The Pennsylvania State University

The Graduate School

Department of Entomology

THE INFLUENCE OF FARMING PRACTICES ON SLUGS AND THEIR

PREDATORS IN REDUCED-TILLAGE FIELD CROPS IN PENNSYLVANIA

A Thesis in

Entomology

by

Margaret Rose Douglas

2012 Margaret R. Douglas

Submitted in Partial Fulfillment of the Requirements for the Degree of

Master of Science

August 2012

ii

The thesis of Margaret Rose Douglas was reviewed and approved* by the following:

John F. Tooker Assistant Professor of Entomology and Extension Specialist Thesis Advisor

Mary E. Barbercheck Professor of Entomology

Heather D. Karsten Associate Professor of Crop Production/Ecology

Gary W. Felton Professor of Entomology Head of the Department of Entomology

*Signatures are on file in the Graduate School

iii

ABSTRACT

Slugs are a destructive pest of myriad field crops in reduced-tillage settings, particularly in the mid-Atlantic and Northeastern United States. Current management options are limited and farmers have expressed a need for improved slug control; however, basic aspects of slug ecology in the region remain obscure. My thesis begins with a review of existing knowledge on slug ecology, natural history, scouting, and management. Through this review, I identified research areas in need of attention for improving integrated management of slugs. In the remaining chapters, I attempt to address several of these needs through laboratory and field experiments. In preliminary laboratory experiments, I explored the potential for common generalist predators to contribute to slug suppression in Pennsylvania. Wolf spiders and cantharid larvae were disinclined to eat slugs; however, two common ground beetle species, Chlaenius tricolor and Pterostichus melanarius, preyed on slugs and protected soybean seedlings from damage, indicating a potentially important role for them in biological control. Next, I examined how crop management decisions influence slugs, insect pests, and predators in reduced-tillage maize systems, in field studies spanning two years. Weed and manure management had few effects on pest activity, crop damage, or predator activity in either year, although cultivation reduced late-season slug activity in 2011. Low-external-input crop rotations using cover crops were somewhat more vulnerable to slugs and European corn borer compared to a control rotation using pre-emptive insect management; however, damage from these pests was likely sub-economic in all rotations. Furthermore, in the second year of the study, the low- external-input rotations supported higher levels of predation on sentinel caterpillars, and comparable or lower levels of slug activity late in the season compared to the higher-input control rotation. Overall the low-external-input rotations were competitive with the more conventional, pre-emptive approach. Finally, I examined the influence of the common seed treatment, thiamethoxam, on slug damage to soybeans in a tritrophic context. In laboratory experiments, thiamethoxam did not significantly influence slug survival or feeding behavior on soybean. However, slugs that were fed for one week on thiamethoxam-infused soybean seedlings were poisonous to the ground beetle, Chlaenius tricolor. My results suggest that neonicotinoid seed treatments do not prevent slug damage to soybean, and in fact may exacerbate slug damage if slugs pass these insecticides to their predators. This possibility bears further evaluation in field studies.

iv

TABLE OF CONTENTS

LIST OF FIGURES ...... vi LIST OF TABLES ...... ix PREFACE ...... xi ACKNOWLEDGEMENTS ...... xii

Introduction ...... 1

Thesis objectives ...... 2 References ...... 4

Chapter 1 Slug (Mollusca: Agriolimacidae, Arionidae) ecology and management in no- till field crops, with an emphasis on the mid-Atlantic region ...... 7

Description of species and life cycles ...... 8 Host-plant species and damage ...... 11 Scouting ...... 14 Environmental influences...... 15 Natural enemies and biological control ...... 17 Management options ...... 19 References ...... 24 Figures ...... 31

Chapter 2 A preliminary laboratory study of slug predation by common generalist predators found in Pennsylvania crop fields ...... 36

Introduction ...... 36 Materials and methods ...... 37 Soybean seeds, slugs, and predators ...... 37 Slug predation experiment ...... 37 Chlaenius tricolor choice study ...... 38 Statistical analyses...... 38 Results ...... 39 Discussion ...... 40 Conclusion ...... 41 References ...... 42 Tables ...... 45 Figures ...... 46

Chapter 3 Dynamics of pest and predatory invertebrates in reduced-tillage maize (Zea mays L.) cropping systems in Pennsylvania ...... 48

Introduction ...... 48 Materials and methods ...... 52 Study site and crop management ...... 52 Stand establishment and early season herbivory ...... 53

v Slug activity-density ...... 53 European corn borer activity ...... 54 Predatory arthropod activity-density ...... 54 Predation on sentinel caterpillars ...... 55 Statistical analyses...... 55 Results ...... 57 Stand establishment and early season herbivory ...... 57 Slug activity-density ...... 58 European corn borer damage ...... 59 Predatory arthropod activity-density ...... 60 Predation on sentinel caterpillars ...... 60 Discussion ...... 61 Conclusion ...... 66 References ...... 67 Tables ...... 75 Figures ...... 84

Chapter 4 A laboratory-based assessment of the influence of thiamethoxam seed treatment on interactions among soybeans (Glycine max), slugs (Deroceras reticulatum) and ground beetles (Chlaenius tricolor) ...... 91

Introduction ...... 91 Materials and methods ...... 93 Seeds, slugs, and beetles ...... 93 Soybean-slug interactions ...... 94 Slug-ground beetle interactions ...... 95 Statistical analyses...... 96 Results ...... 98 Soybean-slug experiments ...... 98 Slug-ground beetle experiments ...... 99 Discussion ...... 100 Conclusion ...... 104 References ...... 105 Tables ...... 113 Figures ...... 115

Appendix A Supplemental data on the natural history of slugs in Pennsylvania field crops ...... 119

Figures ...... 120

Appendix B Supplemental tables describing statistical methods in Chapter 3 ...... 123

Appendix C Supplemental data on the influence of slug damage on grain yield in maize ..... 125

Reference ...... 126 Figure ...... 127

vi

LIST OF FIGURES

Figure 1-1. Gray garden slug (Deroceras reticulatum) (photo: Margaret Douglas, PSU) ..... 31

Figure 1-2. Marsh slug (Deroceras laeve) (photo: Margaret Douglas, PSU) ...... 31

Figure 1-3. Dusky slug (Arion subfuscus) (photo: courtesy of Nick Sloff, PSU) ...... 32

Figure 1-4. Banded slug (Arion fasciatus) (photo: courtesy of Nick Sloff, PSU) ...... 32

Figure 1-5. Slug eggs in the soil (photo: courtesy of Nick Sloff, PSU) ...... 33

Figure 1-6. Slug damage to corn (photo: Margaret Douglas, PSU) ...... 33

Figure 1-7. Juvenile grey garden slug (Deroceras reticulatum) on soybean (photo: courtesy of Nick Sloff, PSU) ...... 34

Figure 1-8. Slug damage to canola (photo: Margaret Douglas, PSU) ...... 34

Figure 1-9. Roofing material used as an artificial slug shelter (photo: Margaret Douglas, PSU) ...... 35

Figure 1-10. Chlaenius tricolor, a slug-eating beetle (photo: courtesy of Ian Grettenberger, PSU) ...... 35

Figure 2-1. Number of plants (out of four) damaged by slugs (mean ± SE) after four days of exposure to slugs and each of four predator species and a no-predator control (n = 9 - 19). These data include pooled data from both D. reticulatum and D. laeve. Bars that do not share a letter differ from each other at P ≤ 0.05 based on Tukey’s test...... 46

Figure 2-2. Slug survival (proportion, pooled across slug species) following a four-day assay with different predator species (n = 9 - 19). Bars that do not share a letter differ from each other at P ≤ 0.05 based on Fisher’s Exact Tests with a Bonferroni correction...... 46

Figure 2-3. Relationship between slug starting mass and log-transformed ending mass for slugs surviving four days in the presence or absence of predators, pooled by slug species (n = 18 for no predator control; n = 9 for cantharid larva; n = 12 for Trochosa; n = 5 for Pterostichus). Only one slug survived in the presence of Chlaenius and so it was not included here. Treatments followed by different letters in the key differ from each other at P ≤ 0.05 based on Tukey’s test...... 47

Figure 2-4. Survival of D. laeve, D. reticulatum, and A. fasciatus in choice arenas with C. tricolor (n = 6 for D. laeve vs. D. reticulatum and n = 5 for D. laeve vs. A. fasciatus)...... 47

vii Figure 3-1. Schematic of maize comparisons in the Sustainable Dairy Cropping Systems project. Cover crop treatments were in place in 2011 but not 2010...... 84

Figure 3-2. Slug activity-density through time by rotation (mean ± SE, n = 4). Values were square root transformed for analysis but here I show the untransformed values. The vertical line in September marks maize silage harvest. In 2010, management practices in spring precluded slug sampling for most of the early season, hence the few data points in spring 2010. Asterisks indicate dates on which there was a significant difference between rotations at P ≤ 0.05 based on ESTIMATE comparisons within dates...... 85

Figure 3-3. Slug activity-density (means ± SE, n = 4) after maize silage harvest in 2011 by rotation and split-plot treatment. Values were square root transformed for analysis but here I show the untransformed values. Bars with different letters are significantly different based on Tukey’s post-hoc comparisons at P ≤ 0.05. C-BM: Control rotation, broadcast manure; C-IM: Control rotation, injected manure; G-RH: Grain rotation, reduced herbicide; G-SH: Grain rotation, standard herbicide...... 86

Figure 3-4. Correlations between number of slugs per trap during spring and damage to corn seedlings at V2 and V5 (n = 12 plots). Slug activity-density was square root transformed for analysis but here I show the untransformed values...... 87

Figure 3-5. Seasonal patterns of activity-density of predatory arthropods captured in pitfall traps in 2010 and 2011, pooled across treatments (mean ± SE, n = 12 plots)...... 88

Figure 3-6. Seasonal patterns of activity-density of predatory arthropods captured in pitfall traps in 2010 and 2011, pooled across treatments (mean ± SE, n = 12 plots)...... 89

Figure 3-7. Predation on sentinel caterpillars in June over 12 hours as a function of time of day, caging treatment, and rotation...... 90

Figure 3-8. Predation on sentinel caterpillars in July over 12 hours as a function of time of day, caging treatment, and rotation...... 90

Figure 4-1. Mean numbers (A) and proportion (B) of slugs surviving by seed treatment: U = Untreated; F = Fungicide only; F+L = Fungicide + low rate thiamethoxam; F+H = Fungicide + high rate thiamethoxam. Data were collected in (A) spring, with 4 small slugs per container (n = 8) and, (B) fall, with one medium slug per container (n = 34). Error bars show one standard error...... 115

Figure 4-2. Slug mass gain (mean % ± SE) by seed treatment: U = Untreated; F = Fungicide only; F+L = Fungicide + low rate thiamethoxam; F+H = Fungicide + high rate thiamethoxam. Data were collected in (A) spring, with 4 small slugs per container (n = 8) and, (B) fall, with one medium slug per container (n = 29-32)...... 115

Figure 4-3. Number of soybean plants (out of 4) damaged by slugs (mean ± SE) as a function of time and seed treatment: U = Untreated; F = Fungicide only; F+L = Fungicide + low rate thiamethoxam; F+H = Fungicide + high rate thiamethoxam. Data were collected in (A) spring, with 4 small slugs per container (n = 8) and, (B) fall, with one medium slug per container (n = 29-32)...... 116

viii Figure 4-4. Soybean biomass (mean ± SE) at the end of seven days, by slugs and seed treatments: U = Untreated; F = Fungicide only; F+L = Fungicide + low rate thiamethoxam; F+H = Fungicide + high rate thiamethoxam. Data were collected in (A) spring, with four small slugs per container (n = 8) and, (B) fall, with one medium slug per container (n = 24 for no slug controls; n = 29-32 for slug treatments). Means that do not share a letter are different at P ≤ 0.05...... 116

Figure 4-5. Proportion of slugs surviving in containers with C. tricolor, as a function of time and slug treatment: U = Untreated (n = 19); F = Fungicide only (n = 17); F+L = Fungicide + low rate thiamethoxam (n = 18); F+H = Fungicide + high rate thiamethoxam (n = 18)...... 117

Figure 4-6. Symptoms of beetle posioning seen during seven days after exposure to slugs that had fed on soybeans with various seed treatments: U = Untreated (n = 19); F = Fungicide only (n = 17); F+L = Fungicide+ low rate thiamethoxam (n = 18); F+H = Fungicide + high rate thiamethoxam (n = 18)...... 117

Figure 4-7. Proportion of C. tricolor surviving by treatment of its slug prey: U = Untreated (n = 19); F = Fungicide only (n = 17); F+L = Fungicide + low rate thiamethoxam (n = 18); F+H = Fungicide + high rate thiamethoxam (n = 18). Survival curves that do not share a letter in the key are different at P ≤ 0.05...... 118

Figure A-1. Activity-density (mean ± SE) of three pest slug species observed under shelter traps during a field study in central Pennsylvania. Traps were located in both alfalfa and maize plots (n = 4 to 20 plots per date)...... 120

Figure A-2. Biomass of gray garden slugs (Deroceras reticulatum) collected from shelter traps in maize and alfalfa plots in 2011. The black symbol signifies the median biomass on each date...... 121

Figure A-3. Biomass of marsh slugs (Deroceras laeve) collected from shelter traps in maize and alfalfa plots in 2011. The black symbol signifies the median biomass on each date...... 121

Figure A-4. Biomass of banded slugs (Arion fasciatus) collected from shelter traps in maize and alfalfa plots in 2011. The black symbol signifies the median biomass on each date...... 122

Figure C-1. The relationship between slug damage (% defoliation of the lowest four leaves at V7) and ear weight of individual maize plants at harvest (n = 48 plants, split among 4 plots)...... 127

ix

LIST OF TABLES

Table 2-1. Sample sizes for slug predation study...... 45

Table 3-1. Management practices in maize plots in the three crop rotations...... 75

Table 3-2. Management practices and timing of management activities in maize split- plots across the three crop rotations. IM = injected manure; BM = broadcasted manure; RH = reduced herbicide; SH = standard herbicide...... 76

Table 3-3. Maize establishment (mean ± SE, n = 4) at V5 as a percentage of the target number of seeds planted, by year and rotation. Values marked with different letters are statistically different at P ≤ 0.05 based on Tukey’s post-hoc comparisons...... 77

Table 3-4. Slug damage (mean ± SE, n = 4) to maize seedlings at V2 and V5 by rotation. Values marked with different letters are statistically different at P ≤ 0.05 based on ESTIMATE comparisons within stages...... 77

Table 3-5. Slug damage to maize seedlings (mean ± SE, n = 4) at V2 and V5 by cover crop treatment in the Forage rotation. Values marked with different letters are statistically different at P ≤ 0.05 based on ESTIMATE comparisons within stages...... 77

Table 3-6. Percentage of maize seedlings severed by cutworms (mean ± SE, n = 4) by year and rotation...... 78

Table 3-7. European corn borer damage (mean ± SE, n = 4) by year and rotation. Values were square root transformed for analysis, but untransformed values are shown here. Values marked with different letters are statistically different at P ≤ 0.05 based on Tukey’s post-hoc comparisons...... 78

Table 3-8. Adult carabid species collected via pitfall trapping, 2010 and 2011...... 79

Table 3-9. Seasonal totals of predatory arthropods captured in pitfall traps in 2010 (mean ± SE, n = 4). Data were square root transformed for analysis but raw values are shown here. Values marked with different letters are significantly different based on Tukey’s post-hoc tests at P ≤ 0.05. Lowercase letters signify split-plot comparisons. IM = inject manure; BM = broadcast manure; RH = reduced herbicide; SH = standard herbicide...... 80

Table 3-10. Seasonal totals of predatory arthropods captured in pitfall traps in 2011 (mean ± SE, n = 4). Data were square root transformed for analysis but raw values are shown here. Values marked with different letters are significantly different based on Tukey’s post-hoc tests at P ≤ 0.05. Uppercase letters signify rotation comparisons. IM = inject manure; BM = broadcast manure; RH = reduced herbicide; SH = standard herbicide...... 81

x Table 3-11. Repeated measures analysis of predation on sentinel caterpillars in June 2011...... 82

Table 3-12. Predators observed during sentinel prey experiment, June 2011, three hours after sentinel caterpillars were deployed...... 82

Table 3-13. Repeated measures analysis of predation on sentinel caterpillars in July 2011...... 82

Table 3-14. Predators observed in sentinel prey experiment, July 2011, three hours after sentinel caterpillars were deployed...... 83

Table 3-15. Correlations between predation on sentinel caterpillars on 7/20/11 and numbers of arthropods in pitfall samples from 7/21/11 to 7/23/11. Predator activity- densities were square root transformed for analysis...... 83

Table 4-1. Sample sizes for the soybean-slug study with medium D. reticulatum and soybean seedlings, carried out in 3 trials...... 113

Table 4-2. Sample sizes for the slug-beetle study with medium D. reticulatum and C. tricolor, carried out in 3 trials...... 113

Table 4-3. ANOVA table for tests of the effects of slugs and seed treatment on soybean biomass at the end of seven days (small slug experiment)...... 113

Table 4-4. ANOVA table for tests of the effects of seed treatment and trial on slug mass change (medium slug experiment)...... 114

Table 4-5. Linear mixed model repeated measures analysis to test for the effect of seed treatment on soybean damage over seven days of slug feeding (medium slug experiment)...... 114

