Graduate School

Certificate for Approving the Dissertation

We hereby approve the Dissertation of David Jon Stasek

Candidate for the Degree: Doctor of Philosophy

______Director Thomas O. Crist

______Reader Ann L. Rypstra

______Reader Bruce A. Steinly

______Michael J. Vanni

______Graduate School Representative Martin Henry H. Stevens ABSTRACT

POPULATION RESPONSES OF A GENERALIST INSECT PREDATOR AND ITS PREY TO PATCH CHARACTERISTICS IN FORAGE CROPS

by David Jon Stasek

There is a large body of literature documenting the effect of habitat features on populations of single species, but there are fewer studies documenting the effects of patch characteristics on community and ecosystem processes. Specifically, there are few studies that document the effect of dispersal among habitat patches on species interactions. Using an experimental field of forage crops, I assessed the population response of a generalist predatory insect, the damsel bug ( Nabis spp.), to prey density and the patch characteristics of size, matrix type, and habitat fragmentation. I then determined the functional response and preference of the damsel bug to two common leafhopper species in the forage-crop system: the constricted leafhopper ( Agallia constricta ) and the clover leafhopper ( Ceratagallia agricola ). Finally, I studied how dispersal among habitat patches and leafhopper and damsel bug density affected the survival of A. constricta using connected experimental mesocosms. In the experimental field, damsel bugs had higher abundances in large patches and patches surrounded by an orchard-grass matrix. Both damsel bug nymphs and adults aggregated in patches with high densities of constricted leafhoppers and aphids. There was no effect of habitat fragmentation. In functional response experiments, the damsel bug displayed a Type I functional response to both species of leafhoppers and showed a preference for A. constricta . In experimental mesocosms, there was no difference in the survival probabilities of leafhoppers in any treatment when isolation and conspecific density were varied. Damsel bug density and dispersal did affect the survival probability of leafhoppers with leafhoppers having lower survival rates in patches that were not isolated from one another. This study is one of the first experimental tests of varying dispersal to assess the survival probability of species within a metacommunity, as well as assessing the effects of patch characteristics on interspecific interactions. Results from this research also indicate that the damsel bug may be a candidate to help control A. constricta in agroecosystems.

POPULATION RESPONSES OF A GENERALIST INSECT PREDATOR AND ITS PREY TO PATCH CHARACTERISTICS IN FORAGE CROPS

A DISSERTATION

Submitted to the Faculty of

Miami University in partial

fulfillment of the requirements

for the degree of

Doctor of Philosophy

Department of Zoology

David Jon Stasek

Miami University

Oxford, OH

2009

Dissertation Advisor: Thomas O. Crist TABLE OF CONTENTS

List of Tables iii

List of Figures iv

Acknowledgements vi

1 Literature Review 1

2 The effects of landscape features and prey density 14 and composition on a generalist insect predator in red clover and grass forage crops

3 The predation rate and functional response of a 37 generalist insect predator to leafhopper prey: the roles of prey preference and habitat complexity

4 The effects of dispersal and predator density on 55 survival time of insects in an insect-red clover metacommunity

5 Synthesis 75

ii LIST OF TABLES Chapter 2

Table 1. Results of generalized linear models for treatment 29 effects on damsel bug nymphs

Table 2. Herbivore covariates of best-fitting treatment models 30 for damsel bug nymphs.

Table 3. Results of generalized linear models for 31 treatment effects on damsel bug adults.

Table 4. Herbivore covariates of best-fitting treatment 32 models for damsel bug adults.

Chapter 4

Table 1. Survival probabilities (±SE) and mean (±SD) 70 A. constricta surviving per day in response to varied isolation and leafhopper density. “Isolation” is the percentage of time per week that connecting tubes were open to movement. “Density” is the density of leafhoppers per mesocosm. “Time” is the hours since the experiment commenced.

Table 2. Survival probabilities (±SE) and mean (±SD) 71 A. constricta surviving per day in response to varied isolation and damsel bug density. “Isolation” is the percentage of time per week that connecting tubes were open to movement. “Density” is the density of damsel bugs per mesocosm. “Time” is the hours since the experiment commenced. Different letters after isolations indicate a significant difference in survival probability (P<0.05).

Chapter 5

Table 1. Summary of damsel bug responses towards 85 A. constricta and C. agricola . Results come from both field and mesocosm experiments. “Adult aggregation” refers to when prey species were added as covariates into generalized linear models to determine which prey species damsel bug adults aggregated.

iii LIST OF FIGURES

Chapter 2

Figure 1. Design of the experimental red clover field. 33 Dark, speckled areas represent red clover patches, light grey areas represent orchard grass, and white areas represent bare ground.

Figure 2. Damsel bug density in July for A) nymphs 34 and adults (B and C) plotted against prey density. The pea aphid was the best-fitting model with prey in July for nymphs, while a model with both the pea aphid and the constricted leafhopper was the best-fitting model in July for adults. Both damsel bug and prey densities were ln-transformed.

Figure 3. Damsel bug nymph density plotted against prey 35 density in September. The best-fitting model contained the main effects of size and matrix type and the constricted leafhopper (A and B) and the spotted alfalfa aphid (C and D). Both damsel bug and prey densities were ln-transformed.

Figure 4. Damsel bug adult density in September. Matrix 36 type was the best model to explain damsel bug adult abundance. A model with both A) the spotted alfalfa aphid and B) the constricted leafhopper was the best-fitting model with prey added. Both damsel bug and prey densities were ln-transformed.

Chapter 3

Figure 1. The proportion of prey eaten plotted against prey 51 density for A) Agallia constricta and B) Ceratagallia agricola . The damsel bug displayed a Type I functional response to both species with a near- constant proportion of prey eaten at each density.

Figure 2. Calculated preference of the damsel bug to 52 A) A. constricta and B) C. agricola at five densities (n=5). Whiskers represent 25 th and 75 th percentiles, the dark line indicates the median, and the edges of the boxes represent the 10 th and 90 th deciles.

iv Figure 3. Survival probability of A. constricta at five 53 densities with one damsel bug present over the course of 5 d. Treatments with a different letter are significantly different from each other (P= 0.05).

Figure 4. The effect of habitat complexity and damsel bug 54 density on the predation rate of C. agricola . More leafhoppers were consumed in the single damsel bug, bare ground treatment than in the single damsel bug, clover treatment. Error bars are ±1 SD.

Chapter 4

Figure 1. Diagram of experimental unit from two views 72 A) Above and B) Ground level. Each circle or cylinder represents a community connected by tubing to allow dispersal of insects among communities. A pot of red clover was placed in each cylinder.

Figure 2. Survival probability of A. constricta in the 73 absence of predation at densities of A) 25 leafhoppers per mesocosm and B) 50 leafhoppers per mesocosm.

Figure 3. Survival probability of A. constricta with 74 A) 1 damsel bug per mesocosm and B) 2 damsel bugs per mesocosm.

Chapter 5

Figure 1. Arthropod food web in experimental 86 forage-crop ecosystems. Interaction strengths are indicated by the thickness of connecting lines. Box line thickness indicates the importance of the organism in the food web.

v ACKNOWLEDGEMENTS

I want to thank my advisor Tom Crist for all of his support and guidance over the past six years. I also wish to thank my committee members Ann Rypstra, Bruce Steinly, Hank Stevens, and Mike Vanni for all of their insightful suggestions and support over the course of my dissertation and master’s research. Many thanks go to Ashley Boerger, Stephanie Nguyen, and James Radl for all of their help in the field and laboratory. Thank you to Drs. John Miller and Michael Melampy for all of their inspiration and advice. Finally, a big thanks to my parents, Dennis and Nancy, and to all of my relatives and friends for all of their love and support. I could not have done it without you.

vi Chapter 1

Literature Review

As natural habitats are altered by human activities, it is increasingly important to determine the impact of these alterations on population, community, and ecosystem processes. As a result of habitat loss, animals must often move among local habitat patches to find food resources (Hanski et al. 1994, Kuussaari et al. 1996), mating opportunities (Kuussaari et al. 1996, Šálek and Marhoul 2008), or over-wintering sites (Thies and Tscharntke 1999, Tscharntke et al. 2002), thus linking population dynamics among patches. There is now a large body of literature documenting how the alteration of patch characteristics affects population dynamics and species abundance. Characteristics such as habitat patch size (Gustafson and Gardner 1996, Moilanen and Hanski 1998), isolation (Moilanen and Hanski 1998), shape (Diamond 1975, Stamps et al. 1987), arrangement (Holyoak 2000), and quality (Kuussaari et al. 1996, Thomas et al. 2001) all affect population dynamics of species in local habitat patches in fragmented landscapes. In addition to patch characteristics, the type of edge habitat surrounding a patch (Stamps et al. 1987, Schtickzelle and Baguette 2003), as well as the intervening matrix habitat (Ricketts 2001, Stasek et al. 2008), influence dispersal rates among suitable habitat patches. Metapopulation biology was first developed for biological control programs to screen for effective natural enemies given that pest dispersal among fields would allow for recolonization following local extinction (Levins 1969). A classic metapopulation model assumes that all patches are equal in size and isolation, local dynamics are not synchronized with other local patches, emigration and immigration rates are low enough to allow asynchronous local dynamics, and all patches are susceptible to extinction and colonization (Levins 1969). Since Levins’ seminal paper, several different metapopulation models have been developed based on the dispersal rates of individuals and patch dynamics. The rate of dispersal will influence the degree to which population dynamics are synchronized among habitat patches which will effect persistence time of the metapopulation. In a

1 classic metapopulation, between 5 and 30% of individuals in a population will move among habitat patches (Levins 1969, Hanski et al. 1994, Hill et al. 1996). A patchy metapopulation has greater dispersal rates than a classic metapopulation such that dynamics among patches are highly synchronized and the metapopulation behaves as a single, large population (Harrison 1991, Harrison and Taylor 1997, Sutcliffe et al. 1997). An island-mainland metapopulation contains one, large patch with a low extinction probability which serves as a source for colonists for other local patches (Harrison et al. 1988). A species in a patchy landscape may display attributes of more than one of these models; therefore, the dynamics that drive the metapopulation should be used to determine the persistence of the metapopulation (Hill et al. 1996, Harrison and Taylor 1997, Sutcliffe et al. 1997). Despite the large body of literature on single-species metapopulations, there are far fewer studies on the role of dispersal among suitable habitat patches on trophic interactions. Trophic interactions in ecological communities have often been studied in isolation from other suitable habitat patches (Paine 1966, Beckerman et al. 1997, Schmitz et al. 1997, Schmitz 2003). Metacommunity biology is a rapidly growing field of ecology which examines the effect of dispersal rates on biodiversity and interspecific interactions such as predation, parasitism, herbivory, and competition (Leibold et al. 2004, Holyoak et al. 2005). Leibold et al. (2004) and Holyoak et al. (2005) characterize four major perspectives of metacommunity dynamics within patchy landscapes. While these perspectives are simplifications, they can serve as a starting point to determining the underlying dynamics of a metacommunity. The patch-dynamic perspective extends the theory of island biogeography (MacArthur and Wilson 1967). It assumes that all patches are similar and undergo both stochastic and deterministic extinction and recolonization. Dispersal rates among patches are low; otherwise, a dominant competitor or predator could drive species to local and possibly regional extinction (Holyoak et al. 2005). There must also be a tradeoff between competition and colonization ability in order for competitors to coexist (Holyoak et al. 2005). The species-sorting perspective assumes patches are dissimilar, and that variation in local habitat quality determines population dynamics. This perspective is similar to traditional niche theory (MacArthur 1958, Pianka 1966). The mass-effects perspective is

2 a multispecies model of source-sink dynamics and the rescue effect where population densities, or “mass”, influence dispersal rates among local communities (Brown and Kodric-Brown 1977, Shmida and Wilson 1985, Pulliam 1988, Holyoak et al. 2005). As in the species-sorting perspective, variation in habitat quality among patches is key to species distributions and coexistence. Unlike the species-sorting perspective, however, dispersal rates may be high enough to synchronize dynamics among habitat patches (Holyoak et al. 2005), or create spillover of species to less suitable patches (Rand et al. 2006). The neutral perspective assumes all individuals of trophically similar species have the same fitness, ecological drift leads to local extinction, and migration from the larger metacommunity results in recolonization. These models are considered a null hypothesis for the other three perspectives (Bell 2000, Hubbell 2001). Neutral models assume species have the same extinction and speciation rates and can be considered one endpoint of a continuum of coexistence mechanisms (Holyoak et al. 2005). While there is a well-developed body of metacommunity theory, there are few experimental tests of this theory. Studies with protists and microbes in laboratory microcosms show that persistence time and diversity increase with intermediate levels of dispersal in the presence of predation (Holyoak and Lawler 1996a, 1996b, Cadotte and Fukami 2005, Hauzy et al. 2007) and with trophically similar species (Cadotte 2006), while Forbes and Chase (2002) found no effect of dispersal. Kneitel and Miller (2003) examined the effect of dispersal rate on the abundance of organisms. They determined that local protozoan and rotifer abundance within pitcher plant communities increased with increasing dispersal rates.

There are even fewer studies that experimentally control dispersal rates of individuals to examine interspecific interactions. Howeth and Leibold (2008) determined the effect of varying dispersal rates on tri-trophic interactions, with dispersal dampening trophic cascades in all dispersal treatments relative to isolated communities. Laboratory studies of host-parasitoid interactions have shown that persistence time of predator and prey populations increases at intermediate levels of dispersal (Bonsall et al. 2002, 2005, Bull et al. 2006). Intermediate levels of dispersal for both predators and their prey allow predator and prey metapopulations to persist the longest due to local habitat patches having a higher probability of being recolonized than at low dispersal rates (Crowley

3 1981, Nachman 1987, Holyoak and Lawler 1996b). Intermediate levels of dispersal also permit the greatest abundances of predators as local populations are often rescued from extinction by immigrants from other local populations (Brown and Kodric-Brown 1977, Holyoak and Lawler 1996b).

Most empirical metacommunity studies are conducted using protozoans and microbes due to their rapid generation times. This makes it possible to determine the effect of dispersal on interspecific interactions, biodiversity, and the persistence of the metacommunity across multiple generations in a manner that is both logistically and temporally feasible (but see Bonsall et al. 2002, 2005, Bull et al. 2006). However, it is not possible to measure the number of protists and microbes moving among local patches. This does not allow for calculation of the survival probability of species in the metacommunity, nor is it possible to examine the mechanistic basis for movement of individuals among local habitat patches.

Several factors will influence whether or not an organism will disperse among habitat patches. The vagility of the organism (Ricketts 2001), the permeability of the edge and matrix habitat (Stamps et al. 1987, Ricketts 2001, Stasek et al. 2008), patch quality (Kuussari et al. 1996, Thomas et al. 2001, Binckley and Resetarits 2007), the density of conspecifics (Fonseca and Hart 1996, Matthysen 2005), predation risk (Resetarits 2005), patch arrangement (Bull et al. 2006), the number of habitat patches (Bonsall et al. 2002, 2005), connectivity (Matthiessen et al. 2007), and the presence of superior competitors (Tilman 1994) may all influence dispersal among local habitat patches.

It is important to understand the causes of dispersal because metacommunity persistence requires dispersal among habitat patches (Holyoak et al. 2005). Since predators often have lower dispersal rates than their prey, the alteration of patch characteristics has profound impacts on their persistence in the metacommunity (Kareiva 1987, Van Nouhuys and Hanski 2002). By determining which factors hinder or enhance movement among patches, landscapes can be managed to help promote predator or prey persistence.

4 The impact of dispersal on interspecific interactions has applications to agriculture for biological control programs. Many biological control agents are specialists on one species of prey, such as the beetle Rodolia cardinalis (Coleoptera: Coccinellidae) and the fly Cryptochaetum iceryae (Diptera: Cryptochaetidae) on the cottony-cushion scale (Quezada and DeBach 1973), or the parasitoid wasp Aphidius ervi (Hymenoptera: Braconidae) on the pea aphid (Rauwald and Ives 2001, Gross et al. 2005), while some pest species have no known specialist parasitoids. However, due to their trophic position, predator and parasitoid populations are often characterized as small and susceptible to extinction (Kareiva 1987, Kruess and Tscharntke 1994, 2000, Holt et al. 1999, Kruess 2003). Many of these species need additional resources not found in agricultural fields, such as supplemental food sources and over-wintering sites (Thomas et al. 1991, Thies and Tscharntke 1999, Elliott et al. 2002, Tscharntke et al. 2002, Tylianakis et al. 2004). Therefore, it is necessary to quantify the dispersal rates of predators and parasitoids between habitat types to determine the viability of these populations. Specialists are often considered good biological control agents due to their rapid generation times relative to their prey and because they have minimal effects on non- target species (Huffaker and Messenger 1976, DeBach and Rosen 1991). Often there is a time lag from the time of parasitoid adult emergence to the depression of prey populations while the parasitoid larvae develop inside the host (Snyder and Ives 2003). In contrast, generalist predators are often thought to be poor biological control agents due to their broad diets and low numerical responses relative to their prey (Huffaker and Messenger 1976, DeBach and Rosen 1991). At odds with this generalization, however, is a recent body of research that has determined that they may be effective agents despite these limitations (Snyder and Wise 2001, Cardinale et al. 2003, Snyder and Ives 2003). Specifically, generalists may act in concert with specialist parasitoids by depressing target pest populations while the specialist parasitoid larva is developing within the host (Snyder and Ives 2003). Though some parasitoid larvae are killed when the generalist predator consumes the host, the generalist predator guild may have an additive, rather than an antagonistic, effect on prey suppression (Snyder and Ives 2003). Generalists may

5 also depress a non-target species to allow increased parasitism by a specialist parasitoid on the target pest species (Cardinale et al. 2003). My dissertation research had several aims. First, I determined the population response of a generalist insect predator, the damsel bug ( Nabis spp., Heteroptera: Nabidae), to altered habitat size and arrangement in an experimental agroecosystem of red clover, orchard grass, and bare ground. There are many studies that document the response of individual species to landscape characteristics, but there are fewer studies which examine these effects on interspecific interactions (but see Kruess and Tscharntke 1994, 2000, Steffan-Dewenter and Tscharntke 1999, Tscharntke et al. 2002, Van Nouhuys and Hanski 2002, Diekötter et al. 2007). There are also few studies that document prey preferences in damsel bug nymphs and adults (Donahoe and Pitre 1977, Nadgauda and Pitre 1978, Sloderbeck and Yeargan 1983), and only one study documents a switch in prey preference from nymphs to adults (Nadgauda and Pitre 1986). Therefore, I also examined if the damsel bug aggregated with certain prey species based on its developmental stage. After determining the population responses in larger experimental patches, I used laboratory mesocosms to examine the functional responses of the damsel bug to two of its leafhopper prey, the constricted leafhopper ( Agallia constricta , Homoptera: Cicadellidae) and the clover leafhopper ( Ceratagallia agricola , Homoptera: Cicadellidae). Both prey species are pests upon red clover, alfalfa, and other forage legumes, and were two of the most common herbivores in the experimental field. My aims for this part of the study were to first determine the functional response of the damsel bug to each prey species and then assess if the damsel bug preferred one prey species to the other. I then examined the effect of varying habitat complexity and the presence of alternative prey on the incidence of cannibalism among damsel bugs, which are known to cannibalize one another (Lattin 1989). Finally, I studied the effect of varying connectivity and damsel bug density on the survival probability and dispersal rates of leafhoppers in experimental mesocosms. While several studies have examined predator-prey metapopulation dynamics (Holyoak and Lawler 1996a, 1996b, Bonsall et al. 2002, 2005, Bull et al. 2006), there are no studies which document the effect of varying the connectivity of habitat patches on predator-prey interactions to simulate the endpoints of a classic metapopulation and a patchy

6 population. These approaches allowed me to examine the response of the damsel bug to its insect prey at the mesocosm and field levels, as well as helping to determine if the damsel bug is a candidate for the biological control of the leafhoppers A. constricta and C. agricola .

7 References Cited

Beckerman, A.P., Uriarte, M. and Schmitz, O.J. 1997. Experimental evidence for a behavior-mediated trophic cascade in a terrestrial food chain. Proceedings of the National Academy of Sciences 94: 10735-10738.

Bell, G. 2000. The distribution of abundance in neutral communities. American Naturalist 155: 606-617.

Binckley, C.A. and Resetarits, Jr., W.J. 2007. Effects of forest canopy on habitat selection in treefrogs and aquatic insects: implications for communities and metacommunities. Oecologia 153: 951-958.

Bonsall, M.B., French, D.R., and Hassell, M.P. 2002. Metapopulation structures affect persistence of predator-prey interactions. Journal if Animal Ecology 71: 1075- 1084.

Bonsall, M.B., Bull, J.C., Pickup, N.J., and Hassell, M.P. 2005. Indirect effects and spatial scaling affect the persistence of multispecies metapopulations. Proceedings of the Royal Society of London B 272: 1465-1471.

Brown, J.H. and Kodric-Brown, A. 1977. Turnover rates in insular biogeography: effect of immigration on extinction. Ecology 58: 445-449.

Bull, J.C., Pickup, N.J., Hassell, M.P., and Bonsall, M.B. 2006. Habitat shape, metapopulation processes and the dynamics of multispecies predator-prey interactions. Journal of Animal Ecology 75: 899-907.

Cadotte, M.W. 2006. Metacommunity influences on community richness at multiple spatial scales: a microcosm experiment. Ecology 87: 1008-1016.

Cadotte, M.W. and Fukami, T. 2005. Dispersal, spatial scale, and species diversity in a hierarchically structured experimental landscape. Ecology Letters 8: 548-557.

Cardinale, B.J., Harvey, C.T., Gross, K., and Ives, A.R. 2003. Biodiversity and biocontrol: emergent impacts of a multi-enemy assemblage on pest suppression and crop yield in an agroecosystem. Ecology Letters 6: 857-865.

Crowley, P.H. 1981. Dispersal and the stability of predator-prey interactions. American Naturalist 118: 673-701.

DeBach, P. and Rosen, D. 1991. Biological control by natural enemies. Cambridge University Press, New York, NY, USA.

Diamond, J.M. 1975. The island dilemma: lessons of modern biogeographic studies for the design of nature reserves. Biological Conservation 7: 129-146.

8 Diekötter, T., Haynes, K.J., Mazeffa, D., and Crist, T.O. 2007. Direct and indirect effects of habitat area and matrix composition on species interactions among flower- visiting insects. Oikos 116: 1588-1598.

Donahoe, M.C. and Pitre, H.N. 1977. Reduviolus roseipennis behavior and effectiveness in reducing numbers of Heliothis zea on cotton. Environmental Entomology 6: 872-876.

Elliott, N.C., Kieckhefer, R.W., and Beck, D.A. 2002. Effect of aphids and the surrounding landscape on the abundance of Coccinellidae in cornfields. Biological Control 24: 214-220.

Fonseca, D.M. and Hart, D.D. 1996. Density-dependent dispersal of black fly neonates is mediated by flow. Oikos 75:49-58.

Forbes, A.E. and Chase, J.M. 2002. The role of habitat connectivity and landscape geometry in experimental zooplankton metacommunities. Oikos 96: 433-440.

Gross, K., Ives, A.R. and Nordheim, E.V. 2005. Estimating fluctuating vital rates from time-series data: a case study of aphid biocontrol. Ecology 86: 740-752.

Gustafson, E.J. and Gardner, R.H. 1996. The effect of landscape heterogeneity on the probability of patch colonization. Ecology 77: 94-107.

Hanski, I., Kuussaari, M., and Nieminen, M. 1994. Metapopulation structure and migration in the butterfly Melitaea cinxia . Ecology 75: 747-762.

Harrison, S. 1991. Local extinction in a metapopulation context: an empirical evaluation. Biological Journal of the Linnean Society 42: 73-88.

Harrison, S. and Taylor, A.D. 1997. Empirical evidence for metapopulation dynamics. pp. 27-42 in I.A. Hanski and M.E. Gilpin, editors. Metapopulation Biology: Ecology, Genetics, and Evolution. Academic Press, San Diego, CA, USA.

Harrison, S., Murphy, D.D., and Ehrlich, P.R. 1988. Distribution of the bay checkerspot butterfly, Euphydryas editha bayensis : Evidence for a metapopulation model. American Naturalist 132: 360-382.

Hauzy, C., Hulot, F.D., and Loreau, M. 2007. Intra- and interspecific density-dependent dispersal in an aquatic prey-predator system. Journal of Animal Ecology 76: 552- 558.

Hill, J.K., Thomas, C.D. and Lewis, O.T. 1996. Effects of habitat patch size and isolation on dispersal by Hesperia comma butterflies: implications for metapopulation structure. Journal of Animal Ecology 65: 725-735.

9 Holt, R.D., Lawton, J.H., Polis, G.A., and Martinez, N.D. 1999. Trophic rank and the species-area relationship. Ecology 80: 1495-1504.

Holyoak, M. 2000. Habitat patch arrangement and metapopulation persistence of predators and prey. American Naturalist 156: 378-389.

Holyoak, M. and Lawler, S.P. 1996a. Persistence of an extinction-prone predator-prey interaction through metapopulation dynamics. Ecology 77: 1867-1879.

------1996b. The role of dispersal in predator-prey metapopulation dynamics. Journal of Animal Ecology 65: 640-652.

Holyoak, M., Leibold, M.A., Mouquet, N., Holt, R.D., and Hoopes, M.F. 2005. Metacommunities: a framework for large-scale community ecology. pp. 1-31 in Holyoak, M., Leibold, M.A. and Holt, R.D., editors. Metacommunities: Spatial dynamics and ecological communities. University of Chicago Press, Chicago, IL, USA.

Howeth, J.G. and Leibold, M.A. 2008. Planktonic dispersal dampens temporal trophic cascades in pond metacommunities. Ecology Letters 11: 245-257.

Hubbell, S.P. 2001. The Unified Thoery of Biodiversity and Biogeography. Princeton University Press, Princeton, NJ, USA.

Huffaker, C.B. and Messenger, P.S. (editors). 1976. Theory and practice of biological control. Academic Press, New York, NY, USA.

Kareiva, P. 1987. Habitat fragmentation and the stability of predator-prey interactions. Nature 326: 388-390.

Kneitel, J.M. and Miller, T.E. 2003. Dispersal rates affect species composition in metacommunities of Serracenia purpurea inquilines. American Naturalist 162: 165-171.

Kruess, A. 2003. Effects of landscape structure and habitat type on a plant-herbivore- parasitoid community. Ecography 26: 283-290.

Kruess, A. and Tscharntke, T. 1994. Habitat fragmentation, species loss, and biological control. Science 264: 1581-1584.

------2000. Species richness and parasitism in a fragmented landscape: experiments and field studies with insects on Vicia sepium . Oecologia 122: 129-137.

Kuussaari, M., Nieminen, M. and Hanski, I. 1996. An experimental study of migration in the Glanville fritillary butterfly Melitaea cinxia . Journal of Animal Ecology 65: 791-801.

10 Lattin, J.D. 1989. Bionomics of the Nabidae. Annual Review of Entomology 34: 383- 400.

Leibold, M.A., Holyoak, M., Mouquet, N., Amarasekare, P., Chase, J.M., Hoopes, M.F., Holt, R.D., Shurin, J.B., Law, R., Tilman, D., Loreau, M., and Gonzalez, A. 2004. The metacommunity concept: a framework for multi-scale community ecology. Ecology Letters 7: 601-613.

Levins, R. 1969. Some demographic and genetic consequences of environmental heterogeneity for biological control. Bulletin of the Entomological Society of America 15: 237-240.

MacArthur, R.H. 1958. Population ecology of some warblers of northeastern coniferous forests. Ecology 39: 599-619.

MacArthur, R.H. and Wilson, E.O. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton , NJ, USA.

Matthiessen, B., Gamfeldt, L., Jonsson, P.R., and Hillebrand, H. 2007. Effects of grazer richness and composition on algal biomass in a closed and open marine system. Ecology 88: 178-187.

Matthysen, E. 2005. Density-dependent dispersal in birds and mammals. Ecography 28: 403-416.

Moilanen, A. and Hanski, I. 1998. Metapopulation dynamics: effects of habitat quality and landscape structure. Ecology 79: 2503-2515.

Nachman, G. 1987. Systems analysis of acarine predator-prey interactions. II. The role of spatial processes in system stability. Journal of Animal Ecology 56: 267-281.

Nadgauda D. and Pitre, H.N. 1978. Reduviolus roseipennis feeding behavior: acceptability of tobacco budworm larvae. Journal of the Georgia Entomological Society 13: 304-308.

------1986. Effects of temperature on feeding, development, fecundity, and longevity of Nabis roseipennis (Hemiptera: Nabidae) fed tobacco budworm (Lepidoptera: Noctuidae) larvae and tarnished plant bug (Hemiptera: Miridae) nymphs. Environmental Entomology 15: 536-539.

Paine, R.T. 1966. Food web complexity and species diversity. American Naturalist 100: 65-75.

Pianka, E.R. 1966. Latitudinal gradients in species diversity: A review of concepts. American Naturalist 100: 33-46.

11 Pulliam, H.R. 1988. Sources, sinks, and population regulation. American Naturalist 132: 652-661.

Quezada, J.R. and DeBach, P. 1973. Bioecological and population studies of the cottony- cushion scale, Icerya purchasi Mask., and its natural enemies, Rodolia cardinalis Mul. and Cryptochaetum iceryae Will., in Southern California. Hilgardia 41: 631- 688.

Rand, T.A., Tylianakis, J.M., and Tscharntke, T. 2006. Spillover edge effects: the dispersal of agriculturally subsidized insect natural enemies into adjacent natural habitats. Ecology Letters 9: 603-614.

Rauwald, K.S. and Ives, A.R. 2001. Biological control in disturbed agricultural systems and the rapid recovery of parasitoid populations. Ecological Applications 11: 1224-1234.

Resetarits, Jr., W.J. 2005. Habitat selection behaviour links local and regional scales in aquatic systems. Ecology Letters 8: 480-486.

Ricketts, T.H. 2001. The matrix matters: effective isolation in fragmented landscapes. American Naturalist 158: 87-99.

Šálek, M. and Marhoul, P. 2008. Spatial movements of prey partridge Perdix perdix : male-biased spring dispersal and effect of habitat quality. Journal of Ornithology 149: 329-335.

Schmitz, O.J. 2003. Top predator control of plant biodiversity and productivity in an old- field ecosystem. Ecology Letters 6: 156-163.