Table 4-6. ANOVA table for tests of the effects of slugs and seed treatment on soybean biomass at the end of seven days of slug feeding (medium slug experiment)...... 114

Table B-1. Mixed model analysis for cross-rotation comparisons, including split-plot effects...... 123

Table B-2. Mixed model analysis for the Forage rotation, including split-plot and split- split-plot effects (2011 only): ...... 123

Table B-3. Repeated measures mixed model analysis for cross-rotation comparisons, including split-plot effects (e.g. with two time points, t = 2)...... 124

xi

PREFACE

Chapter 1 of this thesis was multiply authored with Margaret R. Douglas as the first author and John F. Tooker as the second author. MRD conducted much of the literature search and drafted the manuscript. JFT contributed to the literature search and also helped draft the manuscript.

xii

ACKNOWLEDGEMENTS

It truly took a village to produce this thesis. In the Tooker lab, I thank Ian Grettenberger, Eric Bohnenblust, Anjel Helms, and Anthony Vaudo for their valuable feedback, camaraderie, and hands-on help during the conception and execution of this thesis. Special thanks go to Ian for sharing his impressive skills in insect photography. Sarah Kossak introduced me to the lab and was brave enough to accompany me on my first forays in the lab truck. Andrew Aschwanden assisted with virtually every experiment at one time or another, and was always ready with a helping hand and a winning attitude. Many undergraduates contributed field and laboratory assistance including Amber Delikat, Nate Blunk, Hoonie Kwon, Glendon Taylor, Allie Schoffner, and Curtis London. Janet Teeple also assisted with laboratory studies. I wish to especially thank Christy Rose Aulson, who became my right-hand woman in all manner of slug explorations. I am also thankful for the guidance and encouragement of my advisor, Dr. John Tooker, who besides always having an open door for my (many) questions, was known to accompany me at odd hours of the night in search of slugs. I appreciate John’s bottomless and infectious enthusiasm for all things bug or slug. I was fortunate to be part of the group of scientists and extension specialists involved in the Sustainable Dairy Cropping Systems project. In particular, I thank Drs. Bill Curran, Roger Koide, Doug Beegle, Pete Kleinman, and Tom Richard, as well as Craig Altemose, Ron Hoover, Ginny Ischler, and our fearless leader Dr. Heather Karsten, for their guidance and support. In addition, working in the project was far more enjoyable in the company of Kristen Haider, Elina Snyder, Stephanie Bailey, Emily Duncan, and Robb Meinen, many of whom also provided field assistance on several occasions. Special thanks are due to Dr. Glenna Malcolm for initiating me into the mysterious ways of SAS and managing to accommodate the insect and slug work in the larger scheme of the project against all odds. Scott Harkom and the entire farm crew at Rock Springs went above and beyond the call of duty to carry out field operations without dislodging pitfall and slug traps. My committee members, Drs. Heather Karsten and Mary Barbercheck, were supportive all along the way of this research. Their constructive feedback greatly improved the quality of this thesis. I am also thankful to Mary’s technician Christy Mullen for helping to collect ground beetles for several lab experiments. I benefited from my interactions with many other faculty and

xiii graduate students in the Department of Entomology as well. In particular, Drs. Chris Mullin and Jim Frazier provided helpful insights in the planning of the seed treatment study, and Dr. Shelby Fleischer kindly allowed me to glean reference specimens from his carabid collection. I thank Becky Heinig for assisting with night observations. I am also thankful for the excellent support staff in the Department of Entomology, with special thanks to Nick Sloff for help with photography. At other institutions, Dr. Ron Hammond (Ohio State), Joanne Whalen (University of Delaware), Grant Troop (formerly of Penn State Cooperative Extension) and Dr. Bob Byers (retired from USDA-ARS) improved the quality of the slug review in Chapter 1. Dr. Byers also shared his impressive library of slug literature and knowledge of slug biology. Dr. Tim Pearce and Bob Davidson at the Carnegie Museum of Natural History provided helpful advice on mollusk and beetle identification. Finally, I am truly indebted to my friends and family for their encouragement and understanding during this program. My parents, Doug and Karen Douglas, have nurtured my curiosity from the beginning and continue to cheer on my scientific pursuits with surprising intensity. I thank Katie and Laura for keeping me grounded and often providing much-needed comic relief. Last but certainly not least, I thank Bill Freese for his endless support, grocery shopping, garden tending, helping on too many Saturdays, and for sharing the highs and lows of this experience with me. For funding these studies I thank the USDA Northeast Sustainable Agriculture and Education program, the research society Sigma Xi, the Maryland Grain Board, and the Department of Entomology.

Introduction

Agriculture faces several major challenges now and in the future. Paramount among these is to feed a growing population while conserving the soil, water, and biodiversity that ultimately make farming possible (Vitousek 1994, Altieri 1999, Tilman et al. 2001, 2002, Foley et al. 2011). Agro-ecological research can help to develop cropping systems that bolster biological processes to supply nutrients and regulate pests, thereby requiring fewer external inputs and minimizing agricultural pollution (Vitousek 1994, Robertson and Swinton 2005). Importantly, such low- external-input systems also have the potential to improve the economic sustainability of farming by decreasing costs while maintaining productivity (e.g. Liebman et al. 2008). Despite progress along these lines, there is still much work to be done to develop farming systems that are productive, profitable, and environmentally sound (Robertson and Swinton 2005, Foley et al. 2011). Overcoming agricultural challenges will require not one but many solutions, appropriate to particular geographies. Although food is traded on global markets, growing food remains a necessarily local affair, rooted in local conditions and constraints. Field crops are a major part of the landscape in Pennsylvania, in part because they help support the state’s large dairy and other livestock industries (ERS 2012). As in much of the Northeast and mid-Atlantic, Pennsylvania’s field crop growers contend with heavy and rocky soils, sloping land, and unpredictable weather patterns including periods of heavy rain and drought. Farming practices in Pennsylvania also have important consequences for the fate of regional resources, such as the sensitive Chesapeake Bay (NRCS 2011). In this context, reducing or eliminating tillage offers several important advantages in field crop production. Since the birth of soil conservation techniques over 90 years ago (Nelson 1997), conservation tillage has been further developed by farmers, agricultural researchers, and agribusinesses (Nelson 1997, Bennett 2000), and encouraged through farm policy (Bennett 2000, NRCS 2006). Some of the advantages of no-till farming in particular include protecting soil from wind and water erosion (Montgomery 2007), conserving soil moisture (Norwood 1998, DeFelice et al. 2006), requiring fewer farming operations at planting (Phillips et al. 1980), and using less fuel (Phillips et al. 1980, West and Marland 2002). Today, more than a third of U.S. cropland is farmed without tillage (Horowitz et al. 2010) and in Pennsylvania no-till adoption is even higher.

2 Between 2005 and 2010, no-till became the dominant management style in maize, Pennsylvania’s largest acreage crop (ERS 2011). However, as with many farming practices, reducing or eliminating tillage entails agronomic and environmental trade-offs (Phillips et al. 1980). No-till systems often rely heavily or entirely on herbicides for weed control, a practice that can lead to water pollution and herbicide-resistant weed populations (Mortensen et al. 2012). Without the ability to incorporate manure or other nutrients into the soil, no-till systems in the Northeast and mid-Atlantic are also at risk for increased loss of nutrients into waterways (Kleinman et al. 2009). Finally, slugs are virtually unknown in field crops under conventional tillage in this region, but are a formidable pest once tillage is reduced or eliminated (Hammond and Byers 2002). Slugs are broad generalists that feed on virtually every crop species and can hinder crop establishment (South 1992, Barker 2002). In some cases, farmers have even returned to tillage to manage destructive slug populations (Willson and Eisley 1992, Hammond et al. 1996). The continued success of conservation tillage will therefore depend in part on developing strategies for slug management. In this thesis, I report the results of ecological investigations toward this end.

Thesis objectives

In Chapter 1, my advisor and I review the ecology and management of slugs in no-till field crops, with emphasis on the mid-Atlantic region. The goal of this review was to pull together important facets of slug biology, natural history, scouting, and management, to inform scientists, extension specialists, and land managers, and to identify gaps in understanding. We identified several areas in need of research, including: 1) better documenting slug natural history in the region, 2) identifying potential slug predators, and 3) studying the influence of management practices on slugs, their predators, and the balance between them. The remaining chapters contribute to addressing these needs. In Chapter 2, I used laboratory assays to screen common arthropod predators for their ability to prey on common slug species. These assays included soybean seedlings so that I could assess not only outright predation, but also effects of predators on slug feeding behavior. In Chapter 3, I used field studies to explore the influence of insect, weed, and nutrient management practices on pests, predators, and predation services in reduced-tillage maize systems. I focused mainly on slugs as pests although important insect pest species (black

3 cutworm [Agrotis ipsilon], European corn borer [Ostrinia nubilalis]) were also included. This study examined how practices designed to address some of the challenges of reduced-tillage systems (weed and manure management) influence invertebrate dynamics. The other major goal of this study was to determine if low-external-input maize systems could compete in terms of pest management with a more conventional system relying on pre-emptive insect management tactics including broadcast insecticides and transgenic traits for insect resistance. In Chapter 4, I further studied the interaction between insecticides and slug management. I used laboratory studies to examine the influence of the common neonicotinoid seed treatment thiamethoxam on slug damage to soybeans in a tritrophic context. The goal of this study was to test whether slugs are 1) susceptible to this insecticidal seed treatment and 2) able to transmit these insecticides from treated seedlings to their beetle predators. Together, these studies contribute to knowledge of slug ecology and integrated pest management in reduced-tillage farming systems.

4

References

Altieri, M. A. 1999. The ecological role of biodiversity in agroecosystems. Agriculture, Ecosystems and Environment 74: 19-31. Barker, G. M. 2002. Molluscs as Crop Pests. CABI Publishing, Wallingford, U.K. Bennett, H. H. 2000. Reconstructing the farm landscape: the spread of conservation tillage in the United States. pp. 255-295. In: C. M. Coughenour and S. Chamala (eds.) Conservation Tillage and Cropping Innovation: Constructing the New Culture of Agriculture. Iowa State University Press, Ames, IA. DeFelice, M. S., P. R. Carter, and S. B. Mitchell. 2006. Influence of tillage on corn and soybean yield in the United States and Canada. Crop Management. Accessed 6/16/12: www.plantmanagementnetwork.org. doi:10.1094/CM-2006-0626-01-RS. Economic Research Service. 2011. Crop Production Practices. Agricultural Resource Management Survey. U. S. Department of Agriculture, Washington, D.C. Accessed 6/1/12: http://www.ers.usda.gov/Data/ARMS. Economic Research Service. 2012. State Fact Sheets: Pennsylvania. U.S. Department of Agriculture, Washington, D.C. Accessed 6/16/12: http://www.ers.usda.gov/StateFacts/HTML2PDF/PA-Fact-Sheet.pdf Foley, J. A., N. Ramankutty, K. A. Brauman, E. S. Cassidy, J. S. Gerber, M. Johnston, N. D. Mueller, C. O’Connell, D. K. Ray, P. C. West, C. Balzer, E. M. Bennett, S. R. Carpenter, J. Hill, C. Monfreda, S. Polasky, J. Rockström, J. Sheehan, S. Siebert, D. Tilman, and D. P. M. Zaks. 2011. Solutions for a cultivated planet. Nature 478: 337-342. Hammond, R. B., and R. A. Byers. 2002. Agriolimacidae and Arionidae as pests in conservation- tillage soybean and maize cropping in North America, pp. 301-314. In G. M. Barker (ed.), Molluscs as Crop Pests. CAB International, Wallingford, UK. Hammond, R. B., J. A. Smith, and T. Beck. 1996. Timing of molluscicide applications for reliable control in no-tillage field crops. Journal of Economic Entomology 89: 1028- 1032. Horowitz, J., R. Ebel, and K. Ueda. 2010. “No-Till” farming is a growing practice. Economic Information Bulletin No. 70. USDA-ERS, Washington, DC.

5 Kleinman, P. J. A., A. N. Sharpley, L. S. Saporito, A. R. Buda, and R. B. Bryant. 2009. Application of manure to no-till soils: phosphorous losses by sub-surface and surface pathways. Nutrient Cycles in Agroecosystems 84: 215-227. Liebman, M., L. R. Gibson, D. N. Sundberg, A. H. Heggenstaller, P. R. Westerman, C. A. Chase, R. G. Hartzler, F. D. Menalled, A. S. Davis, and P. M. Dixon. 2008. Agronomic and economic performance charateristics of conventional and low-external-input cropping systems in the central Corn Belt. Agronomy Journal 100(3): 600-610. Montgomery, D. R. 2007. Soil erosion and agricultural sustainability. Proceedings of the National Academy of Sciences 104(33): 13268-13272. Mortensen, D. A., J. F. Egan, B. D. Maxwell, M. R. Ryan, and R. G. Smith. 2012. Navigating a critical juncture for sustainable weed management. BioScience 62(1): 75-84. Nelson, P. J. 1997. To hold the land: soil erosion, agricultural scientists, and the development of conservation tillage techniques. Agricultural History 71(1): 71-90. Norwood, C. A. 1998. Water use and yield of dryland row crops as affected by tillage. Agronomy Journal 91(1): 108-115. Robertson, G. P. and S. M. Swinton. 2005. Reconciling agricultural productivity and environmental integrity: a grand challenge for agriculture. Frontiers in Ecology and the Environment 3(1): 38-46. South, A. 1992. Terrestrial Slugs: Biology, Ecology and Control. Chapman & Hall, London, UK. Tilman, D., J. Fargione, B. Wolff, C. D’Antonio, A. Dobson, R. Howarth, D. Schindler, W. H. Schlesinger, D. Simberloff, and D. Swackhamer. 2001. Forecasting agriculturally driven global environmental change. Science 292: 281-284. Tilman, D., K. G. Cassman, P. A. Matson, R. Naylor, and S. Polasky. 2002. Agricultural sustainability and intensive production practices. Nature 418: 671-677. Natural Resources Conservation Service. 2006. Conservation resource brief: Soil erosion. U.S. Department of Agriculture, Washington, D.C. Natural Resources Conservation Service. 2011. Assessment of the effects of conservation practices on cultivated cropland in the Chesapeake Bay Region. U.S. Department of Agriculture, Washington, D.C. Phillips, R. E., G. W. Thomas, R. L. Blevins, W. W. Frye, and S. H. Phillips. 1980. No-tillage agriculture. Science 208(4448): 1108-1113. Vitousek, P. M. 1994. Beyond global warming: Ecology and global change. Ecology 75(7): 1861- 1876.

6 West, T. O. and G. Marland. 2002. A synthesis of carbon sequestration, carbon emissions, and net carbon flux in agriculture: comparing tillage practices in the United States. Agriculture, Ecosystems, and Environment 91(1-3): 217-232. Willson, H. R., and J. B. Eisley. 1992. Effects of tillage and prior crop on the incidence of five key pests on Ohio corn. Journal of Economic Entomology 85: 853 – 859.

Chapter 1

Slug (Mollusca: Agriolimacidae, Arionidae) ecology and management in no- till field crops, with an emphasis on the mid-Atlantic region

This chapter has been published in the Journal of Integrated Pest Management with the citation:

Douglas, M. R. and J. F. Tooker. 2012. “Slug (Mollusca: Agriolimacidae, Arionidae) ecology and management in no-till field crops, with an emphasis on the mid-Atlantic region.” Journal of Integrated Pest Management 3(1): C1-C9. DOI: http://dx.doi.org/10.1603/IPM11023.

Slugs have been one of the most serious pests of crops grown in no-tillage systems since these conservation-based farming practices were first adopted in North America (Gregory and Musick 1976). Slugs thrive in the low-disturbance, residue-rich environments characteristic of no-till fields, and now with no-till farming practiced on more than 88 million acres (35.5%) of US cropland (Horowitz et al. 2010), slugs have become a prominent pest in parts of the United States with high no-till adoption rates. For instance, in a recent Pennsylvania survey, over 80% of no- till growers identified slugs as their most challenging pest problem (n = 61, J. Tooker, unpublished data). Heavy and perennial slug damage can even convince some frustrated growers to return to tillage to control their heavy slug populations (Willson and Eisley 1992, Hammond et al. 1996, J. F. Tooker, personal observation). One area of heavy no-till adoption has been the mid-Atlantic region, where no-till farming has been encouraged to limit agricultural run-off into streams and other bodies of water that flow into the Chesapeake Bay (USDA-NRCS 2011). In Pennsylvania in 2009, for example, 58% (1.2 million acres) of corn (Zea mays L.), soybeans (Glycine max (L.) Merrill), barley (Hordeum vulgare L.), wheat (Triticum spp. L.), and oats (Avena sativa L.) were farmed without tilling (USDA-NASS 2009). In particularly cool and wet years, like 2009, much of this acreage can be at risk for slug damage, but it is has been estimated that slugs cause stand reductions and yield loss on approximately 20% of no-till acres annually (J. Whalen Univ. of Delaware, personal communication). Consistent with this, 20% of growers in our survey indicated that they experience trouble with slugs every year, and an additional 47% see significant slug damage every two to three years.

8 Compounding the ongoing challenge posed by slugs is a limited understanding of slug ecology across the region and, beyond tillage, a paucity of reliable slug control tactics. Growers have access to just a few active ingredients that are labeled against slugs, and many growers rely on homespun solutions that are not very well evaluated. Here, we review the natural history and ecology of slugs with a focus on mid-Atlantic field crops, discuss the available scouting and management options, particularly for corn, and highlight areas in need of research.