Schmitz, O.J., Beckerman, A.P., and O’Brien, K.M. 1997. Behaviorally mediated trophic cascades: effects of predation risk on food web interactions. Ecology 78: 1388- 1399.

Schtickzelle, N. and Baguette, M. 2003. Behavioral responses to habitat patch boundaries restrict dispersal and generate emigration-patch area relationships in fragmented landscapes. Journal of Animal Ecology 72: 533-545.

Shmida, A. and Wilson, M.V. 1985. Biological determinants of species diversity. Journal of Biogeography 12: 1-20.

Sloderbeck, P.E. and Yeargan, K.V. 1983. Comparison of Nabis americoferus and Nabis roseipennis (Hemiptera: Nabidae) as predators of the green cloverworm (Lepidoptera: Noctuidae). Environmental Entomology 12: 161-165.

Snyder, W.E. and Ives, A.R. 2003. Interactions between specialist and generalist natural enemies: parasitoids, predators, and pea aphid biocontrol. Ecology 84: 91-107.

12 Snyder, W.E. and Wise, D.H. 2001. Contrasting trophic cascades generated by a community of generalist predators. Ecology 82: 1571-1583.

Stamps, J.A., Buechner, M., and Krishnan, V.V. 1987. The effects of edge permeability and habitat geometry on emigration from patches of habitat. The American Naturalist 129: 533-552.

Stasek, D.J., Bean, C. and Crist, T.O. 2008. Butterfly abundance and movements among prairie patches: the roles of habitat quality, edge, and forest matrix permeability. Environmental Entomology 37: 897-906.

Steffan-Dewenter, I. and Tscharntke, T. 1999. Effects of habitat isolation on pollinator communities and seed set. Oecologia 121: 432-440.

Sutcliffe, O.L., Thomas, C.D., and Peggie, D. 1997. Area-dependent migration by ringlet butterflies generates a mixture of patchy population and metapopulation attributes. Oecologia 109: 229-234.

Thies, C. and Tscharntke, T. 1999. Landscape structure and biological control in agroecosystems. Science 285: 893-895.

Thomas, J.A., Bourn, N.A.D., Clarke, R.T., Stewart, K.E., Simcox, D.J., Pearman, G.S., Curtis, R., and Goodger, B. 2001. The quality and isolation of habitat patches both determine where butterflies persist in fragmented landscapes. Proceedings of the Royal Society London Biological Sciences 268:1791-1796.

Thomas, M.B., Wratten, S.D., and Sotherton, N.W. 1991. Creation of “island” habitats in farmland to manipulate populations of beneficial arthropods: predator densities and emigration. Journal of Applied Ecology 28: 906-917.

Tilman, D. 1994. Competition and biodiversity in spatially structured habitats. Ecology 75: 2-16.

Tscharntke, T., Steffan-Dewenter, I., Kruess, A., and Thies, C. 2002. Contribution of small habitat fragments to conservation of insect communities of grassland- cropland landscapes. Ecological Applications 12: 354-363.

Tylianakis, J.M., Didham, R.K., and Wratten, S.D. 2004. Improved fitness of aphid parasitoids receiving resource subsidies. Ecology 85: 658-666.

Van Nouhuys, S. and Hanski, I. 2002. Colonization rates and distances of a host butterfly and two specific parasitoids in a fragmented landscape. Journal of Animal Ecology 71: 639-650.

13 Chapter 2

The effects of landscape features and prey density and composition on a generalist insect predator in red clover and grass forage crops

The characteristics of habitat patches play important roles in the population dynamics of individual species. Patch attributes such as quality (Kuussaari et al. 1996, Thomas et al. 2001), size (Gustafson and Gardner 1996, Moilanen and Hanski 1998), isolation (Moilanen and Hanski 1998), shape (Diamond 1975, Stamps et al. 1987), and arrangement (Holyoak 2000) all influence the population dynamics of species in patchy landscapes. The surrounding matrix habitat (Ricketts 2001, Stasek et al. 2008) and the type of edge surrounding habitat patches (Stamps et al. 1987, Schtickzelle and Baguette 2003) also influence dispersal rates of individuals among patches or provide additional food resources and mating opportunities. Fewer studies document how these characteristics influence species interactions such as pollination (Steffan-Dewenter and Tscharntke 1999, Diekötter et al. 2007), parasitism (Kruess and Tscharntke 1994, 2000, Tscharntke et al. 2002, Van Nouhuys and Hanski 2002), and predation (Kareiva 1987). Recently, however, this is a rapidly growing area of interest due to the applications of this research to ecosystem services such as the pollination of crops and biological control of pests (Tscharntke et al. 2005, 2007, Kremen et al. 2007). Habitat fragmentation may release herbivores from control by their predators and parasitoids, potentially leading to outbreaks of pest species (Kareiva 1987, Kruess and Tscharntke 1994, Schoener et al. 1995, Tscharntke et al. 2002, Van Nouhuys and Hanski 2002). As a result of fragmentation, predators and parasitoids may be unable to colonize isolated habitat patches as quickly as their more vagile prey, and numerical responses of predators may be insufficient to suppress pest species in patchy habitats (Kareiva 1987, Van Nouhuys and Hanski 2002). Specialist parasitoids are at a particularly high risk of local extinction in fragmented landscapes since they are dependent on a narrow range of host species to complete their life cycle (Van Nouhuys and Hanski 2002). If prey populations decrease or become locally extinct, specialist parasitoids may be unable to disperse to other patches to find their host species. In agricultural systems, this risk of extinction may be decreased when there is a diversity of habitat types within or among crop fields or pastures. These additional habitats may contain alternative food sources

14 and over-wintering sites which predators and parasitoids need to complete their life cycle (Thomas et al. 1991, Thies and Tscharntke 1999, Tscharntke et al. 2002). While several successful biological control programs use specialist parasitoids for pests such as the pea aphid (Rauwald and Ives 2001, Gross et al. 2005), cottony-cushion scale (Quezada and DeBach 1973), and the cabbage worm (Cameron and Walker 2002), the use of generalist predators as biological control agents is more controversial. Generalist predators are thought to be ineffective biological control agents due to their broad diets and low numerical response relative to their prey (Huffaker and Messenger 1976, DeBach and Rosen 1991). As a result, generalist predators will often engage in cannibalism or intraguild predation, which may actually release crop pests from predator control (Rosenheim et al. 1993, Snyder and Ives 2001, Snyder and Wise 2001). However, a few studies demonstrate that generalists can be good control agents despite these limitations (Snyder and Wise 2001, Cardinale et al 2003, Snyder and Ives 2003). Since some pests have no known specialist parasitoids, generalist predators may be the only form of biological control available to help control these pests. In this study, I examined the population responses of a generalist predatory insect (damsel bugs, Heteroptera: Nabidae) to the patch characteristics of size, fragmentation, and matrix type, and determined if higher densities of damsel bug nymphs and adults per unit area associate with particular prey species in an experimental field of red clover and orchard grass. I hypothesized that patch size, fragmentation, and the surrounding matrix habitat would influence the abundance of the damsel bug and their herbivorous prey. Large, unfragmented red clover patches surrounded by an orchard-grass matrix were predicted to support higher abundances of herbivorous insects and damsel bugs due to the greater availability of preferred and supplementary food resources.

Materials and Methods

Experimental design

An experimental field of red clover, orchard grass, and bare ground was established in 2004 at the Miami University Ecology Research Center in Oxford, OH.

15 Details of the study design and sampling protocols can be found in Diekötter et al. (2007) and Haynes and Crist (in press). Briefly, 32 plots, 14 × 14 m in area, were sown from seed with red clover ( Trifolium pratense L.) as the focal host plant and orchard grass (Dactylis glomerata L.) as the matrix habitat in half of the plots. The other half of the plots contained a bare-ground matrix habitat. Each plot contained four subplots of red clover. To separate the effects of habitat area from habitat fragmentation per se (Fahrig 2003), clover patches varied in size (4 or 16 m2) and level of fragmentation (connected or separated by 2 m). There were four replicates of each combination of area, fragmentation, and matrix habitat according to a full-factorial design (Fig. 1). Four control plots of orchard grass were also established, but these were not sampled for this study. Plots were weeded regularly throughout the season and Roundup WeatherMAX® herbicide (Monsanto, St. Louis, MO, USA) was applied to the bare-ground matrix habitat and to the area between plots every month.

Insect sampling Plots were sampled once in June, July, and September of 2005 using a D-vac® suction-sampler to provide a uniform sampling area of 0.086 m 2. A sample was obtained from a randomly selected location within each quadrat of a clover subplot for a plot total of 16 clover samples. An additional four samples were taken from the surrounding matrix habitat at random locations from each quadrat of the plot for a total of 16 matrix samples. Samples were frozen, and insects were later preserved in alcohol and sorted in the laboratory. A synoptic collection of the insects collected is in the Crist laboratory at Miami University. Damsel bugs (Nabis spp., Heteroptera: Nabidae) were the most abundant predator and seven species of herbivores comprised over 90% of the herbivore abundance (Crist and Haynes, unpublished data). Therefore, I restricted my analyses to these eight species. All damsel bugs were identified to genus because it was not possible to identify damsel bug nymphs to species. Herbivores were identified to species. The seven most abundant herbivores were the spotted alfalfa aphid ( Therioaphis maculata Buckton, Homoptera: Aphididae), the pea aphid ( Acyrthosiphon pisum Harris, Homoptera: Aphididae), the meadow spittlebug ( Philaenus spumarius L., Homoptera: Cercopidae), the tarnished plant bug ( Lygus lineolaris (Palisot de Beauvois), Heteroptera:

16 Miridae), the potato leafhopper ( Empoasca fabae Harris, Homoptera: Cicadellidae), the clover leafhopper ( Ceratagallia agricola (Hamilton), Homoptera: Cicadellidae), and the constricted leafhopper ( Agallia constricta Van Duzee, Homoptera: Cicadellidae). The clover leafhopper prefers red clover as its host plant, the spotted alfalfa aphid prefers alfalfa but will feed on clovers, and the pea aphid prefers legumes. Therefore, these three species can be considered specialists (Tscharntke et al. 2002). The tarnished plant bug, constricted leafhopper, meadow spittlebug, and the potato leafhopper will feed on multiple families of host plants and are considered generalists. Damsel bugs are generalist predators that prey on many different arthropod taxa including aphids, leafhoppers, mirids, and will cannibalize conspecifics (Lattin 1989), though they prefer aphids (Flinn et al. 1985, Östman and Ives 2003). Damsel bugs are diurnal, folivorous predators that actively forage for prey using chemoreception and vibrations (Irwin and Shepard 1980, Braman and Yeargan 1989, 1990, Freund and Olmstead 2000, Schotzko and O’Keeffe 1989). I counted damsel bug nymphs and adults separately since their developmental stage likely influenced their movement ability and aggregation patterns with herbivores.

Data Analysis Damsel bug and prey densities were expressed at the plot level for analyses. Species densities in each plot were weighted according to the total area of clover and matrix habitat in each plot. Weighted densities in the clover and matrix habitat were then summed and divided by the total area of the plot (196 m 2) to give a density (per m 2) for the overall plot. Densities of damsel bugs and their prey were ln-transformed prior to analysis. Plot-level densities of damsel bugs were analyzed using generalized linear models (R Development Core Team 2008). Densities of damsel bug nymphs and adults at the plot level were modeled as variables in separate analyses for the July and September sampling periods. Damsel-bug counts were too low in June to conduct analyses. The main effects of habitat area, fragmentation, and matrix type were tested along with all possible interactions. The best-fitting model was determined by hierarchical selection and the lowest AIC value. After selecting the best model for experimental treatments,

17 prey densities for each species were then added as covariates into the model to determine if damsel bug adults and nymphs aggregate to certain prey species. Prey covariates were retained in the model using AIC criteria. My analysis of predator responses to prey densities assume there are no time lags in the predator’s numerical response to changes in prey densities. Regression models do not account for this time lag, and predator densities may be positively or negatively correlated to prey density (Carpenter et al. 1985); therefore, any inference of predation rate based on the prey covariates in the model must be done cautiously. Here, I infer a positive response of a predator to prey as an indication that damsel bug predators aggregate to their prey, possibly based on patch characteristics. Experimental manipulations of prey and predator density are necessary to determine if the predator does indeed feed on the potential prey item (see Chapters 3 and 4), and whether there are bottom-up effects of prey on predators (positive relationships) or top-down effects of predators on prey density (negative relationships) (Diekötter et al. 2007).

Results Damsel bug nymph abundance in July was best predicted using patch size, with larger patches containing higher nymph abundance (Table 1, Fig. 2). A competing model included the main effects of size and fragmentation as well as a size-fragmentation interaction. A second competing model included the main effects of patch size and matrix type. When prey abundances were added as covariates, only the pea aphid was retained in the best-fitting model (Table 2). For September nymphs, the best-fitting model included the main effects of patch size and matrix type and a positive interaction between patch size and matrix type, with small patches surrounded by an orchard grass matrix containing greater nymph abundances than small patches surrounded by a bare- ground matrix (Table 1, Fig. 3). A competing model included the main effect of matrix type. The spotted alfalfa aphid and the constricted leafhopper were retained as prey in the best-fitting model (Table 2). Abundance of adult damsel bugs in July was influenced by patch size and the abundance of constricted leafhoppers and pea aphids. A competing model included the main effects of patch size and matrix type. Larger patches contained a greater abundance

18 of damsel bug adults (Tables 3 and 4, Fig. 2). In September, the best-fitting model contained only the treatment effect of matrix type with patches surrounded by an orchard- grass matrix containing greater abundances of adults (Table 3, Fig. 4). A competing model included the main effects of fragmentation and matrix type. The best-fitting model with prey covariates contained the spotted alfalfa aphid and the constricted leafhopper (Table 4).