Description of species and life cycles

Slugs are close relatives of snails - essentially snails without a shell. They are legless, soft-bodied creatures with four front tentacles, two that carry the eyes and two that operate like antennae. Slugs also have a covering of slimy mucus all over their bodies. In addition to the mucus on the outside of the body, when slugs travel they secrete mucus from the pedal gland, located at the anterior end of the animal (South 1992). This mucus aids in slug locomotion and leaves behind a characteristic “slime trail” that can be a valuable clue of their presence. Different species vary in color and pattern, but all are various earth tones such as gray, brown, or orange. Again varying by species and age, slugs can range in size from a few millimeters to several centimeters. Over fifteen slug species occur in the mid-Atlantic United States (Pearce 2008), but only four appear to be common in field crops: Deroceras reticulatum Müller, (gray garden, or gray field, slug, Figure 1), Deroceras laeve Müller (marsh slug, Figure 2), Arion subfuscus Draparnaud (dusky slug, Figure 3), and Arion fasciatus Nilsson (banded slug, Figure 4). While some of these species are not commonly associated with damage, D. reticulatum appears to be the most economically important, often occurring in the largest numbers and most often associated with crop damage (Hammond and Byers 2002). This is a medium-sized (up to 5 cm long), light to dark gray slug that produces sticky, white mucus when disturbed (an identifying characteristic of this species). It often has a mottled appearance, although pattern can vary greatly. Deroceras reticulatum appears to be native to Western Europe, where it is a common pest of agricultural crops (Kerney and Cameron 1979), but has been introduced to most parts of the world, frequently becoming a serious pest in areas of introduction, including North America where it was established by 1843 (Chichester and Getz 1969). Deroceras reticulatum appears particularly well-adapted to crop fields. It seems to have less restrictive water requirements than many slug

9 species so it can survive better in crop fields, where it also encounters less competition from other slug species (South 1992). Moreover, D. reticulatum is less common in natural habitats (Chichester and Getz 1973). Deroceras laeve (Figure 2) looks similar to D. reticulatum but is often darker, occasionally being nearly black, and produces clear, watery mucus when disturbed (an identifying characteristic of this species). As an adult, it is smaller than D. reticulatum (up to 2.5 cm). Historically D. laeve had a Holarctic distribution so it evolved in North America, but this species has also been moved worldwide and some populations in the U.S. appear to have been introduced from Europe and elsewhere (South 1992). It has a very wide habitat range, including crop fields and gardens, as well as a variety of natural habitats including moist woodlands and marshy areas (Getz 1959, Chichester and Getz 1973). In most cases it does not appear to reach the high population densities of D. reticulatum in field crops of the mid-Atlantic, although it can be a significant pest in some settings, including greenhouses and irrigated vegetable production (Godan 1983, South 1992). Arion subfuscus (Figure 3) is a large (up to 8 cm) tan, brown, or even orange-looking slug that produces orange mucus upon disturbance (an identifying characteristic of this species). It also appears to be native to Western Europe, but is more common in northern portions of Europe, and has been introduced to many countries where it is an occasional crop pest (Kerney and Cameron 1979, South 1992). In Pennsylvania, it is more common in crop fields in western counties. Arion subfuscus is now thought to be a cryptic species complex containing at least two distinct species, A. subfuscus s.s. and A. fuscus, which can only be distinguished on the basis of genital characteristics (Pinceel et al. 2004, 2005). Both species are present in the Northeastern United States (Barr et al. 2009). Arion fasciatus (Figure 4) is a medium-size slug as an adult (up to 5 cm), has a dark lateral band extending down each side of the body, and on the dorsal side often has a thin, dashed white line, which is faintly evident near the tail of juveniles. Like A. subfuscus, A. fasciatus appears to be introduced to the U.S. from Western and Northern Europe, but confusion between it and two other similar-looking species (A. silvaticus and A. circumscriptus) limits the reliability of historic records and distribution maps (Kerney and Cameron 1979). Arion fasciatus can be distinguished from these two other species by the presence of a yellowish or orangish band below each lateral body band (Chichester and Getz 1973). That being said, the species status of these three closely-related taxa remains in flux (Geenen et al. 2006). Arion fasciatus is often easy to

10 find in mid-Atlantic crop fields, particularly pre-planting in spring, but rarely appears to be associated with crop damage. All slugs are hermaphrodites, but their mating systems are species-specific and can be quite complex. Genetic data suggest that D. reticulatum is predominantly outcrossing, usually mating with other individuals to reproduce, while other species of pest slug may be more prone to self-fertilization (McCracken and Selander 1980). Mating, egg-laying, hatching, and development are not well synchronized even within a single species, so slugs of various stages of development can be found at many times of year. This makes slug activity difficult to predict, but generally speaking slugs are most active and damaging April-June and then again in September and October (Godan 1983). Spring-time damage can often be caused by newly hatched D. reticulatum, which appears to have a more synchronized life cycle in this region than some other species. In central Pennsylvania, a large portion of the population seems to overwinter as eggs. In Ohio, both eggs and adults of D. reticulatum have been reported to overwinter, although it appears that the newly hatched juveniles are still responsible for most crop damage in spring (Hammond et al. 1996). In Delaware, adults and juveniles of D. reticulatum are both present in the spring. Although it was unclear prior to 2010 whether adults or hatchlings are most responsible for crop damage in spring, recent surveys indicate that juveniles appear to cause the most damage (J. Whalen Univ. of Delaware, personal communication). Slug eggs are small, gelatinous spheres or ovals found under residue or in the soil (Figure 5). The eggs are often found in clumps but may also occur singly. These eggs tend to hatch in Pennsylvania in early to mid May, which is about the time when corn and soybean are in vulnerable seedling stages. Deroceras reticulatum juveniles, which resemble adults but are smaller, grow through the spring and summer, emerging at night and during rains and usually hiding during the day, mature in the late summer or early fall, mate, and lay eggs in the fall. It is often mature slugs of D. reticulatum that are responsible for damage to fall-planted small grains, forages, and cover crops. Eggs from these individuals overwinter and hatch the following spring. Despite this general pattern, it is apparent that D. reticulatum eggs can often be found throughout the year, and some areas will see in early autumn a significant hatch of eggs and damage from juveniles (South 1992). It may appear that this fall hatch is the result of a second generation, and some areas of Europe see a second generation. However it is more likely that it is a result of eggs laid during spring and early summer by individuals that survived the winter and resumed activity in spring (Hunter and Symonds 1971). In lab studies, the maximum life span of D. reticulatum is twelve months or so and the shortest generation time would be at least six

11 months because eggs can take anywhere from two to five months to hatch (South 1992). Individual D. reticulatum can lay several hundred eggs in their lifetime (Port and Port 1986). The life cycles of the other slug species common in mid-Atlantic crop fields also are reported to be annual (South 1992), but in mid-Atlantic states they are not synchronized with D. reticulatum. Deroceras laeve and A. fasciatus appear to overwinter more often as adults or juveniles rather than eggs because large individuals of these species are common in early spring (personal observation). This matches the observation that D. reticulatum adults were fairly sensitive to winter cold in Ontario, whereas D. laeve was quite tolerant (Rollo and Shibata 1991). Adults of Arion subfuscus were reported to die in summer in central New York (Beyer and Saari 1978). Some Arion species have been reported to live as many as two years, particularly if they are unsuccessful at finding a mate (South 1992). As with D. reticulatum, populations of these other species can be significantly influenced by hard winters, which can kill adults and juveniles, but thick snow packs can insulate slugs against the cold and allow more to survive the winter. Despite a basic understanding of slug life cycles, regional variation in slug phenology is poorly documented. Improved monitoring efforts, particularly for D. reticulatum, could help growers and pest managers better anticipate and manage slug damage, for instance by timing crop planting in spring to avoid periods of greatest slug activity.

Host-plant species and damage

Slugs can feed on a wide range of host-plant species and are well known as agricultural pests. In no- and reduced-tillage field crop production, they are considered serious pests of many crop species, including wheat, barley, oats, rye (Secale cereal L.), corn, soybeans, tobacco (Nicotiana tabacum L.), canola (Brassica napus L. or Brassica rapa L.), alfalfa (Medico sativa L.), and other cereals and leguminous forages (Godan 1983, Hammond and Stinner 1987, South 1992, Barratt et al. 1994, Cook et al. 1996, Hammond et al. 1999, Byers 2002). Most slug- induced crop damage occurs within a month of planting when crops are vulnerable seedlings (Byers et al. 1983, South 1992). Slugs damage crops by feeding on the seed, resulting in plant mortality prior to emergence and poor crop stands, and then damage seedlings as plants emerge from the ground (South 1992). Slugs feed by scraping with their radula on the surface of their food, which can include seeds, roots, stems, leaves, and flowers.

12 In corn and many small grains, slugs scrape strips in the leaves, leading first to window- pane damage, and then to leaf shredding (Figure 6). In soybeans, slugs create craters in cotyledons (Figure 7), then ragged holes in leaves, but cause plant mortality by killing the apical meristem. Similar ragged holes to those seen on soybeans are seen on slug-damaged canola, alfalfa, and other broadleaf crops (Figure 8). There are reports of complete defoliation of some crops, including tobacco, under extreme population densities (Godan 1983). Slime trails, often associated with slug damage, can be used to confirm the presence of slugs in a field. Seedlings are especially at risk when the seed furrow or slot is left open, creating dark, cool slug “highways” leading right to the next seedling. For potatoes and some horticultural crops (e.g., grapes), slugs can even vector diseases, but this phenomenon does not appear to have been reported for field crops (South 1992). For many crop species, the economic impact of slug feeding has been hard to quantify. For corn, plants can outgrow apparently heavy damage with little yield loss. However, a given amount of slug damage can correspond to more or less yield loss depending on weather conditions during and after slug feeding (Byers and Calvin 1994). For soybeans, damage can be quite severe if slugs reduce plant populations (Barratt et al. 1994), but so long as slugs do not kill seedlings, soybeans can withstand significant defoliation without suffering significant yield loss (Hammond 2000). In forages, slugs can kill seedlings during establishment, contributing to lowered yields in the establishment year (Byers and Templeton 1988). Through selective feeding, slugs can also decrease the amount of legumes in mixed forage stands, leading to a less desirable grass- or weed-heavy mix (Byers 2002). For cereals, slug-thinned stands often have increased tillering, mitigating yield losses (South 1992), although yield losses can be significant under high slug densities (Barratt et al. 1994). A large body of literature has tried to clarify slug, and particularly D. reticulatum, feeding preferences (reviewed in South 1992). Much of this research, however, provides only limited information because feeding assays often use leaf disks, detached leaves, or crushed leaves that have been incorporated into agar (e.g. Cates and Orians 1975, Dirzo 1980, Rathcke 1985, Molgaard 1986, Cook et al. 1996), with the exception of a few studies using whole plants (e.g. Kozlowski and Kozlowska 2004). Living plants are better able to mobilize defenses in response to herbivory and perhaps fend off some slug feeding, but confining slugs to live plants and quantifying the amount of damage can be challenging. Plant architecture may also be quite important to slug preferences in a field setting, as slugs tend to feed heavily on leaves near the soil surface (personal observation). Nevertheless, existing preference work appears to indicate

13 that D. reticulatum prefers plant species in Fabaceae, Brassicaceae, Asteracaeae, and cultivated cereals (South 1992). In some experiments, D. reticulatum preferred clover species (Trifolium pretense L., T. repens L.) and weedy plant species such as narrowleaf plantain (Plantago lanceolata L.), dandelion (Taraxacum officinale L.), shepherd’s purse (Capsella bursa-pastoris L.), and lamb’s quarters (Chenopodium album L.), suggesting that slightly weedy fields, or fields deliberately underseeded with a preferred clover species, might help limit crop damage by providing slugs alternative food sources (Cook et al. 1996, 1997; Peters et al. 2000, Brooks et al. 2003). Experiments conducted in wheat indicate that slightly weedy fields or intercropping might reduce slug damage (Cook et al. 1997, Brooks et al. 2005). Further studies are needed to see if these strategies would apply to other row crops (i.e. corn and soybeans). In pastures and old fields, slugs are strong drivers of plant community composition because they preferentially feed on seedlings of certain species (e.g., Peters et al. 2000). Within crop plant species, oats appear less palatable than barley, which is preferred less than rye and wheat, perhaps due to host-plant chemistry (Duthoit 1964, Godan 1983, South 1992). Alfalfa and red clover are preferred over bird’s-foot trefoil (Byers and Bierlein 1982). Some plant species show intraspecific variation in susceptibility to slug feeding and chemistry may also explain these patterns (South 1992, Peters et al. 2000). Certain potato varieties, for instance, are less susceptible to slugs than others, perhaps due to the higher levels of trypsin inhibitors they produce (Port and Port 1986). Similarly in oilseed rape, slug damage is inversely related to concentrations of glucosinolates in young seedlings (Glen et al. 1990). In wheat, however, twelve varieties were similarly preferred by D. reticulatum (Cook et al. 1996). Slugs, and D. reticulatum in particular, appear capable of “learning” because they can avoid unpalatable plant varieties after limited exposure (Gouyon et al. 1983). In addition to live plants, pest slugs have also been documented to eat fungi, plant residue, and occasionally one another or other invertebrates (Pallant 1972, Fox and Landis 1973, Jennings and Barkham 1975, Beyer and Saari 1978, Lundgren et al. 2006). The Arion species in particular are thought to feed more heavily on fungi (Chichester and Getz 1973, Beyer and Saari 1978). As mentioned above, slugs can also survive on soil organic matter (Miles et al. 1931). The extent to which slugs feed on these alternative foods in field crops is unknown, but may be important in fully understanding slug population dynamics and relationships with crop plants. For instance, slug feeding activity may hasten decomposition and thereby alter soil nutrient dynamics (e.g. Theenhaus and Scheu 1996).

14 Scouting

Economic thresholds are not available to guide slug control decisions for most crop species. One research effort established economic injury levels (EIL) for slug damage to corn seedlings in wet and dry years, but associated economic thresholds were not developed (Byers and Calvin 1994). The EIL for corn ranged from 2-20% leaf area removed in a warm, wet year, and from 39-59% leaf area removed in dry years, depending on the value of the crop and the cost of the control tactic (Byers and Calvin 1994). The variability in this EIL makes it difficult to implement, but reemphasizes the point made above that slug damage to crops can be difficult to quantify in part because plants have most of the growing season to recover from sub-lethal damage. In general, many crop species can recover from significant defoliation during early vegetative stages. Corn hybrids appear capable of withstanding at least 40% defoliation during early growth without reductions in yield (Vorst 1986). In soybeans, 50% defoliation of the first unifoliate leaflets caused only minor yield loss, while 50% defoliation of the first trifoliate leaflets caused no yield loss at all (Hammond 2000). More concerning are stand reductions caused by heavy slug feeding (Hammond 2000), most likely to occur in dicotyledonous crops where the growing point is prone to slug grazing. Despite the lack of economic thresholds to help prevent economic loss, scouting for slugs is still useful because it can help identify areas with large slug populations and identify fields at risk. Farmers tend to know which of their fields historically have been troubled by slugs, but it has been our experience that slug populations in many fields catch farmers by surprise. Estimating absolute density of slugs usually requires soil sampling (South 1964, Hunter 1968), and unfortunately is labor-intensive and impractical for farmers or crop consultants. A number of less-intensive techniques can provide insight into relative slug populations. Specialists in Ohio recommend scouting for adults in fall to identify potential problem fields and to get a relative idea of population size which can help predict spring populations (R. Hammond, personal communication). In spring prior to seeding, slug eggs and overwintered slugs can be found by looking under crop residue, especially on mild days soon after rain. Another approach to find slugs is to place artificial shelters in the field, such as roofing shingles (Figure 9), old boards, wet cardboard, or anything that will create a dark, cool, moist environment. An evaluation of artificial shelters made from various materials concluded that black roofing shingles wrapped in aluminum foil were most effective shelters for assessing populations of D. reticulatum, D. laeve, and A. fasciatus (Schrim and Byers 1980). Several days after putting

15 shelters out, slugs can be found under the shelters during the day. Because shelters can warm up in the heat of the day, it is best to check them in the morning or evening to have maximum potential for detecting slugs (Hommay et al. 2003). Once crops have emerged, slugs can be found by inspecting crops in the evening with a flashlight. With all of these methods, it is important to look closely because juvenile slugs can be very small. Also in early spring, the absence of juvenile slugs could signal not that slug populations are low, but rather that they have not yet hatched from their eggs. Regular scouting can help identify when egg hatch is to be expected. To more widely document slug populations, extension specialists and educators in mid- Atlantic states have recently made an effort to standardize sampling protocols. We have settled on widely available white rolled roofing (Owens Corning; color Shasta White), which are cut into 1 × 1 ft pieces using a reciprocating saw and utility blade. The white color helps reflect sunlight, keeping them cooler than darker shingles. Shingles are then placed randomly in the field. We move residue aside and have shingles rest directly on the soil where they can better act as artificial shelters. While shingle traps are a crude sampling technique, our ongoing research indicates that the number of slugs beneath shingles is roughly correlated with damage to corn and alfalfa seedlings in Pennsylvania, and the strength of this relationship can be improved by increasing the number of shingles and averaging the number of slugs found under shingles on multiple dates. Overall, priority for slug scouting should be given to: 1) fields with a history of slug problems, 2) fields with abundant surface residue, and 3) fields that are low-lying and/or with heavy soil. Significant slug populations are most likely to materialize when a mild winter is followed by a wet spring. Wet falls may also foster high slug populations the following spring, since abundant soil moisture encourages egg-laying (Willis et al. 2008). Scouts should pay closest attention to new crop growth on successive scouting periods to determine if plants are outgrowing slug damage. Finally, the weather forecast can help inform slug management decisions. If mild, wet, and cloudy weather is expected, slug damage will likely continue and crop growth will be slow, whereas warm and dry conditions are likely to favor plant recovery over slug activity.