Discussion My hypothesis that patch size and matrix type would affect damsel bug abundance was partially supported. Both damsel bug adults and nymphs responded positively to increased patch size in July. Patch size also influenced the abundance of nymphs in September with larger patches containing a higher abundance of nymphs. Matrix type also affected damsel bug nymph and adult abundance in September with clover patches surrounded by an orchard-grass matrix harboring greater abundances of damsel bugs. Fragmentation did not affect damsel bug nymph or adult abundance in any month. Large habitat patches often contain greater abundances of a species than small habitat patches (Kruess and Tscharntke 2000, Tscharntke et al. 2002). This has been attributed to greater abundance of resources for both predators and their prey, such as food or refuges from predators (Root 1973, Kareiva 1985). Large patches are also necessary to maintain predator and parasitoid populations due to their trophic position, which are characterized by small, variable populations (Kareiva 1987, Kruess and Tscharntke 1994, 2000, Holt et al. 1999, Kruess 2003). Clover patches surrounded by an orchard-grass matrix are more complex than patches surrounded by a bare-ground matrix. Since damsel bugs are active predators (Irwin and Shepard 1980, Freund and Olmstead 2000), this complex habitat may have decreased their encounters with conspecifics and other arthropod predators, such as wolf spiders, web-building spiders, and carabid beetles, all of which are common in our system (cf. Finke and Denno 2002). Increased habitat complexity, such as the presence of thatch, plant species diversity, and plant structural diversity, has been shown to increase natural enemy abundance (Langellotto and Denno 2004). Decreased encounter

19 rates in the complex habitat would decrease intraguild predation and, therefore, increase the predation rate on pests and strengthen trophic cascades if they are present (Finke and Denno 2003, 2004, 2005). Patches surrounded by an orchard-grass matrix likely led to a spillover of individuals from the clover into the matrix. The orchard grass provided an alternative host plant for generalist herbivores (Haynes and Crist, in press). The increased habitat area of the clover plus the grass matrix allowed for increased abundances of the potato leafhopper, constricted leafhopper, meadow spittlebug, and tarnished plant bug per plot. Damsel bugs likely responded to this increased herbivore abundance by spilling over from the red clover to the orchard grass, as has been observed for other generalist predators (Rand and Tscharntke 2007). The bare-ground matrix likely provided few or no foraging opportunities for damsel bugs, so it would be expected that damsel bugs would have lower abundances in patches surrounded by bare ground. Damsel bugs may need the more complex habitat patches relative to patches surrounded by a bare-ground matrix to complete their life cycle. Orchard grass is an effective over-wintering site for predatory arthropods (Thomas et al. 1991). Since it is a tussock-forming grass, the temperature beneath the grass is more stable than in mat- forming grass species (Thomas et al. 1991). Presumably, part of the adult damsel bug population may have over-wintered in the orchard grass during the winter of 2004-2005, and hence had greater abundances in patches surrounded by an orchard-grass matrix during 2005. Damsel bug adults and nymphs appeared similar in their responses to prey abundance with both nymphs and adults aggregating in patches with greater abundances of aphids and constricted leafhoppers. Damsel bugs are generalist predators that will consume a variety of prey including aphids, leafhoppers, mirids, and will cannibalize conspecifics (Lattin 1989). There are few studies that document feeding preferences between damsel bug nymphs and adults (Donahoe and Pitre 1977, Nadgauda and Pitre 1978, Sloderbeck and Yeargan 1983). Nadgauda and Pitre (1986) documented a switch in food preference as nymphs became adults. In their study, Nadgauda and Pitre (1986) found that damsel bug nymphs consumed more tobacco budworm larvae (Heliothis virescens , Lepidoptera: Noctuidae) than tarnished plant bug nymphs ( Lygus lineolaris ,

20 Heteroptera: Miridae). This result was attributed to the tobacco budworm’s lower mobility and weaker defenses than the tarnished plant bug. They also documented that adult damsel bugs consumed more tarnished plant bug nymphs, possibly due to their higher nutritional content which is needed for reproduction, though they did not test this hypothesis. This supports my result that damsel bug nymphs aggregated in patches with high densities of aphids due to their limited mobility and defenses. Damsel bugs often prefer aphids to other taxa of arthropod prey (Flinn et al. 1985, Östman and Ives 2003). Therefore, it was not surprising that both nymphs and adults aggregated in patches with high abundances of aphids. Aphids are wingless for most of the year, have no armored defense, move slowly, and have been documented to not avoid approaching damsel bugs (Flinn et al. 1985). Their main method of defense is to drop off plants when approached by predators, though aphids are also capable of defending with their frontal horns (Montgomery and Nault 1977, Dixon 1985). In addition, some species of aphids sequester toxins from their host plants (Wink and Witte 1991), release defensive secretions (Dixon 1985), and use ants to deter predation (Dixon 1958). Aphids are also less likely to drop from a host plant when approached by damsel bugs than the larger coccinellid beetles (Losey and Denno 1998), which are considered the most important generalist predator in suppressing pea aphid outbreaks (Frazer et al. 1981, Gutierrez et al. 1990). Damsel bug nymphs, in particular, would be more successful in attacking and subduing aphids than adult leafhoppers which can leap away from predators. Damsel bug nymphs may be more inclined to prey on aphids than on other prey types. However, further tests such as functional response and prey preference experiments are needed to confirm that nymphs will prey upon both species of aphids; therefore, I can only conclude that nymphs aggregate with aphids in the field (sensu Carpenter et al. 1985). Damsel bug adults prey on aphids for the same reasons as the damsel bug nymphs, but they also aggregated in areas of high constricted leafhopper densities, suggesting that adults prey on constricted leafhoppers in addition to aphids. A leafhopper’s main defenses are to hop, fly, or remain still to escape predation. Leafhoppers are also larger than aphids, and may therefore provide more nutrients per individual to the damsel bug. Perhaps more importantly, constricted leafhoppers were in

21 such great abundance that damsel bugs were able to easily locate and capture them. In July, constricted leafhoppers were 4x more abundant than spotted alfalfa aphids and 35x more abundant than pea aphids, while the constricted leafhopper was 2x more abundant than spotted alfalfa aphids in September. Finally, studies on the functional response of damsel bugs to constricted leafhoppers showed high predation rates (Chapter 3). This study determined that the patch characteristics of size and matrix type influenced which patches supported the greatest abundances of damsel bugs. Damsel bug nymphs and adults also aggregated in patches that contained similar prey items. Nymphs and adults both aggregated in patches with aphids and the constricted leafhopper. Few studies have documented the effect of landscape features on species interactions. This study provides evidence that the features of the landscape influence the aggregation of predators and their prey, though more research is needed to determine the extent of control damsel bugs exert on their prey in the field.

22 References Cited

Braman, S.K. and Yeargan, K.V. 1989. Intraplant distribution of three Nabis species (Hemiptera: Nabidae), and impact of N. roseipennis on green cloverworm populations in soybean. Environmental Entomology 18: 240-244.

------1990. Phenology and abundance of Nabis americoferus , N. roseipennis , and N. rufusculus (Hemiptera: Nabidae) and their parasitoids in alfalfa and soybean. Journal of Economic Entomology 83: 823-830.

Cameron, P.J. and Walker, G.P. 2002. Field evaluation of Cotesia rubecula (Hymenoptera: Braconidae), an introduced parasitoid of Pieris rapae (Lepidoptera: Pieridae) in New Zealand. Environmental Entomology 31: 367-374.

Carpenter, S.R., Kitchell, J.F., and Hodgson, J.R. 1985. Cascading trophic interactions and lake productivity. Bioscience 35: 634-639.

DeBach, P. and Rosen, D. 1991. Biological control by natural enemies. Cambridge University Press, New York, NY, USA.

Diamond, J.M. 1975. The island dilemma: lessons of modern biogeographic studies for the design of nature reserves. Biological Conservation 7: 129-146.

Diekötter, T., Haynes, K.J., Mazeffa, D., and Crist, T.O. 2007. Direct and indirect effects of habitat area and matrix composition on species interactions among flower- visiting insects. Oikos 116: 1588-1598.

Dixon, A.F.G. 1958. Escape responses shown by certain aphids to the presence of the coccinellid, Adalia decempunctata (L.). Transactions of the Royal Entomological Society London 10: 319-334.

------1985. Aphid Ecology. Blackie and Sons, London, UK.

Donahoe, M.C. and Pitre, H.N. 1977. Reduviolus roseipennis behavior and effectiveness in reducing numbers of Heliothis zea on cotton. Environmental Entomology 6: 872-876.

Fahrig, L. 2003. Effects of habitat fragmentation on biodiversity. Annual Review of Ecology and Systematics 34: 487-515.

Finke, D.L. and Denno, R.F. 2002. Intraguild predation diminished in complex-structured vegetation: implications for prey suppression. Ecology 83: 643-652.

23 ------. 2003. Intra-guild predation relaxes natural enemy impacts on herbivore populations. Ecological Entomology 28: 67-73.

------. 2004. Predator diversity dampens trophic cascades. Nature: 429: 407-410.

------. 2005. Predator diversity and the functioning of ecosystems: the role of intraguild predation in dampening trophic cascades. Ecology Letters 8: 1299-1306.

Flinn, P.W., Hower, A.A., and Taylor, R.A.J. 1985. Preference of Reduviolus americoferus (Hemiptera: Nabidae) for potato leafhopper nymphs and pea aphids. Canadian Entomologist 117: 1503-1508.

Frazer, B.D., Gilbert, N., Ives, P.M., and Raworth, D.A. 1981. Predator reproduction and the overall predator-prey relationship. Canadian Entomologist 113: 1015-1024.

Freund, R.L. and Olmstead, K.L. 2000. Role of vision and antennal olfaction in habitat and prey location by three predatory heteropterans. Environmental Entomology 29: 721-732.

Gross, K., Ives, A.R., and Nordheim, E.V. 2005. Estimating fluctuating vital rates from time-series data: a case study of aphid biocontrol. Ecology 86: 740-752.

Gustafson, E.J. and Gardner, R.H. 1996. The effect of landscape heterogeneity on the probability of patch colonization. Ecology 77: 94-107.

Gutierrez, A.P., Hagen, K.S., and Ellis, C.K. 1990. Evaluating the impact of natural enemies: a multitrophic perspective. pp. 81-107 in Mackauer, M., Ehler, L.E., and Roland, J. editors. Critical Issues in Biological Control. Intercept, Andover, UK.

Haynes, K.J. and Crist, T.O. 2009. Insect herbivory in an experimental agroecosystem: the relative importance of habitat area, fragmentation, and the matrix. Oikos In press .

Holt, R.D., Lawton, J.H., Polis, G.A., and Martinez, N.D. 1999. Trophic rank and the species-area relationship. Ecology 80: 1495-1504.

Holyoak, M. 2000. Habitat patch arrangement and metapopulation persistence of predators and prey. American Naturalist 156: 378-389.

Huffaker, C.B. and Messenger, P.S. (editors). 1976. Theory and practice of biological control. Academic Press, New York, NY, USA.

Irwin, M.E. and Shepard, M. 1980. Sampling predaceous Hemiptera on soybeans. pp. 505-531 in Kogan, M. and Herzog, D. editors. Sampling Methods in Soybean Entomology. Springer-Verlag, New York, NY, USA.

24 Kareiva, P. 1985. Finding and losing host plants b Phyllotreta: patch size and surrounding habitat. Ecology 66: 1809-1816.

------1987. Habitat fragmentation and the stability of predator-prey interactions. Nature 326: 388-390.

Kremen, C., Williams, N.M., Aizen, M.A., Gemmill-Herren, B., LeBuhn, G., Minckley, R., Packer, L., Potts, S.G., Roulston, T., Steffan-Dewenter, I., Vázquez, D.P., Winfree, R., Adams, L., Crone, E.E., Greenleaf, S.S., Keitt, T.H., Klein, A.M., Regetz, J., and Ricketts, T.H. 2007. Pollination and other ecosystem services produced by mobile organisms: a conceptual framework for the effects of land- use change. Ecology Letters 10: 299-314.

Kruess, A. 2003. Effects of landscape structure and habitat type on a plant-herbivore- parasitoid community. Ecography 26: 283-290.

Kruess, A. and Tscharntke, T. 1994. Habitat fragmentation, species loss, and biological control. Science 264: 1581-1584.

------2000. Species richness and parasitism in a fragmented landscape: experiments and field studies with insects on Vicia sepium . Oecologia 122: 129-137.

Kuussaari, M., Nieminen, M., and Hanski, I. 1996. An experimental study of migration in the Glanville fritillary butterfly Melitaea cinxia . Journal of Animal Ecology 65: 791-801.

Langellotto, G.A. and Denno, R.F. 2004. Responses of invertebrate natural enemies to complex-structured habitats: a meta-analytical synthesis. Oecologia 139: 1-10.

Lattin, J.D. 1989. Bionomics of the Nabidae. Annual Review of Entomology 34: 383- 400.

Losey, J.E. and Denno, R.F. 1998. The escape response of pea aphids to folivar-foraging predators: factors affecting dropping behavior. Ecological Entomology 23: 53-61.

Moilanen, A. and Hanski, I. 1998. Metapopulation dynamics: effects of habitat quality and landscape structure. Ecology 79: 2503-2515.

Montgomery, M.E. and Nault, L.R. 1977. Comparative response of aphids to the alarm pheromone, (E)-β-farnesene. Entomologia experimentalis et applicata 34: 347- 348.

Nadgauda D. and Pitre, H.N. 1978. Reduviolus roseipennis feeding behavior: acceptability of tobacco budworm larvae. Journal of the Georgia Entomological Society 13: 304-308.

25 ------1986. Effects of temperature on feeding, development, fecundity, and longevity of Nabis roseipennis (Hemiptera: Nabidae) fed tobacco budworm (Lepidoptera: Noctuidae) larvae and tarnished plant bug (Hemiptera: Miridae) nymphs. Environmental Entomology 15: 536-539.

Östman, Ö. and Ives, A.R. 2003. Scale-dependent indirect interactions between two prey species through a shared predator. Oikos 102: 505-514.

R Development Core Team. 2008. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3- 900051-07-0, URL http://www.R-project.org .

Rand, T.A. and Tscharntke, T. 2007. Contrasting effects of natural habitat loss on generalist and specialist aphid natural enemies. Oikos 116: 1353-1362.

Rauwald, K.S. and Ives, A.R. 2001. Biological control in disturbed agricultural systems and the rapid recovery of parasitoid populations. Ecological Applications 11: 1224-1234.

Ricketts, T.H. 2001. The matrix matters: effective isolation in fragmented landscapes. American Naturalist 158: 87-99.

Root, R.B. 1973. Organization of a plant-arthropod association in simplae and diverse habitats: the fauna of collards ( Brassica oleracea ). Ecological Monographs 43: 95-124.

Rosenheim, J.A., Wilhoit, L.R., and Armer, C.A. 1993. Influence of intraguild predation among generalist insect predators on the suppression of an herbivore population. Oecologia 96: 439-449.

Schoener, T.W., Spiller, D.A., and Morrison, L.W. 1995. Variation in the hymenopteran parasitoid fraction on Bahamian islands. Acta Oecologia 16:103-121.

Schotzko, D.J. and O’Keeffe, L.E. 1989. Comparison of sweep net, d-vac, and absolute sampling, and diel variation of sweep net sampling estimates in lentils for pea aphid (Homoptera: Aphididae), nabids (Hemiptera: Nabidae), lady beetles (Coleoptera: Coccinellidae), and lacewings (Neuroptera: Chrysopidae). Journal of Econonomic Entomology 82: 491-506.

26 Schtickzelle, N. and Baguette, M. 2003. Behavioral responses to habitat patch boundaries restrict dispersal and generate emigration-patch area relationships in fragmented landscapes. Journal of Animal Ecology 72: 533-545.

Snyder, W.E. and Ives, A.R. 2001. Generalist predators disrupt biological control by a specialist parasitoid. Ecology 82: 705-716.

------2003. Interactions between specialist and generalist natural enemies: parasitoids, predators, and pea aphid biocontrol. Ecology 84: 91-107.

Snyder, W.E. and Wise, D.H. 2001. Contrasting trophic cascades generated by a community of generalist predators. Ecology 82: 1571-1583.

Stamps, J.A., Buechner, M., and Krishnan, V.V. 1987. The effects of edge permeability and habitat geometry on emigration from patches of habitat. American Naturalist 129: 533-552.

Stasek, D.J., Bean, C., and Crist, T.O. 2008. Butterfly abundance and movements among prairie patches: the roles of habitat quality, edge, and forest matrix permeability. Environmental Entomology 37: 897-906.

Steffan-Dewenter, I. and Tscharntke, T. 1999. Effects of habitat isolation on pollinator communities and seed set. Oecologia 121: 432-440.

Thies, C. and Tscharntke, T. 1999. Landscape structure and biological control in agroecosystems. Science 285: 893-895.

Thomas, J.A., Bourn, N.A.D, Clarke, R.T., Stewart, K.E., Simcox, D.J., Pearman, G.S., Curtis, R., and Goodger, B. 2001. The quality and isolation of habitat patches both determine where butterflies persist in fragmented landscapes. Proceedings of the Royal Society London Biological Sciences 268:1791-1796.

Tscharntke, T., Bommarco, R., Clough, Y., Crist, T.O., Kleijn, D., Rand, T.A., Tylianakis, J.M., Van Nouhuys, S., and Vidal, S. 2007. Conservation biological control and enemy diversity on a landscape scale. Biological Control 43: 294-309.

27 Tscharntke, T., Klein, A.M., Kruess, A., Steffan-Dewenter, I., and Thies, C. 2005. Landscape perspectives on agricultural intensification and biodiversity— ecosystem service management. Ecology Letters: 8: 857-874.

Van Nouhuys, S. and Hanski, I. 2002. Colonization rates and distances of a host butterfly and two specific parasitoids in a fragmented landscape. Journal of Animal Ecology 71: 639-650.

Wink, M. and Witte, L. 1991. Storage of quinolizidine alkaloids in Macrosiphum albifrons and Aphis genistae (Homoptera: Aphididae). Entomologia Generalis 15: 237-254.

28 Table 1. Results of generalized linear models for treatment effects on damsel bug nymphs.

Month Treatment Mean nymph m-2(±SD) AIC Coefficient t P

July Size 30.690 Small patch 0.52±0.6 -0.7447 -5.88 <0.0001 Large patch 2.26±1.5

September Size + Matrix 48.076 +size:matrix

Size Small patch 1.22±1.5 -0.2968 -1.266 0.2160 Large patch 1.01±1.0

Matrix Orchard grass 1.68±1.5 0.1396 0.595 0.55631 Bare ground 0.56±0.6

29 Table 2. Herbivore covariates of best-fitting treatment models for damsel bug nymphs.

Month Treatment Covariate species Mean herbivore m-2 (±SD) AIC Coefficient t P

July Pea aphid 22.511 0.4456 Size Small 0.70±0.7 -0.6652 -5.769 <0.0001 Large 1.12±1.0

September 45.937 Spotted alfalfa aphid 0.2453 1.345 0.1901

Matrix Orchard grass 0.62±0.9 0.04851 0.208 0.8372 Bare ground 0.83±0.9

Size Small patch 0.73±0.9 -0.1519 -0.634 0.5314 Large patch 0.72±1.0

A. constricta 0.34853 1.607 0.1202

Matrix Orchard grass 2.62±1.2 Bare ground 1.57±1.4

Size Small patch 1.60±1.1 Large patch 2.63±1.5

30 Table 3. Results of generalized linear models for treatment effects on damsel bug adults.

Month Treatment Mean adult m-2 (±SD) AIC Coefficient t P

July Size 12.255 Small patch 0.28±0.5 -0.4004 -4.110 0.00028 Large patch 0.89±0.5

September Matrix 51.622 Orchard grass 7.44±3.9 1.5670 8.6940 <0.0001 Bare ground 0.70±0.8

31 Table 4. Herbivore covariates of best-fitting treatment models for damsel bug adults.

Month Treatment Covariate species Mean herbivore m-2 (±SD) AIC Coefficient t P

July A. constricta + 9.8258 Pea aphid Size -0.28113 -2.672 0.0124

A. constricta Small 25.60±19.5 0.12623 1.726 0.0953 A. constricta Large 44.87±23.4

Pea aphid Small 0.70±0.7 0.18101 1.649 0.1104 Pea aphid Large 1.12±1.0

September Spotted alfalfa aphid + 44.071 0.5442 3.007 0.0055 A. constricta 0.2277 1.175 0.2500

Matrix 1.5384 8.700 <0.0001

Spotted alfalfa aphid Bare 0.83±0.9 Spotted alfalfa aphid Grass 0.62±0.9

A. constricta Bare 1.57±1.4 A. constricta Grass 2.62±1.2

N 8 m 14 m 140 m

8 m 8 m 140 m 140

Clover Grass Bare Ground

Figure 1. Design of the experimental red clover field. Dark, speckled areas represent red clover patches, light grey areas represent orchard grass, and white areas represent bare ground.

33 A) B)

Large plots Large plots Small plots Small plots 2 − Damsel bug m bug Damsel 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5

0.0 0.5 1.0− 1.5 2.0 0.0 0.5 1.0− 1.5 2.0 Pea aphid m 2 Pea aphid m 2

Large plots Small plots 2 − Damsel bug m bug Damsel 0.0 0.5 1.0 1.5

0 1 2 3 4− 5 Constricted leafhopper m 2

Figure 2. Damsel bug density in July for A) nymphs and adults (B and C) plotted against prey density. The pea aphid was the best-fitting model with prey in July for nymphs, while a model with both the pea aphid and the constricted leafhopper was the best-fitting model in July for adults. Both damsel bug and prey densities were ln-transformed.

34 A) B)

Large plots Bare-ground matrix Small plots Grass matrix 2 − Damsel bug m bugDamsel 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5

0.0 0.5 1.0 1.5 2.0− 2.5 0.0 0.5 1.0 1.5 2.0− 2.5 Constricted leafhopper m 2 Constricted leafhopper m 2

C) D)

Large patches Bare-ground matrix Small patches Grass matrix 2 − Damsel bug m bugDamsel 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5

0.0 0.5 1.0 1.5− 2.0 0.0 0.5 1.0 1.5− 2.0 Spotted alfalfa aphid m 2 Spotted alfalfa aphid m 2

Figure 3. Damsel bug nymph density plotted against prey density in September. The best-fitting model contained the main effects of size and matrix type and the constricted leafhopper (A and B) and the spotted alfalfa aphid (C and D). Both damsel bug and prey densities were ln- transformed.

35 A) B)

Bare-ground matrix Grass matrix 2 − Damsel bug adult m bugDamsel adult 0 1 2 3 4 0.0 1.0 2.0 3.0

0.0 0.5 1.0 1.5− 2.0 0.0 0.5 1.0 1.5− 2.0 Spotted alfalfa aphid m 2 Constricted leafhopper m 2

Figure 4. Damsel bug adult density in September. Matrix type was the best model to explain damsel bug adult abundance. A model with both A) the spotted alfalfa aphid and B) the constricted leafhopper was the best-fitting model with prey added. Both damsel bug and prey densities were ln-transformed.

36 Chapter 3

The predation rate and functional response of a generalist insect predator to leafhopper prey: the roles of prey preference and habitat complexity

In biological control programs, specialist parasitoids are often used to suppress outbreaks of the target pest species. Generalist arthropod predators are thought to be poor agents for biological control due to their broad diets and low numerical responses relative to their prey (Huffaker and Messenger 1976, Debach and Rosen 1991). For example, generalist predators will often engage in cannibalism or intraguild predation, which may actually release crop pests from predator control (Rosenheim et al. 1993, Snyder and Ives 2001, Snyder and Wise 2001). However, at least three studies demonstrated that generalists can be effective control agents in concert with specialist parasitoids (Cardinale et al. 2003, Snyder and Ives 2003) or with other generalist predators (Snyder and Wise 2001).

The prevalence of intraguild predation and cannibalism should be quantified as part of screening generalist predators for suppression of crop pests. Cannibalism rates may increase at high conspecific densities (Denno et al. 2004). However, greater habitat complexity, such as the addition of leaf litter, increases the predation rate on herbivores and decreases rates of intraguild predation (Finke and Denno 2002, 2006) and cannibalism (Rickers and Scheu 2005, Langellotto and Denno 2006). In the presence of herbivorous prey, generalist predators often will prefer herbivores rather than intraguild predators or conspecifics due to decreased risk of injury or death (Elgar and Crespi 1992, Lucas et al. 1998, Hodge 1999). While increased habitat complexity may decrease intraguild predation and cannibalism, it may also lead to increased prey persistence time due to the creation of a prey refuge (Murdoch et al. 1989, Messina et al. 1997, Magalhães et al. 2007). Therefore, it is necessary to determine the interaction of intraguild predation and habitat complexity in predator-prey relationships.

Functional response experiments are important in determining a predator’s impact on target prey populations and in evaluating if a predator can function as a natural enemy of the target pest species. I determined the predation rate of a common generalist heteropteran predator, the damsel bug ( Nabis spp., Heteroptera: Nabidae), on two

37 leafhopper pests of red clover and other legume forage plants. Both Agallia constricta and Ceratagallia agricola (Homoptera: Cicadellidae) are abundant in red clover. I hypothesized that as density of leafhoppers increased, the predation rate of the damsel bug would also increase until a saturation density was reached. Thus, I predicted to observe a Type II functional response for the damsel bug to both leafhopper species. I also determined if the damsel bug preferred to feed on one leafhopper species versus the other. There were no a priori predictions on leafhopper preferences by damsel bugs due to the scarcity of published literature on these two leafhopper species.

For the effect of habitat complexity, I hypothesized that increased habitat complexity would affect the predation rate of the damsel bug on leafhopper prey. Increased complexity was predicted to decrease the predation rate by providing a refuge for the leafhoppers from predation. I further predicted that increased habitat complexity would decrease cannibalism rates among damsel bugs due to decreased conspecific encounter rates.

Materials and Methods

Study Species

Insects were collected from clover, soybean, and goldenrod fields at the Miami University Ecology Research Center, Oxford, OH, USA during the summers of 2007 and 2008. The clover leafhopper ( Ceratagallia agricola (Hamilton), Homoptera: Cicadellidae) uses red clover as its preferred host plant (Black 1936, Watkins 1941, Hamilton 1998). C. agricola develops from nymph to adult in 18-43 days (Watkins 1941). Agallia constricta Van Duzee (Homoptera: Cicadellidae) is a generalist leafhopper known to consume legumes and grasses (LaHue 1936, Black 1944, Nielson 1968). The development time for this species is not known. A. constricta is most abundant in late June and July, while C. agricola is most abundant in August and September (Schroeder 2007). Damsel bugs ( Nabis spp.) are polyphagous predators that prey on multiple invertebrate taxa including leafhoppers (Lattin 1989, Östman and Ives 2003). Nabids are diurnal, folivorous predators that actively forage for prey using chemoreception and vibrations (Donahoe and Pitre 1977, Irwin and Shepard 1980,

38 Braman and Yeargan 1989, 1990, Schotzko and O’Keeffe 1989, Freund and Olmstead 2000). Development time from nymph to adult ranges from 24-31 days (Nadgauda and Pitre 1986). Nabids are common in many types of agricultural systems such as soybean, alfalfa, and lentils (Irwin and Shepard 1980, Braman and Yeargan 1989, 1990, Schotzko and O’Keeffe 1989). Two species of damsel bugs occurred at my study site, Nabis americoferus (Carayon) and N. roseipennis Reuter, though it was not possible to differentiate between species in the field. Since my experiments used nabids randomly collected from the field, I made the reasonable assumption that the confounding effects of species and sex are negligible (see Östman and Ives 2003).

Functional Response of the Damsel Bug and Leafhopper Survival

I documented the functional response of the damsel bug to the two species of leafhoppers in experimental mesocosms during Summer 2007. Trials were conducted outdoors at the Miami University Ecology Research Center. Medium red clover was grown from seed in 25 cm diameter and 12 cm deep pots. A single pot was placed in a 31 × 31 × 31 cm wire-frame cage covered by “no-see-um” netting (BioQuip Products, Rancho Dominguez, CA, USA). Pots were watered regularly. Prior to the experiment, cages were sprayed with pyrethrin insecticide to eliminate any arthropods present in the top soil or clover. Three days after spraying, damsel bugs and leafhoppers were introduced into mesocosms. Five densities of leafhoppers were used to determine the saturation point of feeding: 10, 20, 40, 60, or 80 adult leafhoppers were added to the enclosure cages. Five replicates were used per treatment. Adult damsel bugs were food- deprived for 2 d, and one damsel bug was added per cage 1 d after the leafhoppers. After 24 hrs, cages were sampled with a portable vacuum to determine the number of leafhoppers eaten. For A. constricta , surviving leafhoppers were returned to the cage from which they were sampled. Cages were resampled daily for an additional 4 d. It was not possible to conduct replacement sampling with C. agricola due to inclement weather.

39 Prey preference

To determine if the damsel bug fed preferentially on one species of leafhopper, I conducted a laboratory experiment in summer 2008 using cylindrical mesocosms covered with “no-see-um” netting and containing transplanted red clover. Mesocosms were 18 cm in diameter and 24 cm high. Total leafhopper densities were the same as in the functional response experiment (10, 20, 40, 60, 80), but a 50:50 ratio of A. constricta : C. agricola was used in each mesocosm to determine preference. Damsel bugs were food- deprived for 2 d, and added to the mesocosms 1 d after the leafhoppers. Five replicates were used at each density. Mesocosms were vacuum-sampled 1 d after the damsel bugs were added to determine the number of each species eaten.

Only female damsel bugs were used in this experiment because females are more voracious than males (Lingren et al. 1968, Donahoe and Pitre 1977, Propp 1982, Ma et al. 2005). This is likely due to the extra energy required for egg production which has been documented in both insects (Chapman 1971) and spiders (Walker and Rypstra 2001, 2002). It is also possible that male and female damsel bugs differ in their consumption rates to maximize fitness. Males maximize their fitness by consuming less than females in order to maximize the time spent searching for a mate, and females maximize their fitness by maximizing consumption to acquire energy needed for reproduction (Schoener 1971). This behavior has been observed in spiders (Givens 1978, Walker and Rypstra 2001) but is unknown in damsel bugs.

Cannibalism and habitat structure

Intraguild predation is common among arthropod predators even when alternative prey is present (Rosenheim et al. 1993, Rosenheim 2005). More complex habitat may provide prey a refuge from predators (Warfe and Barmuta 2004, Magalhães et al. 2007), whereas damsel bugs engage in cannibalism within minutes of being placed together in small empty containers (Stasek, personal observation). The effect of habitat complexity on cannibalism between damsel bug and their predation rates on C. agricola were tested in both simple and complex habitats using cylindrical mesocosms. There were 4 treatments with 5 replicates each. Two treatments were top soil only with either 1 or 2

40 damsel bugs, and the other two treatments were clover with either 1 or 2 damsel bugs. Damsel bugs were collected in the field, fed to satiation with C. agricola , and then deprived of food for 2 wks. Either 1 or 2 food-deprived damsel bugs and 50 C. agricola were added to each mesocosm, with the damsel bugs being added 24 hrs after C. agricola . After 24 hrs, each mesocosm was vacuum-sampled to determine the number of C. agricola and damsel bugs remaining.

Statistical Analysis

The shape of the functional response curve was determined using logistic regression (PROC CATMOD, SAS Institute 2003, Juliano 2001). One outlier at the 60- leafhopper density for A. constricta and one at the 20-leafhopper density for C. agricola were excluded from analyses. Parameters of the functional response model were then estimated using maximum likelihood (PROC MODEL, SAS Institute 2003).

For A. constricta , the role of predation on leafhopper survival was modeled using failure-time analysis (Fox 2001). In failure-time experiments, data are often not normally distributed using any standard transformation because experiments have a definitive start point before which no failures can occur. Predation rates on leafhoppers were also expected to vary during the trial with a high rate at the beginning and leveling off as damsel bugs became satiated. This analysis also allowed for the possibility that some leafhoppers may not be recovered alive or dead at a given sampling period (right- censored data). Leafhopper survivorship was calculated for each density using the estimated cumulative survival function (PROC LIFETEST, SAS Institute 2003). Differences in survivorship curves were analyzed using the Wilcoxon statistic and covariance matrix to calculate Z-statistics for a pairwise comparison of each density (Fox 2001).