Environmental influences

Slugs are quite sensitive to a range of environmental and biological factors. They are most active and damaging in periods of mild and wet weather. Optimal conditions for D.

16 reticulatum are between 17-20° C (63-68° F) and 100% relative humidity, and activity tends to increase when air temperatures drop below 21° C (70° F; Godan 1983, South 1992). Slugs can, however, remain active during colder temperatures and slug feeding, although not much movement, can occur as low as 1° C (34° F; Mellanby 1961). Slugs can also survive several hours of freezing and recover well enough to lay eggs if returned to a preferred temperature (Godan 1983). At high temperatures (27-35° C [81-95° F]), slug activity is substantially inhibited by water loss, but slugs can deal with high temperatures and associated water loss in the following ways (Godan 1983). First, slugs, which are about 80% water, are able to tolerate water loss as high as 50% of their mass, and they can reabsorb water directly through their skin when it becomes available. Second, some slug species have a limited ability to thermoregulate at high temperatures and maintain body temperatures that are considerably cooler than prevailing air temperatures. Third, at high temperatures slugs of the same species tend to huddle together, a tactic presumed to reduce water loss to individuals in the bunch. Finally, slugs can seek shelter from high temperatures by traveling deeper into the soil. Slugs are often found in association with orchardgrass and other bunch-forming grasses (South 1965). Given the sensitivity of slug populations to precipitation and temperature, perhaps it is not surprising that slug populations are expected to shift dramatically with global climate change. For scenarios developed for the United Kingdom, areas prone to slug damage would gain some relief under future climatic conditions whereas areas not typically associated with slugs seem likely to develop significant slug populations (Willis et al. 2006). In addition to reducing activity with high temperatures, slugs have an even stronger sensitivity to changes in light intensity (Godan 1983). Slugs are nocturnal and emerge to feed on aboveground plant material after dusk and then return at dawn to shelters under plant residue, rocks and under the upper soil layers, among other sheltered locations. Rainfall can override this daytime hiding, often causing them to become active during the day (South 1992). During evening, slugs tend to have periods of greater activity. Deroceras reticulatum is most active four to six hours after dark and then again around 3 or 4 AM.; D. laeve is most active around 6 AM (Godan 1983). Slug populations and activity are also strongly influenced by soil types. Slugs are more common on heavy, wet, infrequently tilled soils, and problematic slug species, like D. reticulatum and A. subfuscus, prefer lightly alkaline or neutral soils (Godan 1983). In these heavy soils, slugs exploit holes and gaps to move within soil and can follow these passages several centimeters below the soil surface, allowing them to find shelter when necessary, but also access seeds and

17 newly emerged seedlings (Godan 1983, South 1992). Open seed furrows are ideal habitats for slugs and allow them clear access to an abundance of seeds in a sheltered, often moist environment. Slugs can do particularly well in soils with 3% organic matter or greater, because they can feed on organic matter (Godan 1983). Given their need for a dark, moist microclimate, it is not surprising that slugs and their damage are most common where crop residue is heaviest. Slugs gain shelter from thick residue and can severely damage crops planted into these environments, so farming practices that result in higher levels of residue are likely to increase the amount of damage inflicted by slugs (e.g. Hammond and Stinner 1987). It is worth noting that since 1990 corn yields in Pennsylvania have increased an average of 0.28 Mg ha-1 yr-1; therefore, the amount of stover left after grain harvest has also increased considerably (Grover et al. 2009), potentially contributing to increased troubles with slugs. Management tactics that reduce soil residue have potential to decrease the amount of slug damage.

Natural enemies and biological control

Slugs are preyed upon by a variety of vertebrate and invertebrate natural enemies. Vertebrate predators include frogs, toads, and some snakes (garter snakes, Thamnophis spp.; South 1992); however, it is unlikely that the population densities of these predators are large enough to influence slug populations in agricultural fields. Birds, including poultry, especially ducks and geese, and starlings (Sturnus vulgaris), can eat large numbers of slugs (South 1992, Allen 2004), but some of these species can also cause significant damage to newly sprouted field crops. While these vertebrate predators contribute to slug control, it appears that arthropod predators hold greater potential to suppress slug populations in crop fields. Arthropod predators of slugs include ground beetles (Carabidae), rove beetles (Staphylinidae), firefly larvae (Lampyriidae), marsh flies (Sciomyzidae), harvestmen (Opiliones), wolf spiders (Lycosidae), and centipedes (Chilopoda) (Barker 2004). Certain species of carabid beetles (e.g. Figure 10) appear to be the most significant of these predators in crop fields. For instance in the United Kingdom, the carabid Pterostichus melanarius aggregated in areas of high slug biomass, and ELISA-based gut analysis confirmed that these beetles are significant slug predators (Symondson et al. 1996, Bohan et al. 2000). The potential for ground beetles to suppress slug populations and prevent

18 plant damage has also been amply demonstrated in mesocosm studies (e.g. Asteraki 1993, Oberholzer et al. 2003). However, the significance of arthropod predators to slug suppression in North American crops under field conditions has been little explored. In a recent study in Kentucky strawberries, two of thirteen species of carabids (323 individuals screened) tested positive for presence of slug DNA in their guts in a low slug year (Eskelson et al. 2011). Further studies are needed to identify significant slug predators in North America and their possible contribution to slug management (Thomas et al. 2010). The influence of predators on slugs is reflected in some slug behaviors. For example, slugs appear capable of detecting the presence of ground beetle species that regularly consume slugs, presumably via olfactory cues, and alter their behavior by becoming less active (Armsworth et al. 2005). Under attack, slugs produce copious quantities of defensive mucus, which can gum up the mouthparts of arthropod predators (Mair and Port 2002). Other defense behaviors include “tail-wagging”, descending on a mucus thread (similar to many lepidopteran larvae, Gotwald 1972), and occasional autotomy of the tail (Pakarinen 1994). Of potential slug parasites and pathogens, the most well-known are parasitic nematodes. Ambient levels of slug infection by nematodes are quite low in the United States, and it has even been suggested that release from nematode enemies may be one factor favoring the invasion of North America by European slugs (Ross et al. 2010). In Europe, the species Phasmarhabditis hermaphrodita (Rhabditidae) has been extensively studied and formulated into a biological molluscicide (Nemaslug®) that is now sold in 14 countries (Rae et al. 2007). This nematode is able to infect a range of slug species including D. reticulatum and D. laeve, as well as some but not all slugs in the Arionidae (Grewal et al. 2003). Similar to many entomopathogenic nematodes, this species enters its host as an infective juvenile and is associated with bacteria that are thought to be largely responsible for its pathogenicity (Tan and Grewal 2001a, 2001b). In addition to causing mortality, infection with this nematode rapidly inhibits slug feeding, enhancing its effectiveness in preventing crop damage (Glen et al. 2000). Because P. hermaphrodita is not known to occur naturally in the United States, legislation currently prevents its sale here (Rae et al. 2007). If it were available in the U.S., its high price and short shelf life would likely stifle its economic use in field crops, as it has in Europe (Glen and Symondson 2003), although future improvements may overcome these challenges. Bacteria and other pathogens may be important natural enemies of slugs, but have been little studied (Raut 2004).

19 Management options

Unfortunately, management options for slugs are limited. Moreover, recognized tactics are occasionally ineffective; therefore, an integrated management approach that relies on several control tactics is preferred. Most growers who experience slug problems are committed to no-till or reduced-till practices, so while tillage will certainly help control slugs, it may not be an option. Nevertheless, it is clear that tactics that reduce the amount of surface residue will decrease slug populations. For example, shallow disking (three inches deep) in spring can significantly decrease slug populations (J. F. Tooker and S. Duiker, unpublished data). It is possible that vertical tillage can also provide some relief from slug populations, but we are not aware of empirical work addressing this issue. In addition to providing good habitat for slugs, no-till fields can also harbor improved natural enemy populations when compared to tilled fields (e.g., Witmer et al. 2003); therefore, while no-till fields are prone to slug damage, their stability holds potential to maximize the contribution of predators to improve slug control. These invertebrate predators can be conserved by increasing crop diversity and using insecticides sparingly (e.g., banding insecticides directly over the row rather than broadcasting it over the entire field) in accordance with IPM principles (i.e., use insecticides only when justified economically). Indeed, slug populations have been found to surge after an insecticide application in no-till alfalfa, perhaps due to negative impacts on natural enemies (Grant et al. 1982). Because older crop plants are not as susceptible to slug feeding as young plants, several management tactics aim to foster early plant growth to get crops growing as quickly as possible to try to “outrun” the slug threat. Early planting may give crops a jump on slugs if crops emerge and have significant growth before eggs hatch in large numbers. For instance, early planting in spring can reduce slug damage to new forage stands in Pennsylvania (Byers and Templeton 1988). In contrast to planting early, some growers have tried planting later, after soils are dried and warmed. This approach is meant to encourage quicker germination and growth by the crop during the time when slugs are already active. The choice of early or late planting will likely vary by region, depending on the timing of slug egg hatch relative to crop planting dates. Also, using row cleaners on the front of planters to move crop residue away from the row allows sunlight more access to the soil, increasing soil temperatures and improving crop emergence. However, it has been observed in Ohio that corn fields planted with row cleaners can still have significant slug injury, particularly when slug populations are large because individuals only have to travel a short distance (< 15 inches) from the residue to reach crop plants (R. Hammond,

20 personal communication). Growers can further contribute to better early growth by selecting crop varieties that are rated “excellent” for emergence and seedling vigor. Good agronomic practices such as ensuring seed slots are closed can mitigate some slug damage. The choice of crop rotation (Hammond and Stinner 1987) and cover crop (Vernava et al. 2004) may also influence slug populations and subsequent damage; however more research is needed in these areas. Few chemical controls are available for slugs. Many insecticides, like chlorinated hydrocarbons or organophosphates, do not appear to be toxic to slugs, show inconsistent molluscicidal activity, or require a very large dose to have any influence (Henderson and Triebskorn 2002). Carbamate insecticides, however, can have activity against slugs, and some compounds appear to provide control of slug populations in some settings. Methiocarb has been formulated as a bait, and is used for slug control under the tradename Mesurol® in non-food crop settings such as ornamental production in nurseries and greenhouses (South 1992, Henderson and Triebskorn 2002). Methiocarb was the primary slug control material in the U.S. in the 1980’s, and is still widely used in Europe, but is no longer labeled for field crop use here (R. Hammond, personal communication). Thiodicarb (tradename: Larvin®) was also briefly labeled for use in soybeans (R. Hammond, personal communication). Carbaryl (tradename: Sevin®) has been effective when formulated as a bait, but is ineffective when applied as a spray (South 1992), and does not appear to be labeled for use on slugs in row crops. Another carbamate, methomyl (tradename: Lannate LV®), is currently being explored as a slug control option. In 2010, DuPont (Wilmington, DE) issued in Delaware, Maryland, Pennsylvania, Virginia, and West Virginia a “2(ee) Recommendation” under the Federal Insecticide, Fungicide and Rodenticide Act (FIFRA) for its use in corn and soybeans against slugs, but little efficacy work has been completed thus far. Research conducted in Virginia in 2011 should provide more information (J. Whalen, Univ. of DE, personal communication). Methomyl is known to be toxic to a range of invertebrates and can also alter the behavior of some soil dwelling invertebrates, including earthworms (e.g., Pereira et al. 2009). Neonicotinoid seed treatments, while apparently not lethal to slugs, have shown mixed effects on slug feeding behavior. In the U.K., a clothianidin-based seed treatment (Deter®) is registered in cereals to protect seeds from hollowing by slugs and thereby improve stand establishment. Nonetheless, once seedlings have emerged, neonicotinoids do not appear to reduce slug feeding on leaves (Rose and Oades 2001) and may in some cases even increase slug feeding (Simms et al. 2006). Despite the molluscicidal activity of some insecticides, the primary active ingredient used against slugs is metaldehyde, which is typically formulated into baits (e.g., Deadline® products).

21 Metaldehyde was first developed as a fuel for camp stoves, and in approximately 1934 its molluscicidal activity was accidentally discovered in South Africa (Henderson and Triebskorn 2002). Metaldehyde-based baits were quickly adopted as slug control products because of their selectivity and they still dominate the market (Bailey 2002). There is little doubt that metaldehyde-based baits are effective in controlling slug populations, but a major concern is that most of the baits are somewhat water soluble and rain can diminish their efficacy (Bailey 2002). Additional efficacy issues can arise from crop plants being more attractive than the baits and because individual slugs can stop eating before they receive a lethal dose of metaldehyde (Bailey 2002). Some growers have considered applying baits at planting, but since planting date can coincide fairly closely with slug egg hatch in spring, it is important to ensure that juveniles are present before baits are applied or the baits will lose their effectiveness before the eggs hatch (Hammond et al. 1996). Metaldehyde is also toxic to vertebrates, but recent formulations of metaldehyde-based baits have incorporated mammalian repellents and smaller pellet size, making them less risky to mammals (Bailey 2002). Nevertheless, reports of metaldehyde poisoning (e.g. to dogs) appear to be common in areas where metaldehyde is used regularly (Bailey 2002), although it seems that most cases of poisoning involve animals getting into bags or finding spills on the ground rather than when the product is used according to directions (i.e., broadcast in fields; R. Hammond, personal communication). The focused nature of metaldehyde-based baits may be a benefit to integrated slug management because it allows natural enemies (and other beneficial species like earthworms) to persist in agroecosystems when chemical intervention is necessary (Büchs et al. 1989). While methiocarb and other insecticides can negatively influence populations of some natural enemy species, including carabids and staphylinids (and earthworms) (Bailey 2002), metaldehyde baits have not been reported to cause similar mortality. Pellets based on iron phosphate (e.g., Sluggo products) are also available and are approved by the Organic Materials Review Institute (OMRI) for use in organic systems. These products are also expensive, so use on the large scale typical of field crops may be too costly to be practical. At present, slugs are rarely a problem in organic field crop production since continuous no-till is unusual in these cropping systems, and tillage is likely to prevent significant slug problems in this region. Because slug control can be frustrating, some growers have experimented with home remedies. Chief among these is spraying crops at night when slugs are actively feeding with nitrogen solutions, which act as a contact poison and burn slugs. A common approach is to use a 30% urea-based nitrogen solution, mix it with an equal amount of water, and apply 20 gallons per

22 acre. This tactic is typically repeated a few nights in a row to reach as many slugs as possible and to maximize its effectiveness, and despite potential for burning crop leaves with the high concentration of nitrogen, growers that use this approach believe that the benefits from decreased slug populations outweigh the cost of temporary foliar damage. It should be noted however that there is a wide range of opinion on the efficacy of nitrogen sprays; some growers rely on them whereas others do not believe they are useful (R. Hammond, personal communication). One factor that can undermine this technique is windy conditions that cause slugs to seek shelter out of the reach of nitrogen sprays. Growers using this technique should spray on calm, mild nights when slugs are most likely to be feeding on crop foliage. Other growers will try to control slugs by putting dry ammonium sulfate over their crop rows, generating a salty band that may exclude slugs. Keep in mind that any use of nitrogen or ammonium sulfate needs to comply with a farm’s nutrient management plan. Some farmers with slug problems will choose to use salt-based formulations of herbicides rather than other options with the hope that they might kill or repel slugs. We are not aware of any work evaluating this approach. It appears safe to say that there is not a “silver bullet” for slug problems in no-till crop fields. Many of the tactics discussed above provide some relief under certain circumstances. This inconsistency is problematic; in fact, it is one of the most frustrating features of slug management communicated to us by growers. But inconsistencies can be decreased by employing, in the tradition of IPM, many tactics in concert. For example, we always recommend scouting for slugs to determine where they are problematic. This seems obvious, but our experience is that people infrequently scout for slugs until damage occurs. Even if slugs annually plague certain fields, scout these fields, possibly in fall, but definitely in spring to determine the size of the populations present. Consider the amount of residue the field will have at the time of planting. In fields with a lot of residue, reducing the amount of residue may be prudent if it fits within a grower’s management philosophy. If residue is abundant and slug populations are present, take steps at planting to ensure the crop has the best chance to get up out of the ground quickly. For instance in corn, consider using row cleaners to move debris away from the row. Also, use hybrids rated excellent for emergence and early season vigor, and even a pop-up fertilizer to maximize early season plant growth. Spiked closing wheels may provide some help by ensuring a well-closed seed furrow and good seed-to-soil contact. If in the past these tactics have not seemed to help much, consider adding another tactic, like banding ammonium sulfate over the crop row. If slug populations develop and damage threatens plant survival and stand

23 establishment, be ready to protect your crop with molluscicides. But rather than treating entire fields, consider using metaldehyde-based baits or nitrogren sprays in just the affected areas. No matter the approach, keep records of the growing conditions and what worked. Good records will help refine the most effective approaches to manage slug populations in no-till fields.

24

References

Allen, J. A. 2004. Avian and mammalian predators of terrestrial gastropods, pp. 1-36. In G. M. Barker (ed.), Natural Enemies of Terrestrial Molluscs. CABI Publishing, Oxford, UK. Armsworth, C. G., D. A. Bohan, S. J. Powers, D. M. Glen, and W.O.C. Symondson. 2005. Behavioural responses by slugs to chemicals from a generalist predator. Animal Behaviour 69: 805-811. Asteraki E. J. 1993. The potential of carabid beetles to control slugs in grass-clover swards. Entomophaga 38: 193-198. Bailey, S.E.R. 2002. Molluscidical baits for control of terrestrial gastropods, pp. 33-54. In G. M. Barker (ed.), Molluscs as Crop Pests. CAB International, Wallingford, UK. Barker, G. M. 2004. Natural Enemies of Terrestrial Molluscs. CABI Publishing, Oxford, UK. Barr, N. B., A. Cook, P. Elder, J. Molongoski, D. Prasher, and D. G. Robinson. 2009. Application of a DNA barcode using the 16S rRNA gene to diagnose pest Arion species in the USA. Journal of Molluscan Studies 75: 187-191. Barratt, B.I.P., R. A. Byers, and D. L. Bierlein. 1994. Conservation tillage crop yields in relation to grey garden slug [Deroceras reticulatum (Müller)] (Mollusca: Agriolimacidae) density during establishment. Crop Protection 13: 49-52. Beyer, W. N., and D. M. Saari. 1978. Activity and ecological distribution of the slug, Arion subfuscus (Draparnaud). American Midland Naturalist 100: 359-367. Bohan, D. A., A. C. Bohan, D. M. Glen, W.O.C. Symondson, C. W. Wiltshire, and L. Hughes. 2000. Spatial dynamics of predation by carabid beetles on slugs. Journal of Animal Ecology 69: 367-379. Brooks, A. S., M. J. Crook, A. Wilcox, and R. T. Cook. 2003. A laboratory evaluation of the palatability of legumes to the field slug, Deroceras reticulatum Müller. Pest Management Science 59: 245-251. Brooks, A. S., A. Wilcox, R. T. Cook, and M. J. Crook. 2005. A laboratory-based comparison of a molluscicide and an alternative food source (red clover) as means of reducing slug damage to winter wheat. Pest Management Science 61: 715-20.

25

Büchs, W., U. Heimbach, and E. Czarnecki. 1989. Effects of snail baits on non-target carabid beetles, pp. 245-252. In I. Henderson (ed.) Slugs and Snails in World Agriculture. BCPC Monograph No. 41. British Crop Protection Council, UK. Byers, R. A. 2002. Agriolimacidae and Arionidae as pests in lucerne and other legumes in forage systems of north-eastern North America, pp. 325-335. In G. M. Barker (ed.), Molluscs as Crop Pests. CAB International, Wallingford, UK. Byers, R. A., and D. L. Bierlein. 1982. Feeding preferences of three slug species in the laboratory. Melsheimer Entomological Series 32: 5-11. Byers, R. A., and D. D. Calvin. 1994. Economic injury levels to field corn from slug (Stylommatophora: Agrolimacidae) feeding. Journal of Economic Entomology, 87: 1345- 1350. Byers, R. A., and W. C. Templeton, Jr. 1988. Effects of sowing date, placement of seed, vegetation suppression, slugs, and insects upon establishment of no-till alfalfa in orchardgrass sod. Grass and Forage Science 43: 279-289. Byers, R. A., R. L. Mangan, and W. C. Templeton, Jr. 1983. Insect and slug pests in forage legume seedings. Journal of Soil and Water Conservation 38: 224-226. Cates, R. G., and G. H. Orians. 1975. Successional status and the palatability of plants to generalized herbivores. Ecology 56: 410-418. Chichester, L. F., and L. L. Getz. 1969. The zoogeography and ecology of arionid and limacid slugs introduced into northeastern North America. Malacologia 7: 313-346. Chichester, L. F., and L. L. Getz. 1973. The terrestrial slugs of northeastern North America. Sterkiana 51: 11-42. Cook, R. T., S.E.R. Bailey, and C. R. McCrohan. 1996. Slug preferences for winter wheat cultivars and common agricultural weeds. Journal of Applied Ecology, 33: 866-872. Cook, R. T., S.E.R. Bailey, and C. R. McCrohan. 1997. The potential for common weeds to reduce slug damage to winter wheat: laboratory and field studies. Journal of Applied Ecology 34: 79-87. Dirzo, R. 1980. Experimental studies on slug-plant interactions I. The acceptability of thirty plant species to the slug Agriolimax caruanae. Journal of Ecology 68: 981-998. Duthoit, C.M.G. 1964. Slugs and food preferences. Plant Pathology 13: 73-78. Eskelson, M. J., E. G. Chapman, D. D. Archbold, J. J. Obrycki, and J. D. Harwood. 2011. Molecular identification of predation by carabid beetles on exotic and native slugs in a strawberry agroecosystem. Biological Control 56: 245-253.

26

Fox, L., and B. J. Landis. 1973. Notes on the predaceous habits of the gray field slug, Deroceras laeve. Environmental Entomology 2: 306-307. Geenen, S., K. Jordaens, and T. Backeljau. 2006. Molecular systematics of the Carinarion complex (Mollusca: Gastropoda: Pulmonata): a taxonomic riddle caused by a mixed breeding system. Biological Journal of the Linnaean Society 89: 589-604. Getz, L. L. 1959. Notes on the ecology of slugs: Arion circumscriptus, Deroceras reticulatum, and D. laeve. American Midland Naturalist 61: 485-498. Glen, D. M., and W.O.C. Symondson. 2003. Influence of soil tillage on slugs and their natural enemies, pp. 207-228. In A. El Titi (ed.) Soil Tillage in Agroecosystems. CRC Press, Boca Raton, FL. Glen, D. M., H. Jones, and J. K. Fieldsend. 1990. Damage to oilseed rape seedlings by the field slug Deroceras reticulatum in relation to glucosinolate concentration. Annals of Applied Biology 117: 197-207. Glen, D. M., M. J. Wilson, P. Brain, and G. Stroud. 2000. Feeding activity and survival of slugs, Deroceras reticulatum, exposed to the rhabditid nematode, Phasmarhabditis hermaphrodita: a model of dose response. Biological Control 17: 73-81. Godan, D. 1983. Pest Slugs and Snails: Biology and Control. Springer-Verlag, New York, NY. Gotwald Jr., W. H. 1972. Analogous prey escape mechanisms in a pulmonate mollusk and lepidopterous larvae. Journal of the New York Entomological Society 80: 111-113. Gouyon, P. H., P. Port, and G. Caraux. 1983. Selection of seedlings of Thymus vulgaris by grazing slugs. Journal of Ecology, 71: 299-306. Grant, J. F., K. V. Yeargan, B. C. Pass, and J. C. Parr. 1982. Invertebrate organisms associated with alfalfa seedling loss in complete-tillage and no-tillage plantings. Journal of Economic Entomology 75: 822-826. Gregory, W. W., and G. J. Musick. 1976. Insect management in reduced tillage systems. Bulletin of the Entomological Society of America 22: 302-304. Grewal, S. K., P. S. Grewal, and R. B Hammond. 2003. Susceptibility of North American native and non-native slugs (Mollusca: Gastropoda) to Phasmarhabditis hermaphrodita (Nematoda: Rhabditidae). Biocontrol Science and Technology 13: 119-125. Grover, K. K., H. D. Karsten, and G. W. Roth. 2009. Corn grain yields and yield stability in four long-term cropping systems. Agronomy Journal 101: 940-946. Hammond, R. B. 2000. Simulation of moderate levels of slug injury to soybean. Crop Protection 19: 113-120.

27

Hammond, R. B., and R. A. Byers. 2002. Agriolimacidae and Arionidae as pests in conservation- tillage soybean and maize cropping in North America, pp. 301-314. In G. M. Barker (ed.), Molluscs as Crop Pests. CAB International, Wallingford, UK. Hammond, R. B., and B. R. Stinner. 1987. Seedcorn maggots and slugs in conservation tillage systems in Ohio, USA. Journal of Economic Entomology 80: 680-684. Hammond, R. B., J. A. Smith, and T. Beck. 1996. Timing of molluscicide applications for reliable control in no-tillage field crops. Journal of Economic Entomology 89: 1028- 1032. Hammond, R. B., T. Beck, J. A. Smith, R. Amos, J. Barker, R. Moore, H. Siegrist, D. Slates and B. Ward. 1999. Slugs in conservation tillage corn and soybeans in the eastern Corn Belt. Journal of Entomological Science 34: 467–478. Henderson, I., and R. Triebskorn. 2002. Chemical control of terrestrial gastropods, pp. 1-32. In G. M. Barker (ed.), Molluscs as Crop Pests. CAB International, Wallingford, UK. Hommay, G., J. C. Kienlen, F. Jacky and C. Gertz. 2003. Daily variation in the number of slugs under refuge traps. Annals of Applied Biology 142: 333-339. Horowitz, J., R. Ebel, and K. Ueda. 2010. “No-Till” farming is a growing practice. Economic Information Bulletin No. 70. USDA-ERS, Washington, DC. Hunter, P. J. 1968. Studies on slugs of arable ground, I. Sampling methods. Malacologia 6: 369- 377. Hunter, P. J., and B. V. Symonds. 1971. The leap-frogging slug. Nature 229: 349. Jennings, T. J., and J. P. Barkham. 1975. Food of slugs in mixed deciduous woodland. Oikos 26: 211-221. Kerney, M. P., and R.A.D. Cameron. 1979. A Field Guide to the Land Snails of Britain and North-West Europe. Collins, St. James Place, London, UK. Kozlowski, J., and M. Kozlowska. 2004. Food preferences of Deroceras reticulatum, Arion luscitanicus and Arion rufus for various medicinal herbs and oilseed rape. Journal of Plant Protection Research 44: 239-249. Lundgren, J. G., J. T. Shaw, E. R. Zaborski, and C. E. Eastman. 2006. The influence of organic transition systems on beneficial ground-dwelling arthropods and predation of insects and weed seeds. Renewable Agriculture and Food Systems. 21: 227 – 237. Mair, J., and G. R. Port. 2002. The influence of mucus production by the slug, Deroceras reticulatum, on predation by Pterostichus madidus and Nebria brevicollis (Coleoptera: Carabidae). Biocontrol Science and Technology 12: 325-335.

28

McCracken, G. F., and R. K. Selander. 1980. Self-fertilization and monogenic strains in natural populations of terrestrial slugs. Proceedings of the National Academy of Sciences. 77: 684-688. Mellanby, K. 1961. Slugs at low temperatures. Nature 189: 944. Miles, H. W., J. Wood, and I. Thomas. 1931. On the ecology and control of slugs. Annals of Applied Biology 18: 370-400. Molgaard, P. 1986. Food plant preferences by slugs and snails: a simple method to evaluate relative palatability of the food plants. Biochemical Systematics and Ecology 14: 113- 121. Oberholzer, F., N. Escher, and T. Frank. 2003. The potential of carabid beetles (Coleoptera) to reduce slug damage to oilseed rape in the laboratory. European Journal of Entomology. 100: 81-85. Pallant, D. 1972. The food of the grey field slug, Agriolimax reticulatus (Müller), on grassland. Journal of Animal Ecology 41: 761-769. Pakarinen, E. 1994. Autotomy in arionid and limacid slugs. Journal of Molluscan Studies 60: 19- 23. Pearce, T. A. 2008. Land snails of limestone communities and update of land snail distributions in Pennsylvania. Final Report for Grant Agreement WRCP-04016. Carnegie Museum of Natural History, Pittsburgh, PA. Pereira, J. L., S. C. Antunes, B. B. Castro, C. R. Marques, A.M.M. Gonçalves, F. Gonçalves, and R. Pereira. 2009. Toxicity evaluation of three pesticides on non-target aquatic and soil organisms: commercial formulation versus active ingredient. Ecotoxicology 18: 455-63. Peters, H. A., B. Baur, F. Bazzaz, and C. Körner. 2000. Consumption rates and food preferences of slugs in a calcareous grassland under current and future CO2 conditions. Oecologia 125: 72-81. Pinceel, J., K. Jordaens, N. Van Houtte, A. J. de Winter, and T. Backeljau. 2004. Molecular and morphological data reveal cryptic taxonomic diversity in the terrestrial slug complex Arion subfuscus/fuscus (Mollusca, Pulmonata, Arionidae) in continental north-west Europe. Biological Journal of the Linnaean Society 83: 23-38. Pinceel, J., K. Jordaens, N. Van Houtte, G. Bernon, and T. Backeljau. 2005. Population genetics and identity of an introduced terrestrial slug: Arion subfuscus s.l. in the north-east USA (Gastropoda, Pulmonata, Arionidae). Genetica 125: 155-171.

29

Port, C. M., and G. R. Port. 1986. The biology and behavior of slugs in relation to crop damage and control. Agricultural Zoology Reviews 1: 255-297. Rae, R. G., C. Verdun, P. S. Grewal, J. F. Robertson, and J. M. Wilson. 2007. Biological control of terrestrial molluscs using Phasmarhabditis hermaphrodita – progress and prospects. Pest Management Science 63: 1153-1164. Rathcke, B. 1985. Slugs as generalist herbivores: tests of three hypotheses on plant choices. Ecology 66: 828-836. Raut, S. K. 2004. Bacterial and non-microbial diseases in terrestrial gastropods, pp. 599-612. In G. M. Barker (ed.) Natural Enemies of Terrestrial Molluscs. CABI Publishing, Oxford, UK. Rollo, C. D., and D. M. Shibata. 1991. Resilience, robustness, and plasticity in a terrestrial slug, with particular reference to food quality. Canadian Journal of Zoology 69: 978-987. Rose, P., and L. Oades. 2001. Effects of imidacloprid cereal seed treatment against wireworms and slugs. BCPC Symposium Proceeding: Challenges and Opportunities 76:191-196. Ross, J. L., E. S. Ivanova, P. M. Severns, and M. J. Wilson. 2010. The role of parasite release in invasion of the USA by European slugs. Biological Invasions 12: 603-610. Schrim M., and R. A. Byers. 1980. A method for sampling three slug species attacking sod- seeded legumes. Melsheimer Entomological Series 29: 9-11. Simms, L. C., A. Ester, and M. J. Wilson. 2006. Control of slug damage to oilseed rape and wheat with imidacloprid seed dressings in laboratory and field experiments. Crop Protection 25: 549-555. South, A. 1964. Estimation of slug populations. Annals of Applied Biology 53: 251-258. South, A. 1965. Biology and ecology of Agriolimax reticulatus (Müll.) and other slugs: spatial distribution. Journal of Animal Ecology 34: 403-417. South, A. 1992. Terrestrial Slugs: Biology, Ecology and Control. Chapman & Hall, London, UK. Symondson, W.O.C., D. M. Glen, C. W. Wiltshire, C. J. Langdon, and J. E. Liddell. 1996. Effects of cultivation techniques and methods of straw disposal on predation by Pterostichus melanarius (Coleoptera: Carabidae) upon slugs (Gastropoda: Pulmonata) in an arable field. Journal of Applied Ecology 33: 741-753. Tan, L., and P. S. Grewal. 2001a. Pathogenicity of Moraxella osloensis, a bacterium associated with the nematode Phasmarhabditis hermaphrodita, to the slug Deroceras reticulatum. Applied Environmental Microbiology 67: 5010-5016.

30

Tan, L., and P. S. Grewal. 2001b. Infection behavior of the rhabditid nematode Phasmarhabditis hermaphrodita to the grey garden slug Deroceras reticulatum. Journal of Parasitology 87: 1349-1354. Theenhaus, A., and S. Scheu. 1996. The influence of slug (Arion rufus) and cast material addition on microbial biomass, respiration, and nutrient cycling in beech leaf litter. Biology and Fertility of Soils 23: 80-85. Thomas, A. K., R. J. McDonnell, and J. D. Harwood. 2010. Slugs from the Nearctic: what we need to learn from the Western Palearctic. In: Proceedings of the IOBC/WPRS workgroup on Slugs and Snails: Slug and Snail Control in the 21st Century. U.S. Department of Agriculture-Natural Resources Conservation Service. 2011. Assessment of the effects of conservation practices on cultivated cropland in the Chesapeake Bay Region. USDA, Washington, DC. U.S. Department of Agriculture-National Agricultural Statistics Service. 2009. Pennsylvania Agricultural Statistics 2008-2009. USDA, Harrisburg, PA. Vernava, M. N., P. M. Phillips-Aalten, L. A. Hughes, H. Rowcliffe, C. W. Wiltshire, and D. M. Glen. 2004. Influences of preceding cover crops on slug damage and biological control using Phasmarhabditis hermaphrodita. Annals of Applied Biology, 145: 279-284. Vorst, J. J. 1986. Assessing hail damage to corn. Institute of Agriculture and Natural Resources, Cooperative Extension Service. University of Nebraska NebGuide 803. Willis, J. C., D. A. Bohan, Y. H. Choi, K. F. Conrad, and M. A. Semenov. 2006. Use of an individual-based model to forecast the effect of climate change on the dynamics, abundance and geographical range of the pest slug Deroceras reticulatum in the UK. Global Change Biology 12: 1643-1657. Willis, J. C., D. A. Bohan, S. J. Powers, Y. H. Choi, J. Park, and E. Gussin. 2008. The importance of temperature and moisture to the egg-laying behavior of a pest slug, Deroceras reticulatum. Annals of Applied Biology 153: 105-115. Willson, H. R., and J. B. Eisley. 1992. Effects of tillage and prior crop on the incidence of five key pests on Ohio corn. Journal of Economic Entomology 85: 853 – 859. Witmer J. E., J. A. Hough-Goldstein, and J. D. Pesek. 2003. Ground-dwelling and foliar arthropods in four cropping systems. Environmental Entomology 32: 366-376.