41 Prey preference was determined using the α-vector preference metric:

− α = ln[(ni r i ) / n i ] m − ∑ln[(nj r j ) / n j ] i=1

i =1, 2, m=2 (Chesson 1983) (1)

where α is the proportion of species i in the damsel bug’s diet if both leafhopper species are equally present, ni = the initial number of individuals of leafhopper species i present,

ri = the number of leafhoppers of species i eaten by the damsel bug, nj= the initial number of leafhopper species j present, and rj is the number of leafhopper species j eaten. The difference between A. constricta and C. agricola preference metrics at each density were analyzed to determine if the damsel bug displayed a preference for a particular species of prey across prey densities using a multivariate analysis of variance (PROC GLM, SAS Institute 2003).

The effect of habitat structure on predation rate and cannibalism rates was assessed using linear regression (PROC GLM, SAS Institute 2003). The main effects of habitat complexity and damsel bug density were tested along with their interaction.

Results

For the leafhopper A. constricta , the damsel bug displayed a Type I functional response with no saturation in feeding rate observed within the range of densities used. A nearly constant proportion of prey was eaten at each density. The proportion consumed at the 10-leafhopper treatment was 0.78, while the proportion consumed at the 80- leafhopper treatment was 0.56 (Fig. 1A). From maximum likelihood, the capture efficiency was estimated to be 0.536, suggesting a potentially large effect of damsel-bug predation on A. constricta ’s population growth rate. For C. agricola , the damsel bug also showed a Type I response with no saturation in feeding rate. The capture efficiency for C. agricola was 0.262, indicating a low potential impact of the damsel bug on C. agricola population growth. The proportion of leafhoppers consumed was 0.26 at all densities, about half of the proportion of A. constricta that was consumed (Fig. 1B). The prey

42 preference experiment determined that the damsel bug preferred A. constricta at all densities (Wilks’ Lambda= 0.1230, df= 1, P<0.0001, Fig.2A and B). There was no effect of leafhopper density on the damsel bugs preference for A. constricta (Wilks’ Lambda= 0.8148, df= 1, P= 0.3679).

The survival probability of A. constricta varied across prey densities (Fig. 3). The 20-leafhopper density had a significantly higher survival probability than the 10- leafhopper density (Z= 3.08, df= 4, P=0.00103), 60-leafhopper density (Z= 2.54, df= 4, P= 0.00547), and the 80-leafhopper density treatments (Z= 3.03, df= 4, P= 0.00122). The 40-leafhopper treatment also had a lower survival probability than the 80-leafhopper treatment (Z= 1.88, df= 4, P= 0.0300).

Habitat complexity strongly influenced leafhopper predation by damsel bugs. More leafhoppers were eaten per day in the single damsel bug, bare ground treatment than in the single damsel bug, clover treatment (F=5.97, df=1,8, P=0.040; Fig. 4). A mean of 8.4 ±3.9(SD) and 17.9 ±7.7 C. agricola were eaten in the single damsel bug, clover treatment and single damsel bug, bare ground treatment, respectively. There was no difference in predation rate between the two damsel bug, clover and two damsel bug, bare ground treatments, though the data trended to more leafhoppers being consumed in the bare ground treatment (F=5.00, df=1,8, P=0.056; Fig. 4). A mean of 21.8 ±10.1 and 34.0±6.8 C. agricola were eaten in the clover and bare ground treatments, respectively, with 2 damsel bugs present. Thus, there was no interference between predators in per capita feeding rate between these treatments. No instances of cannibalism were observed in any of the trials.

Discussion

The observed functional response of the damsel bug to both leafhopper species was contrary to my predictions. While I predicted the damsel bug would display a Type II functional response, it displayed a Type I response to both species, indicating that a saturating density was not obtained (Fig. 1). Type I responses are not commonly reported in the literature; however, Type I responses were reported for spiders (Finke and Denno 2002, Denno et al. 2004) and hymenopteran parasitoids (Hoffmann et al. 2002, Luna et

43 al. 2007). Functional response experiments using nabids have reported only a Type II response (Propp 1982, Flinn et al. 1985, Ma et al. 2005).

The 60- and 80-density treatments may have been artificially high, possibly leading the damsel bugs to engage in superfluous killing, whereby prey are only partially consumed. This phenomenon is observed in a variety of taxa, including spiders (Maupin and Riechert 2001), zooplankton (Conover 1966) and damselfly naiads (Johnson et al. 1975). Different parts of the prey item may be of differing quality with the most nutritious parts eaten first (Cook and Cockrell 1978, Givens 1978, Formanowitz 1984). The viscosity of prey parts may also increase as feeding continues. Thus, it would be more difficult for predators to feed that engage in extra-intestinal digestion of their prey, which is the feeding method used by all heteropteran predators (Cook and Cockrell 1978, Pollard 1989, Cohen 2000). Since damsel bugs feed via extra-cellular digestion, superfluous killing may occur at high prey densities and only the most nutritious prey parts are consumed. This also explains why the long-term survival probabilities at the 60- and 80-leafhopper density treatments are lower than the 20-leafhopper density treatment (Fig. 3).

The damsel bug also showed preference for A. constricta when it was presented with both leafhopper species in a 50:50 ratio (Fig. 2). C. agricola may be able to detect an approaching damsel bug sooner than A. constricta and leap or fly more quickly to escape predation. The lower capture efficiency of the damsel bug for C. agricola indicates that this may be the case. C. agricola may also be harder for the damsel bug to manipulate once caught, or it may release or sequester chemicals for defense. Aphids will sequester or release defensive chemicals (Dixon 1985, Wink and Witte 1991), but this is not known for leafhoppers. A final possibility is that A. constricta may be more nutritious than C. agricola . Predators select prey to correct for nutrient deficiencies (Mayntz et al. 2005). Therefore, A. constricta may provide damsel bugs with nutrients that are deficient in other prey species. These questions are all avenues for future experiments.

The increased habitat complexity decreased the predation rate of the damsel bugs upon C. agricola . This decrease can likely be attributed to an increase in refuges for the

44 prey in the structurally-complex habitat (Murdoch et al. 1989, Messina et al. 1997, Magalhães et al. 2007). Since leafhoppers are sedentary unless disturbed, they may be less susceptible to damsel bugs, which actively move through the foliage in search of prey in the heterogeneous habitat (see Finke and Denno 2002). Though damsel bugs hunt primarily by using chemical cues (Freund and Olmstead 2000), the simple bare-ground habitat may allow damsel bugs to use both chemical and visual cues; therefore, leafhoppers may be more susceptible to predation in the bare-ground habitat. The lack of cannibalism was likely due to the availability of alternative prey (Lucas et al. 1998, Hodge 1999, Denno et al. 2004), as well as increased refugia for both leafhoppers and damsel bugs in the complex-habitat treatment. Increased habitat complexity decreased cannibalism and intraguild predation in other studies (Finke and Denno 2002, 2006, Langellotto and Denno 2006). In my experiment, clover likely decreased conspecific encounter rates between damsel bugs, and thus reduced cannibalism rates.

This study has broader applications to agriculture as red clover is often planted as a forage or cover crop, and both leafhopper species are common pests of red clover. Damsel bugs are common in many agroecosystems (Irwin and Shepard 1980, Lattin 1989), and thus may have the ability to function as natural enemies of leafhoppers, possibly in concert with other predatory arthropods such as jumping spiders, wolf spiders, web-building spiders, and carabid beetles. Functional response experiments are often the first step in determining if an organism is a candidate for biological control (Meyling et al. 2003). This study has demonstrated that damsel bugs are voracious predators of A. constricta and C. agricola , with a preference for A. constricta . However, damsel bugs have a lower numerical response than their leafhopper prey (Nadgauda and Pitre 1986), and damsel bugs will also feed on other arthropods in crop fields, such as aphids (Lattin 1989). These two factors may potentially hinder their use as a biological control agent for leafhoppers. Damsel bugs themselves will also be preyed upon by carabid beetles and wolf spiders (Snyder and Wise 2001), both of which are also common in red clover. If intraguild predation is common among predatory arthropods in red clover, it may dampen any trophic cascades that are present (Finke and Denno 2003, 2004, 2005). Future experiments are needed to address these questions.

45 References Cited

Black, L.M. 1936. Some insect and host relationships of the potato yellow dwarf virus. Phytopathology 26: 87.

------. 1944. Some viruses transmitted by Agallian leafhoppers. Proceedings of the American Philosophical Society 88: 132-144.

------. 1990. Phenology and abundance of Nabis americoferus , N. roseipennis , and N. rufusculus (Hemiptera: Nabidae) and their parasitoids in alfalfa and soybean. Journal of Economic Entomology 83: 823-830.

Chapman, R.F. 1971. The Insects Structure and Function. American Elsevier, NY, NY, USA.

Chesson, J. 1983. The estimation and analysis of preference and its relationship to foraging models. Ecology 64: 1297-1304.

Cohen, A. C. 2000. How carnivorous bugs feed. In C.W. Schaefer and A.R. Panizzi (eds). Heteroptera of Economic Importance. CRC Press, New York, NY, USA.

Conover, R.J. 1966. Factors affecting the assimilation of organic matter by zooplankton and the question of superfluous feeding. Limnology and Oceanography 11: 346- 354.

Cook, R.M. and Cockrell, B.J. 1978. Predator ingestion rate and its bearing on feeding time and the theory of optimal diets. Journal of Animal Ecology 47: 529-547.

Debach, P. and Rosen, D. 1991. Biological control by natural enemies. Cambridge University Press, New York, NY, USA.

Denno, R.F., Mitter, M.S., Langellotto, G.A., Gratton, C., and Finke, D.L. 2004. Interactions between a hunting spider and a web-builder: consequences of intraguild predation and cannibalism for prey suppression. Ecological Entomology 29: 566-577.

Dixon, A.F.G. 1985. Aphid Ecology. Blackie and Sons, London, UK.

46 Donahoe, M.C. and Pitre, H.N. 1977. Reduviolus roseipennis behavior and effectiveness in reducing numbers of Heliothis zea on cotton. Environmental Entomology 6: 872-876.

Elgar, M.A. and Crespi, B.J. 1992. Cannibalism: Ecology and Evolution among Diverse Taxa. Oxford University Press, Oxford, UK.

Finke, D.L. and Denno, R.F. 2002. Intraguild predation diminished in complex-structured vegetation: implications for prey suppression. Ecology 83: 643-652.

------. 2003. Intra-guild predation relaxes natural enemy impacts on herbivore populations. Ecological Entomology 28: 67-73.

------. 2004. Predator diversity dampens trophic cascades. Nature: 429: 407-410.

------. 2005. Predator diversity and the functioning of ecosystems: the role of intraguild predation in dampening trophic cascades. Ecology Letters 8: 1299-1306.

------. 2006. Spatial refuge from intraguild predation: implications for prey suppression and trophic cascades. Oecologia 149: 265-275.

Flinn, P.W., Hower, A.A., and Taylor, R.A.J. 1985. Preference of Reduviolus americoferus (Hemiptera: Nabidae) for potato leafhopper nymphs and pea aphids. Canadian Entomologist 117: 1503-1508.

Formanowicz, D.R., Jr. 1984. Foraging tactics of an aquatic insect: partial consumption of prey. Animal Behaviour 32: 774-781.

Fox, G. A. 2001. Failure-time analysis: emergence, flowering, survivorship, and other waiting times. pp. 235-266 in Scheiner, S.M. and Gurevitch, J. editors. Design and Analysis of Ecological Experiments 2 nd edition. Oxford University Press, New York, NY, USA.

Freund, R.L. and Olmstead, K.L. 2000. Role of vision and antennal olfaction in habitat and prey location by three predatory heteropterans. Environmental Entomology 29: 721-732.

Givens, R.P. 1978. Dimorphic foraging strategies of a salticid spider ( Phiddipus audax ). Ecology 59: 309-321.

Hamilton, K.G.A. 1998. The species of the North American leafhoppers Ceratagallia Kirkaldy and Aceratagallia Kirkaldy (Rhynchota: Homoptera: Cicadellidae). The Canadian Entomologist 130: 427-490.

Hodge, M.A. 1999. The implications of intraguild predation for the role of spiders in biological control. Journal of Arachnology 27: 351-362.

47 Hoffmann, M.P., Wright, M.G., Pitcher, S.A., and Gardner, J. 2002. Inoculative releases of Trichogramma ostriniae for suppression of Ostrinia nubilalis (European corn borer) in sweet corn: field biology and population dynamics. Biological Control 25: 249-258.

Huffaker, C.B. and Messenger, P.S. (editors). 1976. Theory and Practice of Biological Control. Academic Press, New York, NY, USA.

Johnson, D.M., Akre, B.G. and Crowley, P.H. 1975. Modeling arthropod predation: wasteful killing by damselfly naiads. Ecology 56: 1080-1093.

Juliano, S.A. 2001. Nonlinear curve fitting: predation and functional response curves. pp. 178-196 in Scheiner, S.M. and Gurevitch, J. editors. Design and Analysis of Ecological Experiments 2 nd edition. Oxford University Press, New York, NY, USA.

LaHue, D.W. 1936. An annotated list of the Bythoscopinae of Indiana (Cicadellidae, Homoptera). Proceedings of the Indiana Academy of Science 45: 310-314.

Langellotto, G.A. and Denno, R.F. 2006. Refuge from cannibalism in complex-structured habitats: implications for the accumulation of invertebrate predators. Ecological Entomology 31: 575-581.

Lattin, J.D. 1989. Bionomics of the Nabidae. Annual Review of Entomology 34: 383- 400.

Lingren, P.D., Ridgway, R.L., and Jones, S.L. 1968. Consumption by several common arthropod predators of eggs and larvae of two Heliothis species that attack cotton. Annals of the Entomological Society of America 61: 613-618.

Lucas, E., Coderre, D., and Brodeur, J. 1998. Intraguild predation among aphid predators: characteristics and influence of extraguild prey density. Ecology 79: 1084-1092.

Luna, M.G., Sanchez, N.E., Pereyra, P.C. 2007. Parasitism of Tuta absoluta (Lepidoptera, Gelechiidae) by Pseudapanteles dignus (Hymenoptera, Braconidae) under laboratory conditions. Environmental Entomology 36: 887-893

Ma, J., Li, Y.Z., Keller, M., and Ren, S.X. 2005. Functional response and predation of Nabis kinbergii (Hemiptera: Nabidae) to Plutella xylostella (Lepidoptera: Plutellidae). Insect Science 12: 281-286.

48 Magalhães, S., van Rijn, P.C.J., Montserrat, M., Pallini, A. and Sabelis, M.W. 2007. Population dynamics of thrips prey and their mite predators in a refuge. Oecologia 150: 557-568.

Maupin, J.L. and Riechert, S.E. 2001. Superfluous killing in spiders: a consequence of adaptation to food-limited environments? Behavioral Ecology 12: 569-576.

Mayntz, D., Raubenheimer, D., Salomon, M., Toft, S., and Simpson, S.J. 2005. Nutrient- specific foraging in invertebrate predators. Science 307: 111-113.