31

Figures

Figure 1-1. Gray garden slug (Deroceras reticulatum) (photo: Margaret Douglas, PSU)

Figure 1-2. Marsh slug (Deroceras laeve) (photo: Margaret Douglas, PSU)

32

Figure 1-3. Dusky slug (Arion subfuscus) (photo: courtesy of Nick Sloff, PSU)

Figure 1-4. Banded slug (Arion fasciatus) (photo: courtesy of Nick Sloff, PSU)

33

Figure 1-5. Slug eggs in the soil (photo: courtesy of Nick Sloff, PSU)

Figure 1-6. Slug damage to corn (photo: Margaret Douglas, PSU)

34

Figure 1-7. Juvenile grey garden slug (Deroceras reticulatum) on soybean (photo: courtesy of Nick Sloff, PSU)

Figure 1-8. Slug damage to canola (photo: Margaret Douglas, PSU)

35

Figure 1-9. Roofing material used as an artificial slug shelter (photo: Margaret Douglas, PSU)

Figure 1-10. Chlaenius tricolor, a slug-eating beetle (photo: courtesy of Ian Grettenberger, PSU)

Chapter 2

A preliminary laboratory study of slug predation by common generalist predators found in Pennsylvania crop fields

Introduction

Despite the rising importance of slugs as pests in no-till cropping systems (e.g. Byers 2002, Hammond and Byers 2002), surprisingly little is known about their natural enemies in North American cropping systems (Thomas et al. 2011). The slug assemblage in Northeastern and mid-Atlantic crop fields typically comprises four species, but is dominated by the introduced gray garden slug, Deroceras reticulatum (Byers 2002, Hammond and Byers 2002, Appendix A). The marsh slug D. laeve, a native congener of D. reticulatum, also occurs in crop fields but is often not as abundant as D. reticulatum, particularly during times of crop establishment (Byers, Barratt, and Calvin 1989, Appendix A). The banded slug, Arion fasciatus, and dusky slug complex, A. subfuscus, both introduced species, can also be present in crop fields although they are rarely associated with crop damage (Hammond and Byers 2002). Numerous epigeal and endogeal arthropod predators co-exist with these slug species, including wolf spiders (Lycosidae) and harvestmen (Opiliones), adult and larval ground (Carabidae) and rove beetles (Staphylinidae), and larval fireflies (Lampyridae) and soldier beetles (Cantharidae). Some members of these taxa have been documented to prey on slugs in other regions (Barker 2004), but their predatory potential in Pennsylvania crop fields is uncertain. To explore the potential for some common ground-dwelling predators in Pennsylvania to prey on D. reticulatum and D. laeve, I conducted preliminary laboratory experiments in autumn 2010. To measure predation, I used tritrophic arenas stocked with soybean seedlings, slugs, and predators. Including plants in the experiment provided insight on whether predators influence slug feeding behavior in addition to how often predators consume slugs outright. Predators that I investigated included wolf spiders (Trochosa sp.), soldier beetle larvae (most likely Chauliognathus sp.), and the ground beetles Pterostichus melanarius and Chlaenius tricolor. Pterostichus melanarius is an introduced species that is known to prey on D. reticulatum in their shared native range in Europe (Symondson et al. 1996, Bohan et al. 2000), and can be quite

37

abundant in Pennsylvania crop fields (Leslie et al. 2007). Chlaenius tricolor is a native species that was found in large numbers in soybean fields in early fall when slug activity was also high, prompting inclusion in this study. Wolf spiders and soldier beetle larvae are abundant in autumn on the soil surface in crop fields and so were included here. In a small follow-up study, I investigated whether C. tricolor shows a preference for the native species D. laeve versus the introduced species D. reticulatum or A. fasciatus.

Materials and methods

Soybean seeds, slugs, and predators

All soybean seeds used in this study were certified organic (variety: Viking 0-2265) and untreated with fungicides or insecticides. Gray garden (D. reticulatum), marsh (D. laeve), and banded (A. fasciatus) slugs were collected in autumn 2010 from old fields, wetlands, and crop fields in the vicinity of State College, PA. Slugs were then kept in the laboratory at room temperature, fed cabbage ad libitum and kept in plastic boxes that were lined with a layer of moist potting soil. I collected predators by hand or with dry pitfall traps in corn, alfalfa, and soybean fields at the Russell E. Larson Agricultural Research Farm at Rock Springs (Pennsylvania Furnace, PA). Following collection in the morning, predators were kept in individual cups with moist paper towels in a growth chamber (21o C, 12:12 L:D), until being used in experiments the following day. In the following text, I report mean ± SD unless otherwise noted.

Slug predation experiment

To determine whether common generalist predators will consume slugs and protect plants from slug damage, I performed laboratory experiments with soybean plants, slugs, and predators. To generate an even arena surface and facilitate slug recovery, dry soil (Premier ® Pro-Mix® BX) was sifted through a 2mm screen, moistened with water (two parts water to one part soil by mass), and added to 16-oz clear plastic containers (Reynolds Del-Pak®) to a depth of ~2.5 cm. In this soil, I planted four, evenly spaced soybean seeds. Sides of the containers were coated with Fluon (BioQuip Products, Inc., Rancho Dominguez, CA) to keep slugs in the arena where

38

predators could attack them (Symondson 1993). Each container received either an individual of D. reticulatum (0.12 ± 0.03 g) or an individual of D. laeve (0.11 ± 0.04 g). Predators were then randomly assigned to containers. Control containers with a slug but no predator were included to measure the influence of slugs on soybeans in the absence of predation (see Table 1 for sample sizes). Containers were kept in a growth chamber for four days (21o C, 12:12 L:D), with the status of plants (undamaged/damaged), slugs (alive/dead), and predators (alive/dead) recorded each day. A plant was considered damaged if there was any slug feeding on the stem, cotyledons, or leaves. At the end of the experiment, slugs were re-weighed.

Chlaenius tricolor choice study

To investigate whether the native ground beetle C. tricolor shows a preference for the native D. laeve over either of the introduced species, D. reticulatum or A. fasciatus, I conducted a small choice study. Containers were prepared as described in the slug predation experiment, except that no soybean seeds were included and soil was added only to a depth of ~1cm. Each container was assigned two size-matched slugs, either one D. laeve and one D. reticulatum (0.19 ± 0.04 g), or one D. laeve and one A. fasciatus (0.15 ± 0.04 g). Adults of C. tricolor were randomly assigned to containers (n = 6 for D. laeve vs. D. reticulatum; n = 5 for D. laeve vs. A. fasciatus). I introduced predators at night, observed containers for the first two hours, and then put them in a growth chamber for five days (21o C, 12:12 L:D). Each day the status of slugs (alive/dead) and predators (alive/dead) were recorded.

Statistical analyses

In the C. tricolor choice study, samples sizes (n = 5 to 6) were too small for formal statistical tests. Instead I report results in graphical form for qualitative interpretation. Unless otherwise noted, I performed all analyses in Minitab version 16 (Minitab, Inc., State College, PA). Using ANOVA, I compared among predator treatments the number of soybean seedlings damaged by slugs at the end of four days, with slug species, predator taxa, and their interaction as fixed factors. I separated means using Tukey’s test when significant effects were detected. Predation on the two slug species was similar across predator treatments and the

39

small sample sizes (n = 4 to 10) precluded an investigation of slug species as a factor in slug mortality. Therefore, I pooled the data across slug species to test for differences in slug survival by predator treatment, using a G-test of independence with the Williams (1976) adjustment (Gotelli and Ellison 2004). This was followed by post-hoc Fisher’s Exact Tests using VassarStat interactive software (http://vassarstats.net; Lowry, Pouphkeepsie, NY) to compare slug survival in each pair of predator treatments, correcting for multiple comparisons using the Bonferroni procedure. To test whether predators influenced slug feeding behavior, I performed a separate analysis on slugs that survived enclosure with predators. ANCOVA was used to examine whether slug species, predator treatment, or their interaction changed the relationship between slug starting and ending mass. Ending mass was log transformed to linearize the relationship between starting and ending mass. Tukey’s test separated means when significant effects were found.

Results

The number of soybean seedlings damaged by slugs varied significantly among predator treatments, but not slug species (Slug species effect: F1,56 = 0.02, P = 0.89; Predator effect: F4,56 =

19.05, P < 0.001; Slug species*Predator effect: F4,56: 2.05, P = 0.10). Slug damage to soybeans was lower in the presence of P. melanarius and C. tricolor compared to no predator, cantharid larva, and Trochosa (Figure 1). Slug survival varied among predator treatments (Gadj = 40.4, d.f. = 4, P < 0.001) and was lowest with C. tricolor, intermediate with P. melanarius, and highest with no predator, cantharid, or Trochosa (Figure 2). Notably, mass of slugs surviving predator exposure varied among predator treatments

(Slug species effect: F1,35 = 0.06, P = 0.80; Predator effect: F3,35 = 7.20, P = 0.001; Slug species*Predator effect: F3,35: 1.21, P = 0.32). Slugs surviving encounters with P. melanarius gained significantly less mass than those in other treatments (Figure 3). In fact, four out of five slugs surviving with P. melanarius lost mass over the four days of the experiment. In the choice study, C. tricolor preyed on D. laeve and D. reticulatum, but did not prey on A. fasciatus (Figure 4). It appeared that C. tricolor had a slight preference for D. laeve over D. reticulatum, although small sample size precludes formal testing of this hypothesis. In observations of C. tricolor shortly after slugs were introduced, roughly half of D. laeve were attacked and killed within two hours of introduction (Figure 4). Defensive behaviors of D. laeve against beetle attack included exudations of mucus and, occasionally, violent spasms of the

40

posterior part of the animal. Chlaenius tricolor followed such attacks by cleaning its mouthparts, legs, and antennae. It often took C. tricolor multiple attempts to successfully subdue its slug prey.

Discussion

In laboratory assays, the ground beetles P. melanarius and C. tricolor preyed upon D. reticulatum and D. laeve, protecting soybean seedlings from damage. For P. melanarius, this result is consistent with European field studies, where this species preyed upon D. reticulatum in crop fields and aggregated in areas of high slug abundance, suggesting this species plays a role in suppressing slug populations (Symondson et al. 1996, Bohan et al. 2000, Symondson et al. 2002). Pterostichus melanarius has been accidentally introduced to both the east and west coasts of North America (Lindroth 1957), and is now a dominant or subdominant member of carabid assemblages in many North American agroecosystems (e.g. Carmona and Landis 1999, Bourassa et al. 2008), including in Pennsylvania (Leslie et al. 2007, Leslie et al. 2009). Given the prevalence of this beetle, further studies on its interactions with slugs in North American cropping systems appear warranted. Furthermore, slugs that survived encounters with P. melanarius weighed less than control slugs, or even lost mass. This result helps to explain why P. melanarius protected seedlings as well as C. tricolor, even though numerically fewer slugs were killed by P. melanarius. Deroceras reticulatum has previously been shown to avoid paper infused with P. melanarius scent (Armsworth et al. 2005); thus the reduced slug feeding I observed in the presence of P. melanarius may have resulted from predator-avoidance behavior by slugs. These laboratory results are consistent with recent research in a range of predator-prey systems highlighting the importance of behavior in mediating the ability of predators to protect plants from herbivore damage (Schmitz et al. 1997, Werner and Peacor 2003, Preisser et al. 2005). Though the present study was preliminary in nature, this is an intriguing area for future studies on slug-predator interactions. Along with P. melanarius, C. tricolor preyed on both D. reticulatum and D. laeve, with only one individual slug in 14 trials surviving the encounter. This result agrees with recent work in which the guts of a third of field-collected C. tricolor tested positive for the remains of D. reticulatum or D. laeve—which was the highest rate of slug predation out of thirteen carabid species tested (Eskelson et al. 2011). In the same study, the two slug species were consumed by

41

C. tricolor with similar frequency, suggesting that this species does not show a preference for D. laeve over D. reticulatum in the field. It is notable that in the current study, C. tricolor never preyed on A. fasciatus, and in fact was never observed even attempting to attack this species. There is scant literature on the basic ecology of C. tricolor, although feeding on molluscs and other animal prey has been previously recorded for this species (Larochelle 1974). Future studies of C. tricolor and other North American carabids could aid in the development of conservation biological control for slugs. Wolf spiders and soldier beetle larvae are prominent predators in soil agroecosystems, where they occur in proximity to slugs. However, in these laboratory studies cantharid larvae did not prey upon slugs, nor did they prevent slug damage to soybean seedlings. These results agree with others that failed to find slug proteins in field-collected Cantharis larvae, and found that these larvae were also reluctant to take D. reticulatum as food in the laboratory (Traugott 2003). As for Trochosa, my finding of low-level slug predation is consistent with previous studies of Lycosidae (Nyffeler and Symondson 2001, Fountain et al. 2005). Given that these taxa failed to attack slugs in enclosed laboratory containers, it seems unlikely that they play an important role in slug predation in the field.

Conclusion

Preliminary studies in the laboratory have identified two common ground beetle species, Pterostichus melanarius and Chlaenius tricolor, as potentially important slug predators in Pennsylvania. These results are consistent with studies of slug predation in Europe and growing research along these lines in North America. Further studies are needed to investigate this potential under field conditions and to examine other predatory taxa for their ability to contribute to slug suppression.

42

References

Armsworth, C. G., D. A. Bohan, S. J. Powers, D. M. Glen, and W.O.C. Symondson. 2005. Behavioral responses by slugs to chemicals from a generalist predator. Animal Behavior 69(4): 805-811. Barker, G. M. 2004. Natural Enemies of Terrestrial Molluscs. CABI Publishing, Oxford, UK. Bohan, D. A., A. C. Bohan, D. M. Glen, W.O.C. Symondson, C. W. Wiltshire, and L. Hughes. 2000. Spatial dynamics of predation by carabid beetles on slugs. Journal of Animal Ecology 69(3): 367-379. Bourassa, S., H. A. Carcamo, F. L. Larney, and J. R. Spence. 2008. Carabid assemblages (Coleoptera: Carabidae) in a rotation of three different crops in southern Alberta, Canada: a comparison of sustainable and conventional farming. Environmental Entomology 37(5): 1214-1223. Byers, R. A. 2002. Agriolimacidae and Arionidae as pests in lucerne and other legumes in forage systems of north-eastern North America, pp. 325-335. In G. M. Barker (ed.), Molluscs as Crop Pests. CAB International, Wallingford, UK. Byers, R. A., B. I. P. Barratt, and D. Calvin. 1989. Comparison between defined-area traps and refuge traps for sampling slugs in conservation tillage crop environments. In: I. F. Henderson, ed. Slugs and Snails in World Agriculture, BCPC Monograph 41, 187-192. Carmona, D. M. and D. A. Landis. 1999. Influence of refuge habitats and cover crops on seasonal activity-density of ground beetles (Coleoptera: Carabidae) in field crops. Environmental Entomology 28(6): 1145-1153. Eskelson, M. J., E. G. Chapman, D. D. Archbold, J. J. Obrycki, and J. D. Harwood. 2011. Molecular identification of predation by carabid beetles on exotic and native slugs in a strawberry agroecosystem. Biological Control 56: 245-253. Fountain, M. T., R. S. Thomas, V. K. Brown, A. C. Gange, P. J. Murray, and W. O. C. Symondson. 2009. Effects of nutrient and insecticide treatments on invertebrate numbers and predation on slugs in an upland grassland: a monoclonal antibody-based approach. Agriculture, Ecosystems and Environment 131: 145-153.

43

Hammond, R. B., and R. A. Byers. 2002. Agriolimacidae and Arionidae as pests in conservation- tillage soybean and maize cropping in North America, pp. 301-314. In G. M. Barker (ed.), Molluscs as Crop Pests. CAB International, Wallingford, UK. Larochelle, A. 1974. A world list of prey of Chlaenius (Coleoptera: Carabidae). Great Lakes Entomologist 7: 147-148. Leslie, T. W., G. A. Hoheisel, D. J. Biddinger, J. R. Rohr, and S. J. Fleischer. 2007. Transgenes sustain epigeal insect biodiversity in diversified vegetable farm systems. Environmental Entomology 36(1): 234-244. Leslie, T. W., D. J. Biddinger, C. A. Mullin, and S. J. Fleischer. 2009. Carabidae population dynamics and temporal partitioning: Response to couples neonicotinoid-transgenic technologies in maize. Environmental Entomology 38(3): 935-943. Lindroth, C. H. 1957. The Faunal Connections between Europe and North America. John Wiley and Sons, New York, NY. Nyffeler, M. and W. O. C. Symondson. 2001. Spiders and harvestmen as gastropod predators. Ecological Entomology 26: 617-628. Preisser, E. L., D. I. Bolnick, and M. F. Bernard. 2005. Scared to death? The effects of intimidation and consumption in predator-prey interactions. Ecology 86(2): 501-509. Schmitz, O. J., A. P. Beckerman, and K. M. O’Brien. 1997. Behaviorally mediated trophic cascades: effects of predation risk on food web interactions. Ecology 78(5): 1388-1399. Symondson, W.O.C. 1993. Chemical confinement of slugs: an alternative to electric fences. Journal of Molluscan Studies 59: 259-261. Symondson, W.O.C., D. M. Glen, C. W. Wiltshire, C. J. Langdon, and J. E. Liddell. 1996. Effects of cultivation techniques and methods of straw disposal on predation by Pterostichus melanarius (Coleoptera: Carabidae) upon slugs (Gastropoda: Pulmonata) in an arable field. Journal of Applied Ecology 33(4): 741 – 753. Symondson, W.O.C., D. M. Glen, A. R. Ives, C. J. Langdon, and C. W. Wiltshire. 2002. Dynamics of the relationship between a generalist predator and slugs over five years. Ecology 83(1): 137-147. Thomas, A.K., R.J. McDonnell, and J.D. Harwood. 2011. Slugs from the Nearctic: what we need to learn from the Western Palearctic. Proceedings of the IOBC/WPRS workgroup on Slugs and Snails: Slugs and Snail Control in the 21st Century. Traugott, M. 2003. The prey spectrum of larval and adult Cantharis species in arable land: an electrophoretic approach. Pedobiologia 47: 161-169.

44

Werner, E. E. and S. D. Peacor. 2003. A review of trait-mediated indirect interactions in ecological communities. Ecology 84(5): 1083-1100.

45

Tables

Table 2-1. Sample sizes for slug predation study.

Deroceras Deroceras Total reticulatum laeve No predator control 10 9 19 Cantharid larva 4 5 9 Trochosa sp. 6 7 13 Pterostichus melanarius 6 5 11 Chlaenius tricolor 7 7 14

46

Figures

Figure 2-1. Number of plants (out of four) damaged by slugs (mean ± SE) after four days of exposure to slugs and each of four predator species and a no-predator control (n = 9 - 19). These data include pooled data from both D. reticulatum and D. laeve. Bars that do not share a letter differ from each other at P ≤ 0.05 based on Tukey’s test.

Figure 2-2. Slug survival (proportion, pooled across slug species) following a four-day assay with different predator species (n = 9 - 19). Bars that do not share a letter differ from each other at P ≤ 0.05 based on Fisher’s Exact Tests with a Bonferroni correction.

47

-0.5

-0.6 )

) -0.7

g

(

t h

g -0.8

i

e w -0.9 Cantharid larva (A)

g Trochosa (A)

n i

d No predator (A)

n -1.0 Pterostichus (B)

e

g u

l -1.1

s

(

g o

L -1.2

-1.3

-1.4 0.050 0.075 0.100 0.125 0.150 0.175 Slug starting weight (g)

Figure 2-3. Relationship between slug starting mass and log-transformed ending mass for slugs surviving four days in the presence or absence of predators, pooled by slug species (n = 18 for no predator control; n = 9 for cantharid larva; n = 12 for Trochosa; n = 5 for Pterostichus). Only one slug survived in the presence of Chlaenius and so it was not included here. Treatments followed by different letters in the key differ from each other at P ≤ 0.05 based on Tukey’s test.

Figure 2-4. Survival of D. laeve, D. reticulatum, and A. fasciatus in choice arenas with C. tricolor (n = 6 for D. laeve vs. D. reticulatum and n = 5 for D. laeve vs. A. fasciatus).

Chapter 3

Dynamics of pest and predatory invertebrates in reduced-tillage maize (Zea mays L.) cropping systems in Pennsylvania

Introduction

Conservation tillage has become increasingly common in recent decades, particularly in the mid-Atlantic and Northeastern U.S. where the rolling topography renders many fields prone to soil erosion, enhancing the benefits of these techniques. Additionally, conservation tillage has been promoted in the region to prevent soil erosion into the Chesapeake Bay (NRCS 2011) and other sensitive bodies of water. Maize is the largest acreage crop in Pennsylvania (nearly 2 million acres), grown extensively for grain and silage (NASS 2011), and most fields are established using conservation tillage. In 2010, approximately 60% of maize acres in Pennsylvania were grown without tillage, and another 9% were grown using reduced-tillage methods (ERS 2011). Reducing or eliminating tillage changes important aspects of the soil environment, including increasing surface residues and minimizing soil disturbance, which can combine to alter invertebrate assemblages of pests, decomposers, and predators (Hendrix et al. 1986, Stinner and House 1990). Because of the potential of conservation tillage to conserve predatory arthropods and the biological control services they can provide (Blumberg and Crossley 1983, Brust et al. 1985, 1986, House and Parmelee 1985, House and Alzugaray 1989, Stinner and House 1990, Witmer et al. 2003), these practices may enhance intrinsic pest regulation of crop fields (House and Brust 1989). The likelihood of realizing this potential, however, will depend in part on the other management practices that are employed in a given cropping system. Some of the management decisions that may influence pest and predatory invertebrates within reduced-tillage systems include insect, weed, and nutrient management strategies. For instance, pre-emptive insect pest management is common in maize. Roughly half of no-till farmers recently surveyed in Pennsylvania routinely use a “preventative” insecticide in spring (J. F. Tooker, unpublished data, n = 59). Insecticides may prevent some early season pest damage, but they may also disrupt populations of generalist predators (e.g., Brust et al. 1985), with uncertain net results. In addition, the vast majority (78%) of the maize acreage in Pennsylvania in

49

2010 was planted with transgenic, insect-resistant maize varieties (i.e., Bt hybrids, ERS 2011), providing pre-emptive protection from certain lepidopteran and/or coleopteran pests. Maize seed is also commonly treated with neonicotinoid insecticides (J. F. Tooker, personal communication), offering another layer of pre-emptive protection from insect pests. The necessity of these pre- emptive pest management strategies is unclear in the context of the reduced-tillage, diverse crop rotations that are common in Pennsylvania. Weed management practices are also well known to influence pest and predatory invertebrates in agroecosystems (Norris and Kogan 2005). Herbicides are a major method of weed management in U.S. field crop production, and in 2010, the average Pennsylvania no-till maize field received four herbicide applications (ERS 2011). This prevalent use of herbicides has led to the evolution of herbicide-resistant weeds (Heap 2012), renewing interest in cultural (e.g., cover crops) and mechanical (e.g., cultivation) strategies to suppress weeds (Mortensen et al. 2012). Cover crops have become regionally popular tools that can contribute to integrated weed management programs; additionally, cover crops can enhance beneficial arthropod populations that can contribute to pest control (Carmona and Landis 1999, Davis and Liebman 2003, Lundgren and Fergen 2010). However, cover crops can also increase the risk of damage from some pest species (Willson and Stinner 1994). Mechanical weed control may disrupt predator populations, though the degree of disturbance depends greatly on the implements used, the timing of disturbance, and the particular taxon. Some implements, such as high-residue cultivators, contribute to weed control while leaving most surface residues intact (Bowman 2002). Tillage that does not fully invert the soil often has varying effects on generalist predators, with some species favored by the disturbance and others disfavored (Kromp 1999, Holland and Luff 2000, Hatten et al. 2007). Certain taxa, such as carabid beetles, have shown inconsistent responses to tillage in general (Belaoussoff et al. 2003). Because cover crops, tillage, and herbicides (e.g., Brust 1990) can each alter arthropod populations and the risk of damage from various pest species, there is a need to evaluate trade-offs along the weed management spectrum. Nutrient management can also alter relationships among crops, pests, and predators (Altieri and Nicholls 2003, Zehnder et al. 2007). Because of the prevalence of dairy and other livestock production in the Northeastern and mid-Atlantic regions of the U.S., soil fertility management in these regions often involves manure, particularly in maize production (ERS 2011). While manure applications can exacerbate damage from certain pest invertebrates (e.g. slugs, Iglesias et al. 2001), they can alleviate damage from others (e.g., European corn borer, Phelan et al. 1995), and can under some conditions enhance populations of generalist predators

50

(Chaing 1970, Purvis and Curry 1984, Settle et al. 1996). In plowed systems, manure can be incorporated into the soil following broadcast application, helping to decrease nutrient loss and pollution from run-off and volatilization. In contrast, manure applied to no-till fields cannot be easily incorporated into soil using traditional equipment, thus nutrient loss and associated pollution are common in no-till regions where manure is prevalent (Meisinger and Jokela 2000, Kleinman et al. 2009). To address the deficiencies of surface-applied manure in no-till systems, agricultural professionals have developed equipment that can inject manure with minimal soil disturbance, leaving surface residues intact (Rotz et al. 2011). It is not known, however, how injecting manure below the soil surface will influence invertebrate assemblages. The goal of this study was to investigate how insect, weed, and nutrient management practices in reduced-tillage maize systems influence pest and predator invertebrate activity. Slugs were a particular focus given their prominence in no-till systems and the paucity of management options available for their control (see Chapter 1). In addition to slugs, I also assessed the influence of management options on black cutworm (Agrotis ipsilon), which is of heightened concern in no-till systems (Johnson et al. 1984, Willson and Eisley 1992), and European corn borer (Ostrinia nubilalis), traditionally considered a major pest of maize in the Northeast (Dillehay et al. 2004). While corn rootworm (Diabrotica spp.) is a major pest complex in maize, I did not include it here because all of the maize in this experiment was grown in rotation with other crops, and Pennsylvania does not host populations of rotation-resistant corn rootworms (Gray et al. 2009). Among predatory arthropods, I focused on ground-dwelling taxa because they are the most likely to respond to soil-based weed and nutrient management practices, and are most relevant for the biological control of soil dwelling pests such as slugs and cutworms. Ground beetles (Carabidae) were the main focus of my efforts because of their documented importance in consuming slugs and insect pests (Kromp 1999, Holland and Luff 2000, Glen and Symondson 2003). My investigation occurred within the context of Penn State’s Sustainable Dairy Cropping Systems project (Penn State 2012), an interdisciplinary experiment started in 2010 to evaluate ecological cropping strategies for Pennsylvania dairy farms. The experiment is evaluating two diverse crop rotations for their ability to produce the forage, feed, and tractor fuel needs of a typical dairy farm while minimizing off-farm inputs and environmental harm. In addition, the experiment includes a high-external-input maize-soybean system for comparison to the diverse rotations. All three rotations include maize. The specific objectives of this study were to:

51

1. Compare a simple two-crop rotation using pre-emptive insect pest management practices to a complex rotation using cover crops and low-external-input insect pest management practices for their influence on invertebrate pest and predator dynamics. 2. Examine for pest and predatory invertebrates the consequences of: a. Reduced-herbicide strategies including high-residue cultivation versus more conventional herbicide-intensive management. b. Nutrient management strategies including: i. Injecting versus broadcasting manure. ii. Using red clover (Trifolium pratense L.) versus hairy vetch (Vicia villosa Roth) as a green manure crop preceding maize.

I hypothesized that for reduced-tillage maize, complex rotations with low-external-input insect pest management would conserve predatory invertebrates and would provide competitive or superior pest control compared to a pre-emptive pest management strategy based on broadcast insecticides and transgenic, insect-resistant maize hybrids. For the weed and nutrient management comparisons, it was challenging to predict the influence of treatments on pest and predatory invertebrates given the potential for trade-offs in many cases. In the weed management comparison, I hypothesized that predatory taxa would have idiosyncractic responses to the treatments, with some suppressed but others favored by the reduced herbicide strategy. Given the known sensitivity of slugs to soil disturbance (Glen and Symondson 2003), I also hypothesized that cultivation in the reduced herbicide treatment would suppress slugs relative to the standard herbicide treatment. In the nutrient management comparison, I hypothesized that injecting manure would disfavor slugs, by minimizing a potential food and shelter resource. During initial observations I noticed that broadcasting manure created a uniform crust over the soil surface, in contrast to a more heterogeneous surface following manure injection, where surface residues remained intact and provided cover. As a result, I predicted that injecting manure would create more favorable habitat and thereby better conserve generalist predators.

52

Materials and methods

Study site and crop management

My research was conducted at Penn State’s Russell E. Larson Agricultural Research Station at Rock Springs (Pennsylvania Furnace, PA), as part of the Sustainable Dairy Cropping Systems project. The study included two, six-year crop rotations that incorporate perennials, cover crops, and high crop diversity (Figure 1). One rotation included maize grown for grain (the ‘Grain’ rotation), and the other included maize grown for silage (the ‘Forage’ rotation). Both of these rotations relied on scouting to manage early season pests and were planted with non-Bt maize varieties. In addition, the study included a simple maize-soybean rotation that served as a low-diversity, high-external-input ‘Control’ rotation. The maize in this rotation was planted with a Bt variety, received a pre-emptive application of pyrethroid insecticide in the spring, and was not preceded by a cover crop. Because of the prevalence of neonicotinoid seed treatments in maize and the difficulty of obtaining untreated seed, maize seed in all rotations was treated with these insecticides. Management practices in the three rotations are summarized in Table 1. Within each rotation, split-plots allowed comparison of manure management or weed management strategies. In the Forage and Control rotations, split-plots compared broadcasting dairy manure on the soil surface to injecting the manure below ground (Figure 1). In the Grain rotation, split-plots compared a standard, herbicide-intensive approach to weed management with a reduced-herbicide approach relying on banded herbicides and high-residue cultivation (Figure 1). In 2011, there was an additional comparison within the Forage rotation; split-split-plots differed in the cover crop preceding corn, red clover or hairy vetch. Split-plot treatments and timing of management activities are summarized in Table 2. Each plot in the experiment measured 27.4 × 36.6 m, and was split into four split-split- plots measuring 27.4 × 9.1 m. Plots were surrounded by grass alleyways and separated from each other by at least 12 m. Each rotation was replicated four times, arranged in a randomized block design. The soil at the site was predominantly a Murrill channery silt loam, with smaller areas of Buchanan channery loam and Hagerstown silt loam.

53

Stand establishment and early season herbivory

To assess maize establishment and damage from early season pests, I randomly selected a three meter section of row from the east and west halves of each split-split-plot when maize reached the two and five leaf stages. All samples were taken at least 1.5 m from plot edges. In each sample, I recorded the number of maize plants and inspected each plant for slug damage, rating them on the following scale: 0: no damage, 1: < 25% leaf area removed, 2: 25-50% leaf area removed, 3: 50-75% leaf area removed, and 4: > 75% leaf area removed (similar to Byers and Calvin 1994). In 2010, a test of this rating system revealed that damage scores of seven independent observers were highly correlated with one another (mean r = 0.86, SD = 0.05). Because slugs feed using a scraping motion, slug damage is distinct from other pest damage, and is also often accompanied by a slime trail that aids in identification. In addition to slug damage, I recorded the number of plants severed by cutworms, along with the number of plants damaged by other caterpillars (e.g., Pseudaleta unipuncta) or billbugs (Spenophorus spp.).

Slug activity-density

To measure slug activity-density, I used square foot pieces of roofing material (made from Owens Corning Rolled Roofing Material, color: Shasta White) as slug shelter traps. Shelter traps provide a relative measure of slug activity (Byers, Barratt, and Calvin 1989), and are the most feasible sampling method for a large experiment such as this that includes many field operations. Starting in mid-April, I randomly placed one shelter in the east and west halves of each split-split-plot, at least 1.5 m from plot edges and each other. I removed shelters to allow for field operations (e.g. manure application) but otherwise left them in the plots continuously. I pushed aside vegetation or residue so that the shelters laid flat on the soil surface, and waited at least two days after putting shelters down before checking them. In 2010, I removed shelters from late June to mid-August during the hottest part of the summer when slug activity on the surface is generally low (Port and Port 1986, Eskelson et al. 2011). In 2011, shelters remained in plots from mid-April to November. Because slugs often leave artificial shelters as they heat up during the day (Hommay et al. 2003), I checked shelters in the morning, with sampling typically completed by noon, and with progressively earlier sampling as temperatures increased through the season. The last sample occurred in each treatment shortly before harvest, in early September for maize

54

grown for silage and in late October for maize grown for grain. I recorded slugs by species (Chichester and Getz 1973, McDonnell et al. 2009), and deposited voucher specimens of the three species with the Frost Entomological Museum at the Pennsylvania State University.

European corn borer activity

To assess damage caused by the European corn borer, I sampled maize plants in late August, shortly before silage harvest. I chose this time to capture cumulative damage from first and second generation larvae (Penn State Agronomy Guide 2011). Other demands on the experiment precluded destructive sampling, so I measured evidence of tunneling that was visible on the outside of the stalks and ears (similar to Witmer et al. 2003). In the fourth and eighth row of each split-split-plot, I sampled three sets of two plants spaced roughly 9 m apart for a total of twelve plants sampled per split-split-plot. Each plant was inspected from top to bottom for entrance holes, and I recorded the number of holes per plant.