Messina, F.J., Jones, T.A. and Nielson, D.C. 1997. Host-plant effects on the efficacy of two predators attacking Russian wheat aphids (Homoptera: Aphidae). Environmental Entomology 26: 1398-1404.

Meyling, N.V., Enkegaard, A., and Brødsgaard, H. 2003. Two Anthocoris bugs as predators of glasshouse aphids- voracity and prey preference. Entomologia Experimentalis et Applicata 108: 59-70.

Murdoch, W.W., Luck, R.F, Walde, S.J., Reeve, J.D. and Yu, D.S. 1989. A refuge for red scale under control by Aphytis : structural aspects. Ecology 70: 1707-1714.

Nadgauda, D. and Pitre, H.N. 1986. Effects of temperature on feeding, development, fecundity, and longevity of Nabis roseipennis (Hemiptera: Nabidae) fed tobacco budworm (Lepidoptera: Noctuidae) larvae and tarnished plant bug (Hemiptera: Miridae) nymphs. Environmental Entomology 15: 536-539.

Nielson, M.W. 1968. The leafhopper vectors of phytopathogenic viruses (Homoptera, Cicadellidae) taxonomy, biology, and virus transmission. USDA Technical Bulletin 1382: 1-386.

Östman, Ö and Ives, A.R. 2003. Scale-dependent indirect interactions between two prey species through a shared predator. Oikos 102: 505-514.

Pollard, S.D. 1989. Constraints affecting partial prey consumption by a crab spider, Diea sp. indet. (Araneae, Thomisidae). Oecologia 81: 392-396.

Propp, G.D. 1982. Functional response of Nabis americoferus to two of its prey, Spodoptera exigua and Lygus hesperus . Environmental Entomology 11: 670-674.

Rickers, S. and Scheu, S. 2005. Cannibalism in Pardosa palustris (Araneae, Lycosidae): effects of alternative prey, habitat structure, and density. Basic and Applied Ecology 6: 471-478.

Rosenheim, J.A. 2005. Intraguild predation of Orius tristicolor by Geocoris spp. and the paradox of irruptive spider mite dynamics in California cotton. Biological Control 32: 172-179.

Rosenheim, J.A., Wilhoit, L.R., and Armer, C.A. 1993. Influence of intraguild predation among generalist insect predators on the suppression of an herbivore population. Oecologia 96: 439-449.

SAS Institute. 2003. SAS for Windows, Release 9.1. SAS Institute, Cary, North Carolina.

Schoener, T.W. 1971. Theory of feeding strategies. Annual Review of Ecology and Systematics 2: 369-404.

Schroeder, B. J. 2007. Effects of landscape structure on generalist and specialist insect herbivores. M.S. Thesis. Miami University.

Snyder, W.E. and Ives, A.R. 2001. Generalist predators disrupt biological control by a specialist parasitoid. Ecology 82: 705-716.

------. 2003. Interactions between specialist and generalist natural enemies: parasitoids, predators, and pea aphid biocontrol. Ecology 84: 91-107.

Snyder, W.E. and Wise, D.H. 2001. Contrasting trophic cascades generated by a community of generalist predators. Ecology 82: 1571-1583.

Walker, S.E. and Rypstra, A.L. 2001. Sexual dimorphism in functional response and trophic morphology in Rabidosa rabida (Araneae: Lycosidae). American Midland Naturalist 146: 161-170.

------. 2002. Sexual dimorphism in trophic morphology and feeding behavior of wolf spiders (Araneae: Lycosidae) as a result of differences in reproductive roles. Canadian Journal of Zoology 80: 679-688.

Warfe, D.M. and Barmuta, L.A. 2004. Habitat structural complexity mediates the foraging success of multiple predator species. Oecologia 141: 171-178.

Watkins, T.C. 1941. Clover leafhopper ( Aceratagallia sanguinolenta Prov.) New York Agricultural Experiment Station (Cornell) Bulletin 758: 3-24.

Wink, M. and Witte, L. 1991. Storage of quinolizidine alkaloids in Macrosiphum albifrons and Aphis genistae (Homoptera: Aphididae). Entomologia Generalis 15: 237-254.

50 A) 1 0.9 0.8 0.7 0.6 0.5

1 0.4 -

d d 0.3 0.2 Observed 0.1 Predicted 0 Agallia density B) 0.7

Proportion of Prey Eaten 0.6 0.5 0.4 0.3 0.2 0.1 0 0 20 40 60 80 Ceratagallia density

Figure 1. The proportion of prey eaten plotted against prey density for A) Agallia constricta and B) Ceratagallia agricola . The damsel bug displayed a Type I functional response to both species with a near- constant proportion of prey eaten at each density.

51 A) Damsel bug preference bug Damsel 0.0 0.2 0.4 0.6 0.8 1.0 5 10 20 30 40 Agallia density

B) Damsel bug preference Damsel 0.0 0.2 0.4 0.6 0.8 1.0 5 10 20 30 40 Ceratagallia density

Figure 2. Calculated preference of the damsel bug to A) A. constricta and B) C. agricola at five densities (n=5). Whiskers represent 25 th and 75 th percentiles, the dark line indicates the median, and the edges of the boxes represent the 10 th and 90 th deciles.

52 1 0.9

0.8 10 0.7 20 0.6 40 0.5 60 80 0.4 Survival Probability Survival 0.3 a 0.2 b b,d 0.1 b a,c 0 0 1 2 3 4 5 6 Time (d)

Figure 3. Survival probability of A. constricta at five densities with one damsel bug present over the course of 5 d. Treatments with a different letter are significantly different from each other (P= 0.05).

53 40 -1 30 eaten d 20

10 Ceratagallia #

0 Single bug, Single bug, 2 bugs, 2 bugs, clover bare clover bare ground ground Treatment

Figure 4. The effect of habitat complexity and damsel bug density on the predation rate of C. agricola . More leafhoppers were consumed in the single damsel bug, bare ground treatment than in the single damsel bug, clover treatment. Error bars are ±1 SD.

54 Chapter 4

The effects of dispersal and predator density on survival time of insects in an insect- red clover metacommunity

Trophic interactions of organisms are often studied in isolated communities (Paine 1966, Beckerman et al. 1997, Schmitz et al. 1997, Schmitz 2003), but it is now known that dispersal among communities is important to local dynamics. Metacommunity biology incorporates dispersal among local communities and how dispersal affects biodiversity and species interactions such as predation, parasitism, herbivory, and competition (Leibold et al. 2004, Holyoak et al. 2005). There is a well- developed body of theory for metacommunity biology, but there are few experimental tests that address how dispersal affects local and metacommunity processes.

Dispersal rates of individuals moving among local communities will determine the extent to which predator and prey dynamics are coupled in patchy landscapes. Dispersal of both predators and their prey may stabilize predator-prey dynamics (Holt 2002, Briggs and Hoopes 2004), but high levels of dispersal may uncouple these dynamics as a result of the predator over-exploiting its prey (Holyoak and Lawler 1996b ). At low dispersal rates, prey may escape in space from predators by colonizing empty habitat patches (Holyoak and Lawler 1996b). Thus, prey populations may experience a high growth rate due to a lack of predation. However, if predators are able to colonize patches with high prey abundance, there may be large-amplitude fluctuations in predator and prey abundance due to an increased numerical response by the predator and a subsequent population crash after prey have been over-exploited by the predator (Holyoak and Lawler 1996b, Kneitel and Miller 2003, Kondoh 2003) .

Intermediate dispersal rates are predicted to increase the persistence time of both predators and their prey due to a greater chance of vacant patches being recolonized than at low dispersal rates (Brown and Kodric-Brown 1977, Crowley 1981, Nachman 1987, Holyoak and Lawler 1996b), but dispersal rates are low enough that the predator cannot over-exploit its prey due to more limited movement among habitat patches than in a single, patchy community. Studies with protist and microbe communities have shown

55 that intermediate dispersal rates increase the persistence and local diversity of species in the presence of predation due to dispersal-limited species being able to colonize new habitat patches and rare species persisting due to either being superior colonizers or emigrating to empty patches (Brown and Kodric-Brown 1977, Holyoak and Lawler 1996a, 1996b, Loreau and Moquet 1999, Moquet and Loreau 2002, Kneitel and Miller 2003, Cadotte and Fukami 2005, Hauzy et al. 2007). These results have also been observed with trophically similar species (Cadotte 2006), while Forbes and Chase (2002) found no effect of dispersal.

Metacommunity persistence is difficult to study in larger organisms due to their longer generation times (but see Bonsall et al. 2002, 2005, Bull et al. 2006). However, the mechanistic basis for movement among habitat patches is often studied using larger organisms so that movement among patches and interspecific interactions can be quantified. This provides insights into mechanisms of metacommunity dynamics operating within generations. Short-term metacommunity dynamics that have been studied on individual observations include processes such as predation (Bonsall et al. 2002, 2005, Bull et al. 2006), herbivory (Matthiessen et al. 2007), and habitat selection (Resetarits 2005, Binckley and Resetarits 2007). The factors influencing dispersal among habitat patches is important to understand as dispersal among habitat patches is required for metacommunity persistence (Holyoak et al. 2005).

My aims in this study were to determine the effect of varying conspecific prey density, predator density, and isolation on the survival probability of the leafhopper Agallia constricta in experimental mesocosms. Density-dependent dispersal occurs in many taxa such as birds, mammals, and insects (Denno and Peterson 1995, Fonseca and Hart 1996, Matthysen 2005). Therefore, I wanted to determine if density-dependent dispersal was present in A. constricta in the absence of a predator. I hypothesized that leafhopper density and isolation would influence the rate of leafhopper dispersal among local communities. Specifically, I predicted greater dispersal among local communities in high-density, no isolation treatments compared to low-density and low-isolation treatments. I also predicted that leafhopper survival time would be greatest in the intermediate isolation treatment in accordance with metapopulation theory (Levins 1969, Hanski et al. 1994, Hill et al. 1996). I also determined the effect of varying isolation and

56 predator density on the survival probability of A. constricta . I predicted that leafhopper survival time would be greatest in the intermediate isolation/ low-predator density treatment.

Methods

Experimental Mesocosms

Cylindrical cages were constructed using “no-see-um netting” covering a wire frame (28 cm d × 40 cm h) and a pot of red clover (30 cm d x 10 cm h) grown from seed (Fig. 1). To remove any arthropods present on clover, pots were sprayed with pyrethrin insecticide before placing in cages. After 2 d, leafhoppers were introduced to cages. Each cage represented a local community of clover, leafhoppers, and predators, and three cages were linked by dispersal to create a metacommunity. Cages were connected using vinyl rain gutter with the sides removed and lined with “no-see-um” netting. Tubes were 10 cm × 50 cm (d × l), and dispersal was controlled by closing the ends of the tubes. There were three levels of isolation: a low level of isolation with the connecting tubes open 5% of the time per wk, or 8 hrs, during the experiment; an intermediate level of isolation with the tubes open to dispersal 30% per wk, or 48 hrs; and no isolation with the tubes open 100% of the time. These three levels of isolation were selected to simulate the dispersal rates of the endpoints of metapopulation dispersal and a single, patchy community. The 5%-dispersal treatment represented the low-end of metapopulation dispersal where each mesocosm would likely behave as an isolated local population. The 30%-dispersal treatment represented the high endpoint of metacommunity dynamics, and the 100%-dispersal treatment represented a single, large, patchy community (cf. Holyoak and Lawler 1996a). I used a random number generator to determine when each replicate would have its tubes open to movement each week that the experiment was conducted. In the experiment with only leafhoppers (see below), 2 replicates did not have their tubes opened due to the experiment being shortened by inclement weather.

57 Preliminary Experiment

To quantify leafhopper movement among mesocosms, leafhoppers were dusted with a different color of fluorescent powder. A preliminary experiment was conducted to determine if damsel bugs prefer to prey on a particular color of powder. I predicted that there would not be a color preference since damsel bugs are known to hunt primarily by chemoreception and vibrations (Freund and Olmstead 2000).

Using mesocosms made from plastic bowls with transplanted red clover (see Chapter 3), 25 A. constricta were dusted with red, blue, or yellow fluorescent powder and added to each mesocosm. A control with no powder was also used. Five replicates were used for each treatment. Damsel bug were food-deprived for 2 d before being added to the mesocosms 1 d after the leafhoppers were added. Mesocosms were vacuum-sampled with a modified Dustbuster® 1 d after damsel bugs were added, and the number of surviving leafhoppers was counted. Results were analyzed using an ANOVA (PROC GLM, SAS Institute 2003). There was no difference in the number of leafhoppers consumed amongst the color treatments and the control (F= 0.820, df= 3, 14, P= 0.506).

Dispersal Experiments

To determine the effect of dispersal on the survival probability of A. constricta in the absence of predation, three dispersal treatments (5%, 30%, 100%) were crossed with two different densities of leafhoppers per mesocosm (25 or 50). Four replicate metacommunities of each treatment were used for a total of 72 communities and 24 metacommunities. Only two mesocosms were stocked with leafhoppers to determine if leafhoppers would colonize an empty mesocosm if local densities were too high. Leafhoppers were dusted with either red or yellow fluorescent powder to quantify movements among communities. Each community was sampled with a suction-sampler 1 d after damsel bugs were introduced and every 2 d thereafter. Surviving leafhoppers were counted and returned to the community where they were sampled. The experiment was only conducted for 5 d due to inclement weather caused by the remnants of Hurricane Ike moving through the Oxford area. A 50-leafhopper, 5%-dispersal replicate and a 50-leafhopper, 30%-dispersal replicate did not have their tubes opened as a result

58 of the experiment being terminated earlier than expected, though I do not believe this influenced my results. I also tested the effect of dispersal on A. constricta survival probabilities with varying the density of predators per community. For this experiment, I used a fixed density of 50 leafhoppers per local community, and I replicated the dispersal levels as in the previous experiment. One mesocosm was intentionally left empty to serve as a refuge from predation. Leafhoppers were dusted with either blue or red fluorescent powder to quantify movements among cages. Additionally, I added either one or two damsel bugs to each community 1 d after leafhoppers were added for a total of either two or four per metacommunity. I used only female damsel bugs in this experiment. Female damsel bugs are more voracious than males (Lingren et al. 1968, Donahoe and Pitre 1977, Propp 1982, Ma et al. 2005), and they may also be energy maximizers that will consume as much prey as possible to acquire the energy needed for reproduction. In contrast, males may spend little time foraging in order to find a mate as quickly as possible (Schoener 1971). Damsel bugs can be sexed by the presence of external claspers on the males (Irwin and Shepard 1980). Each community was sampled with a suction-sampler 1 d after damsel bugs were introduced and every 2 d thereafter. Surviving leafhoppers and damsel bugs were returned to the community from which they were sampled. If a dead damsel bug was found during sampling, it was replaced to maintain predator density throughout the course of the experiment. The experiment was run for 7 d. All experiments were conducted June-September 2008 at the Miami University Ecology Research Center in Oxford, OH.

Statistical Analyses

The role of predation in leafhopper survival and persistence were modeled using failure-time analysis (Fox 2001). Failure-time analysis does not assume a normal distribution of time until death and also allows for the possibility that some leafhoppers may not be recovered alive or dead during a sampling period (right-censored data). Leafhopper survival under different treatment factors of predation (0, 1, or 2 damsel bugs) and dispersal rate (5%, 30%, 100%) were recorded every 2 d for 5 time intervals so

59 that time to mortality can be treated as a distribution of failure times. I expected that leafhopper mortality would not occur at a constant rate as food-deprived damsel bugs were expected to have high feeding rates at the beginning of the experiment and then level off as they became satiated. The predictor variables of predation and dispersal will influence these rates, a pattern that is suited to using non-parametric life-table analysis (PROC LIFETEST, SAS Institute 2003). Linear regression was used to determine if there was an effect of conspecific density, damsel bug density, or dispersal treatment on leafhopper dispersal rates (PROC GLM, SAS Institute 2003).

Results

In the absence of predation, <5% of the leafhoppers moved among habitat patches in all treatments except the 25 leafhoppers/100% dispersal treatment. In this treatment, 18 of 200 leafhoppers (9%) moved among local communities, while in other treatments, between 2 and 16 leafhoppers moved among local communities. Leafhoppers moved among local communities significantly more in the 100% dispersal treatments (F= 38.44, df= 1, P<0.0001), but there was no effect of density on the dispersal rate of leafhoppers (F= 1.0, df= 1, P= 0.3293). Therefore, density-dependent dispersal of leafhoppers among local communities did not occur. There was no difference among treatments in survival time of A. constricta in the absence of predation ( χ²= 1.8637, df= 5, P= 0.8677, Table 1, Fig. 2).

When damsel bugs were present, <5% of leafhoppers moved among local communities in all treatments. I only observed 3 instances of damsel bugs moving among communities. Between 1 and 12 leafhoppers moved among local communities in all treatments. There was no effect of damsel bug density on leafhopper dispersal rates (F= 2.55, df= 1, P= 0.1258), but leafhoppers did disperse among local communities more frequently in the 100% dispersal treatments (F= 8.58, df= 2, P= 0.0020). A. constricta had a higher survival probability in the 1-damsel bug/5%-dispersal treatment compared to the 1-damsel bug/100%-dispersal treatment ( χ²= 8.8613, df= 1, P= 0.0029, Table 2, Fig. 3). Leafhoppers also had higher survival probabilities in the 1-damsel bug/5%-dispersl treatment than in the 2-damsel bug/30% dispersal treatment ( χ²= 9.4088, df= 1, P=

60 0.0022, Table 2, Fig. 3) and the 2-damsel bug/100%-dispersal treatment ( χ²= 4.7514, df= 1, P= 0.0293, Table 2, Fig. 3).

Discussion

My hypothesis that leafhopper density and dispersal rate would affect survival time in the absence of predation was not supported. There was no difference in leafhopper survival times among density and dispersal treatments. Leafhoppers also moved infrequently among local communities with <5% of the total population moving among local communities in all but the 25-leafhopper/100%-dispersal treatment. This suggests that each local population was behaving as a single, isolated population except the 25-leafhopper/100%-dispersal treatment, which behaved as a classic metapopulation since 9% of the total population moved among local communities (Levins 1969, Hanski et al. 1994, Hill et al. 1996, Harrison and Taylor 1997, Stasek et al. 2008). Dispersal rates were determined from short-term experimental trials, however, and rates expressed per generation could be 2-3x times larger.

Despite high conspecific densities and the opportunity to disperse among mesocosms, leafhoppers rarely moved among local communities. Densities of A. constricta can be as high as 108 leafhoppers per m 2 in the field with a mean of 35 leafhoppers per m 2 (Schroeder 2007). This is equivalent to a mean of 2.56 leafhoppers per mesocosm, which have an area of 0.073 m 2. Therefore, densities in the mesocosms were 10-20x greater than the observed field densities for the 25- and 50-leafhopper density treatments, respectively. A meta-analysis by Denno and Peterson (1995) determined that declining host-plant quality is the main factor influencing emigration in sap-feeding insects. All local habitat patches in my mesocosms were of similar quality. There was also no evidence of “hopper burn,” which is a yellowing of plant leaves resulting from saliva injected into the plant stem while the leafhopper feeds. Hopper burn results in stunted growth, delayed maturation, and loss of yield and is commonly observed in agricultural fields infested with some species, such as the potato leafhopper (Kindler et al. 1973, Wilson et al. 1979). It is not known if A. constricta causes hopper burn, but Haynes and Crist (in press) suggested that potato leafhoppers may be more

61 associated with reductions in plant biomass than other herbivores. Therefore, if patch quality was equal to A. constricta , it is not surprising that movement among mesocosms was infrequent. If A. constricta did damage plants to the extent of potato leafhoppers, then movement among habitat patches may have been be more frequent. However, due to inclement weather, I was only able to observe leafhoppers for 5 d.

My hypothesis that damsel bug density and dispersal rate would affect leafhopper survival time was partially supported. I predicted that A. constricta would have the highest survival probability in the single damsel bug, intermediate dispersal treatment. A. constricta had a higher survival probability in the 1-damsel bug/ 5%-dispersal treatment compared to the 1-damsel bug/ 100%-dispersal treatment, the 2-damsel bug/ 100%- dispersal treatment, and the 2-damsel bug/ 30%-dispersal treatment. Previous studies show that predator or parasitoid density does (Hauzy et al. 2007) and does not (French and Travis 2001) influence prey dispersal. In my experiment, <5% of each local leafhopper population dispersed among mesocosms. This indicates that each local leafhopper population behaved as an isolated population (Levins 1969, Harrison and Taylor 1997), and damsel-bug density did not influence leafhopper dispersal rates (French and Travis 2001). Low dispersal in cages may have occurred because leafhoppers had limited opportunities to jump or fly to escape from predation. While movement between mesocosms was observed, dispersal tubes between the sides of mesocosms required leafhoppers walk between mesocosms; if tubes had connected the tops of mesocosms, flying or hopping movements may have been facilitated between cages.

Leafhoppers in the 100%-dispersal treatment were predicted to have lower survival probabilities than the 5%- and 30%-dispersal treatments. I expected the damsel bug to move freely among communities and over-exploit their leafhopper prey. I observed three instances of damsel bugs moving among cages in the 100%-dispersal treatment. Since damsel bugs actively forage for prey, they likely moved freely among local communities and over-exploited their leafhopper prey, in accordance with my predictions and those of metacommunity theory (Holyoak and Lawler 1996b, Freund and Olmstead 2000). It was also predicted that local communities with two damsel bugs would have lower survival probabilities than local communities with only one damsel

62 bug. The combined effect of two damsel bugs preying on A. constricta resulted in slightly higher predation rates compared to the single damsel bug treatment (Table 2). As a result, leafhoppers may have been unable to escape predation in all treatments, resulting in a similar survival probability among dispersal treatments.

There was no difference between the 5%- and 30%-dispersal treatments on leafhopper survival time in the single damsel bug treatment, contrary to my predictions. I predicted that the 30%-dispersal treatment would have the greatest survival time because a greater proportion of leafhoppers would disperse to other local communities to escape predation than in the 5%-dispersal treatment, and damsel bugs would not be able to follow leafhoppers once the tubes were closed (Brown and Kodric-Brown 1977, Crowley 1981, Nachman 1987, Holyoak and Lawler 1996b).

Most experimental metacommunity studies use protists to assess persistence due to their short generation times (Holyoak and Lawler 1996a, 1996b, Forbes and Chase 2002, Kneitel and Miller 2003, Cadotte and Fukami 2005, Cadotte 2006, Hauzy et al. 2007). However, it is important to understand the short-term behaviors that influence dispersal among local communities, such as predation risk (Resetarits 2005), habitat quality (Binckley and Resetarits 2007), patch arrangement (Bull et al. 2006), number of habitat patches (Bonsall et al. 2002, 2005), and connectivity (Matthiessen et al. 2007). My results demonstrated that dispersal and conspecific density had no effect on leafhopper survival in the absence of predation, while damsel bug density and dispersal did affect the survival probability of leafhoppers. This suggests that predator density and dispersal are key to understanding short-term persistence in predator-prey metacommunities; longer-term experiments are needed to determine how predator numerical response and dispersal vary in response to prey patches.

While damsel bugs preyed heavily upon leafhoppers within the first 24 hrs, there was limited movement by leafhoppers among mesocosms afterwards. Leafhoppers are sedentary animals (Östman and Ives 2003), and decreased movement may make them less susceptible to predation, which is observed in planthoppers exposed to predation by spiders (Finke and Denno 2002, 2006). Yet despite remaining sedentary, the predation rate by damsel bugs was still high in all dispersal and density treatments. Damsel bugs

63 likely responded to olfactory or chemical cues given off by the leafhoppers as they fed on plant stems, which is the main method by which damsel bugs hunt (Freund and Olmstead 2000).

This study is one of the first experimental tests of the effect of varying dispersal rates on prey survival. In the absence if predation, A. constricta dispersal rates were low; therefore, most A. constricta populations behaved as isolated populations, and there was no difference in the survival probability of A. constricta in response to varying dispersal treatments and leafhopper density. When damsel bugs were present, A. constricta dispersal rates among local communities were also low, leading to mesocosms behaving as individual communities. The 100%-dispersal treatment had a lower survival probability than the 5%-dispersal treatment when both one and two damsel bugs were present per mesocosm in accordance with metcommunity theory.

Future experiments should focus on varying both predator and prey densities as well as dispersal rates to determine the survival probability of leafhoppers. Variation in intra- and interspecific dispersal rates has only been conducted in a few studies (Bernstein 1984, French and Travis 2001, Hauzy et al. 2007). A combination of A. constricta and other common prey species such as the clover leafhopper and pea and spotted alfalfa aphids should be used to determine if damsel bugs can possibly be used as a biological control agent for A. constricta in the presence of these other pest species in forage crops.

While generalist predators are thought to be poor biological control agents, there is growing evidence that a suite of generalist predators is effective as biological control agents (Cardinale et al. 2003, Snyder and Ives 2003). Wolf spiders (Araneae: Lycosidae) are common generalist predators in many forage crop systems and were abundant at my study site. It should be determined if wolf spiders and other generalist predators such as carabid beetles have an additive, synergistic, or antagonistic effect on the ability of the damsel bug to suppress A. constricta populations when isoaltion and predator density are varied.

64 References Cited

Bernstein, C. 1984. Prey and predator emigration responses in the acarine system Tetranychus urticae- Phytoseiulus persimilis . Oecologia 61: 134-142.

Bonsall, M.B., French, D.R., and Hassell, M.P. 2002. Metapopulation structures affect persistence of predator-prey interactions. Journal if Animal Ecology 71: 1075- 1084.

Briggs, C.J. and Hoopes, M.F. 2004. Stabilizing effects in spatial parasitoid-host and predator-prey models: a review. Theoretical Population Biology 65: 299-315.

Brown, J.H. and Kodric-Brown, A. 1977. Turnover rates in insular biogeography: effect of immigration on extinction. Ecology 58: 445-449.

Cadotte, M.W. 2006. Metacommunity influences on community richness at multiple spatial scales: a microcosm experiment. Ecology 87: 1008-1016.

Cadotte, M.W. and Fukami, T. 2005. Dispersal, spatial scale, and species diversity in a hierarchically structured experimental landscape. Ecology Letters 8: 548-557.

Crowley, P.H. 1981. Dispersal and the stability of predator-prey interactions. American Naturalist 118: 673-701.

65 Denno, R.F. and Peterson, M.A. 1995. Density-dependent dispersal and its consequences for population dynamics. pp. 113-130 in Cappuccino, N. and Price, P.W. ,editors. Population dynamics: new approaches and synthesis. Academic Press, New York, NY, USA.

Donahoe, M.C. and Pitre, H.N. 1977. Reduviolus roseipennis behavior and effectiveness in reducing numbers of Heliothis zea on cotton. Environmental Entomology 6: 872-876.

Finke, D.L. and Denno, R.F. 2002. Intraguild predation diminished in complex-structured vegetation: implications for prey suppression. Ecology 83: 643-652.

------. 2006. Spatial refuge from intraguild predation: implications for prey suppression and trophic cascades. Oecologia 149: 265-275.

Fonseca, D.M. and Hart, D.D. 1996. Density-dependent dispersal of black fly neonates is mediated by flow. Oikos 75:49-58.

Forbes, A.E. and Chase, J.M. 2002. The role of habitat connectivity and landscape geometry in experimental zooplankton metacommunities. Oikos 96: 433-440.

Fox, G. A. 2001. Failure-time analysis: emergence, flowering, survivorship, and other waiting times. pp. 235-266 in Scheiner, S.M. and Gurevitch, J., editors. Design and Analysis of Ecological Experiments 2 nd edition. Oxford University Press, New York, NY, USA.

French, D.R. and Travis, J.M.J. 2001. Density-dependent dispersal in host-parasitoid assemblages. Oikos 95: 125-135.

Freund, R.L. and Olmstead, K.L. 2000. Role of vision and antennal olfaction in habitat and prey location by three predatory heteropterans. Environmental Entomology 29: 721-732.

Hanski, I., Kuussaari, M., and Nieminen, M. 1994. Metapopulation structure and migration in the butterfly Melitaea cinxia . Ecology 75: 747-762.

Hauzy, C., Hulot, F.D., and Loreau, M. 2007. Intra- and interspecific density-dependent dispersal in an aquatic prey-predator system. Journal of Animal Ecology 76: 552- 558.

Haynes, K.J. and Crist, T.O. 2009. Insect herbivory in an experimental agroecosystem: the relative importance of habitat area, fragmentation, and the matrix. Oikos In press .

66 Hill, J.K., Thomas, C.D. and Lewis, O.T. 1996. Effects of habitat patch size and isolation on dispersal by Hesperia comma butterflies: implications for metapopulation structure. Journal of Animal Ecology 65: 725-735.

Holt, R.D. 2002. Food webs in space: On the interplay of dynamic instability and spatial processes. Ecological Research 17: 261-273.

Holyoak, M. and Lawler, S.P. 1996a. Persistence of an extinction-prone predator-prey interaction through metapopulation dynamics. Ecology 77: 1867-1879.

Holyoak, M. and Lawler, S.P. 1996b. The role of dispersal in predator-prey metapopulation dynamics. Journal of Animal Ecology 65: 640-652.

Kindler, S.D., Kehr, W.R., Ogden, R.L., and Schalk, J.M. 1973. Effect of potato leafhopper injury on yield and quality of resistant and susceptible alfalfa clones. Journal of Economic Entomology 66: 1298-1302.

Kneitel, J.M. and Miller, T.E. 2003. Dispersal rates affect species composition in metacommunities of Serracenia purpurea inquilines. American Naturalist 162: 165-171.

Kondoh, M. 2003. Habitat fragmentation resulting in overgrazing by herbivores. Journal of Theoretical Biology 225: 453-460.

Levins, R. 1969. Some demographic and genetic consequences of environmental heterogeneity for biological control. Bulletin of the Entomological Society of America 15: 237-240.

67 Loreau, M. and Mouquet, N. 1999. Immigration and the maintenance of local species diversity. American Naturalist 154: 427-440.

Matthiessen, B., Gamfeldt, L., Jonsson, P.R., and Hillebrand, H. 2007. Effects of grazer richness and composition on algal biomass in a closed and open marine system. Ecology 88: 178-187.

Matthysen, E. 2005. Density-dependent dispersal in birds and mammals. Ecography 28: 403-416.

Mouquet, N. and Loreau, M. 2002. Coexistence in metacommunities: the regional similarity hypothesis. American Naturalist 159: 420-426.