Predatory arthropod activity-density

I used pitfall traps to measure the influence of treatments on the activity-density of ground-dwelling, predatory arthropods. I placed two traps in the seventh row of each split-split- plot, spaced equally along the row (~ 9 m apart), for a total of eight traps/plot. Each trap comprised a 16-oz plastic deli container (11.5-cm diameter, 8-cm tall Reynolds Del Pak ®) that was sunk into the ground so that the edge was level with the soil surface. The lip was removed from an identical container, which was placed inside the first so that it could be easily removed to empty the trap without disturbing the surrounding soil. A white plastic plate (18-cm diameter) supported by nails (8.5-cm long) served as a trap cover and the killing agent was 50% propylene glycol. Traps were opened for 48 h, with the first sample occurring roughly two weeks after maize planting and additional samples about every two and a half weeks until early September (maize silage harvest). The two samples from each split-split-plot were combined into one sample, strained through a fine mesh sieve (1-mm openings) and then transferred into 80% ethanol for later identification. Predatory arthropods were identified to the lowest level possible by me, mostly to family except for Carabidae which were identified to species or genus. I

55

identified carabid species using a regional publication (Bousquet 2010) and a reference collection at the Pennsylvania State University from previous studies of carabids in central Pennsylvania (Leslie et al. 2007, 2009). Other arthropods were identified with trusted references (Kaston et al. 1972, Triplehorn and Johnson 2005, and Ubick et al. 2009). Voucher specimens of carabids and other arthropods identified in this study have been deposited with the Frost Entomological Museum at the Pennsylvania State University.

Predation on sentinel caterpillars

In 2011, I used sentinel waxworm caterpillars (Galleria mellonella) to assess predation (after Lundgren et al. 2006). Once in June (6/14/11) and again in July (7/21/11), I deployed eight caterpillars along the eighth row of each split-split-plot. I pinned caterpillars (mean mass: 0.20g, SD: 0.04g) through their final abdominal segment to a small piece of modeling clay that was buried in the field so that the caterpillar rested on the soil surface. In initial laboratory and field experiments, caterpillars survived in this condition for over 24 hours. Every other caterpillar was enclosed in a cylindrical, hardware cloth cage (9.5-cm tall, 11.5-cm diameter, mesh size: 1.3 cm) topped with a plastic lid to exclude vertebrates. On each sample date, I made two assessments of predator activity: one started at 8:30, and the other at 20:30. Sentinel caterpillars were checked 3 and 12 h after being placed in the fields. After 12 h, any caterpillars that had been attacked or were missing or compromised were replaced with new caterpillars. During night-time sampling, observers used red headlamps or lights covered in red cellophane to minimize disturbance to predators. I recorded caterpillars as whole and alive, partially eaten, or missing at each time point. Caterpillars that died but showed no signs of predation were recorded and excluded from the analysis. In addition, observers recorded the presence of predators in proximity to or eating the caterpillars.

Statistical analyses

To understand the influence of treatments on pests and predatory invertebrates, I compared response variables using linear mixed models (PROC MIXED; SAS 9.2, SAS Institute Inc. 2009) for all analyses unless otherwise stated. To improve their relevance or satisfy

56

assumptions of normality and homoscedasticity, I transformed some of the data prior to analysis. For stand establishment, I converted the number of maize seedlings per sample at V5 into plants per acre, and then divided by the seeding rate for each treatment to calculate establishment as a percentage of the target number of seeds planted. Pitfall trap catches were totaled over the season to compare cumulative predator activity-density in the different treatments. For Carabidae, I separately analyzed species that made up more than one percent of the total carabid catch in each year. Slug counts, European corn borer damage, and pitfall trap catches were square-root transformed. I used three models to examine the influence of experimental treatments on response variables (see details in Appendix B). For variables with a single measurement (maize establishment, European corn borer damage, seasonal pitfall trap catches), I used a model with block as a random factor, rotation as a fixed factor, and weed or manure split-plot treatment as a fixed factor nested within rotation. Because only the Forage rotation included split-split-plot treatments, I performed a separate analysis for this rotation, with block as a random factor, manure split-plot treatment as a fixed factor, and cover crop split-split-plot treatment as a fixed factor. Where the analysis indicated a significant treatment effect (P ≤ 0.05), I used Tukey’s post- hoc tests to separate means. For response variables that were measured at multiple times (slug damage at V2 and V5, slug activity-density over the season), I used repeated measures analyses. The model was similar to that described above, but also included date and its interactions with block, rotation, and split- plot as within-subject factors. I evaluated candidate covariance structures for each response variable and selected an appropriate covariance structure by minimizing AICC scores (Wang and Goonewardene 2004, Littell et al. 2006). This resulted in using the variance components covariance structure for both slug damage and slug activity-density. Where the analysis indicated a significant date by treatment effect, I used the ESTIMATE procedure to separate means within dates (Littell et al. 2006). The analysis of predation on sentinel caterpillars included several additional variables that necessitated a unique repeated measures analysis. The model included block as a random factor, rotation as a fixed factor, and weed or manure split-plot treatments as a fixed factor nested within rotation. In addition, cage treatment (caged or open) was included as a fixed factor, along with its interactions with the other factors. The repeated factor was time of day (day vs. night). To examine whether slug activity-density measured via shelter traps was related to slug damage to maize seedlings, I used correlation. Because slug activity-density was highly variable

57

from sample date to sample date, I averaged values of slug activity-density collected in spring (April to June) for each plot, and then related activity-density to mean slug damage scores using Pearson’s correlation (Minitab v. 16, Minitab, Inc., State College, PA). Because of differences in the timing of field operations each year, the spring average represented three sample dates in 2010 and nine sample dates in 2011. I used a similar approach to examine whether predator activity-density measured via pitfall traps was correlated with predation on sentinel caterpillars under vertebrate exclusion cages in 2011. Because of low predation in June, this analysis was restricted to July (sentinel prey sample: 7/20/11; pitfall sample: 7/21/11-7/23/11). I used field observations of predation on tethered caterpillars to choose taxa that would be expected to correlate with predation. The taxonomic groups tested in the analysis included ants (Formicidae), the dominant ground beetle Pterostichus melanarius, all large Carabidae (≥ 9mm), and all predatory arthropods combined. I used Pearson’s correlation (Minitab v. 16) to test for a significant relationship between catches of these taxa and caterpillar predation as observed three and twelve hours after the start of the experiment. Both variables were averaged at the plot level. In the following text, I report mean ± SE unless otherwise noted.

Results

Stand establishment and early season herbivory

Weed and manure split-plot treatments did not significantly influence maize establishment in either year (2010: F3,9 = 1.48, P = 0.28; 2011: F3,9 = 0.34, P = 0.80). However, maize establishment was marginally more successful in the Control and Grain rotations than in the Forage rotation in 2010 (F2,6 = 4.78, P = 0.057) , and in 2011, was more successful in the

Control rotation than in the low-external-input Forage and Grain rotations (F2,6 = 17.86, P < 0.01;

Table 3). Within the Forage rotation, maize establishment did not differ between manure (F1,3 =

0.02, P = 0.90) or cover crop (F1,3 = 1.03, P = 0.39) treatments, although there was a marginal manure by cover crop interaction (F1,3 = 6.57, P = 0.08). This trend appeared to be driven by greater establishment in the red clover (RC) versus the hairy vetch (HV) split-split plot within the injected manure (IM) treatment (IM-RC: 91 ± 3%, IM-HV: 85 ± 2%).

58

Slug damage was the most prevalent type of herbivore damage on maize seedlings at V2 and V5 in both years. In 2010, there were no significant effects of manure, weed, or insect management practices on the percentage of seedlings damaged by slugs (Rotation effect: F2,6 =

0.05, P = 0.95, Split-plot effect: F3,9 = 1.68, P = 0.24; Table 4) or the damage rating of seedlings

(Rotation effect: F2,6 =0.26 , P = 0.78, Split-plot effect: F3,9 = 1.40, P = 0.30; Table 4). In 2011, after the plots had been established for one year, slug damage was greater in the Forage rotation than in the Control and Grain rotations, measured both as percentage of seedlings damaged (F2,6 =

14.0, P = 0.05; Table 4) and as damage rating (F2,6 =13.0 , P = 0.01; Table 4). In contrast, no significant differences in slug damage were observed among manure or weed management split- plot treatments (percentage of seedlings damaged: F3,9 = 2.09, P = 0.17; damage rating: F3,9 = 2.34, P = 0.14). Within the Forage rotation, slug damage was greater following hairy vetch than red clover (percentage damaged: F1,3 = 11.85, P = 0.04; average rating: F1,3 = 11.62, P = 0.04; Table 5). Cutworm damage was too low for formal statistical analysis in either year, but was below 1.5% in all rotations (Table 6). Defoliation caused by other insect pests at V2 and V5 was generally minor (data not shown).

Slug activity-density

Three slug species were found at this site: the gray garden slug (Deroceras reticulatum), the marsh slug (D. laeve) and the banded slug (Arion fasciatus). Because they can all damage crops, slug species were pooled into one category for analysis. In 2010, slug activity-density did not differ significantly among rotations, weed, or manure management treatments either before or after maize silage harvest (Before harvest, Rotation: F2,6 = 1.47, P = 0.30, Date*Rotation: F12,108 =

0.74, P = 0.71, Split-plot: F3,9 = 0.83, P = 0.51, Date*Split-plot: F18,108 = 0.64, P = 0.86; After harvest, Rotation: F1,3 = 0.62, P = 0.49, Date*Rotation: F6,72 = 1.48, P = 0.20, Split-plot: F2,6 =

0.39, P = 0.69, Date*Split-plot: F12,72 = 0.52, P = 0.90; Figure 2). In 2011, before maize silage harvest, the influence of rotation on slug activity-density changed over time, as reflected in a significant Date*Rotation interaction (F32,288 = 1.78, P < 0.01). In spring, the low-external-input Forage rotation experienced elevated slug activity-density on several spring dates relative to the two other rotations (Figure 2). However, by early September, slug activity-density in the Control rotation surpassed the Grain rotation, and slug activity-density in the Forage rotation was intermediate between the two (Figure 2). In the

59

separate analysis within the Forage rotation, there were no significant differences in slug activity- density by manure split-plot treatment (F1,3 = 0.17, P = 0.71), cover crop split-split-plot treatment

(F1,3 = 1.25, P = 0.34), or their interactions with each other (F1,195 = 0.74, P = 0.39). After maize silage harvest in 2011, there were significant effects of both split-plot treatment (F2,6 = 14.11, P < 0.01) and rotation (F1,3 = 9.73, P = 0.05) on slug activity-density. Overall, the Control rotation had higher slug activity-density than the low-external-input Grain rotation (Figure 2). Consistent with my hypothesis, the reduced herbicide treatment experienced lower slug activity-density than the standard herbicide treatment in the Grain rotation (Figure 3). This depressed slug activity-density in the reduced herbicide treatment also appeared to be driving the significant difference between the Grain and Control rotations, since the standard herbicide split-plot in the Grain rotation did not differ significantly from either manure split-plot in the Control rotation (Figure 3). In both 2010 and 2011, slug activity-density in spring as measured by shelter traps was significantly correlated with slug damage to maize seedlings at V2 and V5 (Figure 4).

European corn borer damage

In 2010, European corn borer damage was, as expected, lowest in the Control rotation where Bt hybrids were planted (F2,6 = 59.48, P < 0.01). Damage was intermediate in the Forage rotation, and highest in the Grain rotation (Table 7). In 2011, damage was again lowest in the Bt hybrids of the Control rotation, which received significantly less damage than maize plants in the

Forage and Grain rotations (F2,6 = 23.78, P < 0.01; Table 7). Damage did not differ significantly among weed or manure treatments in 2010 (F3,9: 0.64, P = 0.61), and differed only marginally among split-plot treatments in 2011 (F3,9 = 3.16, P = 0.08)—a difference that appeared in the Forage rotation to be driven by a trend for higher damage in the broadcast manure (BM) treatment than the injected manure (IM) treatment (mean number of entrance holes in BM: 0.45 ± 0.08; IM: 0.25 ± 0.04). However, in the separate analysis within the Forage rotation, corn borer damage did not differ significantly by manure split-plot treatment (F1,3 = 3.52, P = 0.16), cover crop split-split-plot treatment (F1,3 = 0.45, P = 0.55), or their interaction (F1,3 = 0.27, P = 0.64). Taken together, these results suggest that weed and nutrient management tactics did not strongly influence European corn borer damage.

60

Predatory arthropod activity-density

In 2010, pitfall trapping yielded 8180 predatory arthropods of which 1323 were carabid adults. In 2011, pitfall trapping yielded 7167 predatory arthropods of which 1416 were carabid adults. Other prominent predatory taxa captured in pitfall traps included ants (Formicidae), rove beetles (Staphylinidae), wolf spiders (Lycosidae), and sheet web spiders (Linyphiidae). Modest numbers of harvestmen (Opiliones) and crickets (Gryllidae) were also captured. Each of these predatory taxa appeared to have distinct seasonal patterns of abundance (Figures 5, 6). Among Carabidae, Pterostichus melanarius was by far the dominant species, accounting for 76% of carabids captured in 2010 and 47% of carabids captured in 2011, and was active throughout the sampling period (Figure 5). In addition to P. melanarius, I found at least 30 carabid species in pitfall samples (Table 8). In 2010, the first year plots were established, there were few differences among insect, weed, or manure management treatments in the number of predatory arthropods captured in pitfall traps over the season (Table 9). The sole exception was in linyphiid spiders, which tended to be higher where manure was broadcasted, compared to where it was injected (Table 9). In 2011, there were again few differences among weed or manure treatments, but there were more widespread, yet idiosyncratic, differences among rotations in the number of predatory arthropods captured over the season (Table 10). In partial agreement with my hypothesis, all predators combined had highest activity-density in the low-external-input Forage rotation but lowest activity-density in the low-external-input Grain rotation, with activity-density in the Control rotation intermediate between the two (Table 10).

Predation on sentinel caterpillars

In June 2011, predation on sentinel caterpillars differed by caging treatment (Table 11). Caterpillars under vertebrate exclusion cages were less likely to be attacked than uncaged caterpillars (Figure 7). No other factors caused significant differences in caterpillar predation in June. In addition, few arthropod predation events were observed (Table 12), although temperatures during this experiment dropped below 10°C, likely contributing to low levels of arthropod activity. Birds (esp. American robins, Turdus migratorius) were observed consuming uncaged caterpillars, mainly in the morning.

61

In July 2011, there was a significant interaction of Time*Rotation*Cage (Table 13), driven by greater predation under vertebrate exclusion cages in the Forage and Grain rotations than in the Control rotation at night, consistent with my hypothesis of enhanced arthropod predation services in the low-external-input rotations (Figure 8). Ants were the primary arthropods observed attacking caterpillars during the day, while at night ants and ground beetles were observed attacking caterpillars along with a variety of other taxa (Table 14). Pitfall catches of large (> 9mm) carabids were positively correlated to nighttime predation on sentinel caterpillars, at both three and twelve hours after caterpillars were deployed (Table 15). Pitfall catches of ants or total predators were not significantly correlated to caterpillar predation either day or night (Table 15).

Discussion

The main objective of this study was to compare in reduced-tillage maize systems the influence on pest and predatory invertebrates of complex rotations with fewer external inputs to a simpler rotation with a pre-emptive pest management strategy. An additional objective was to explore the influence of weed and manure management options on these same invertebrate dynamics. With few exceptions, weed and manure management treatments appeared to have little effect on the activity of pest and predatory invertebrates. This result may have been related to limitations inherent to the study, such as the generally low power of pitfall trap data to resolve treatment differences (Lopez et al. 2005) or the movement of organisms between split-plots (Thomas et al. 2006). Alternatively, it may be that the treatments had not been in place long enough to strongly influence invertebrate dynamics. My study occurred during the first two years of the larger experiment; future years will shed light on the long-term influence of these practices. Slug activity-density did show a response to weed management treatment. Consistent with my hypothesis, the reduced herbicide treatment in the Grain rotation experienced lower slug activity-density than the standard herbicide treatment later in the season (Figure 3). Although slug suppression occurred after the period when maize is vulnerable to slugs, this effect could have consequences for slug populations over time. This is particularly the case in cropping systems where high-residue cultivation is used in successive years, as in the present study. The dominant pest slug, Deroceras reticulatum, has an annual life cycle with mating and egg-laying occurring

62

mainly in the late summer and fall (see Chapter 1, Appendix A), giving rise to the following year’s population. My results therefore lead to the hypothesis that high-residue cultivation may reduce the prevalence of slugs in the following crop. Previous studies on the effect of non- inversion tillage on slug populations in Europe have reached mixed conclusions (Glen and Symondson 2003), and it appears that more research is needed to better predict the ability of shallow tillage to influence slug populations, in a wider range of geographies. In contrast to the weed and manure management treatments, I found widespread differences for maize grown in complex rotations with low-external-input insect management tactics versus a pre-emptive strategy in a simpler rotation, particularly in the second year of the study when cover crop treatments had been established. Contrary to my hypothesis, crop establishment was more successful in the Control rotation than in the two other rotations in 2011, perhaps because the broadcast insecticide application and Bt traits limited early season insect damage, or because cover crop residue at planting interfered with seed placement in the low- external-input rotations. Despite this statistical difference, all three rotations in both years were established successfully, with realized plant populations within the recommended range for grain and silage production on typical Pennsylvania soils (grain: 28,000 – 30,000 plants; silage: 30,000 – 34,000 plants; Penn State Agronomy Guide 2011). Early season invertebrate pests are one of the factors that can decrease stand establishment in agronomic crops. Black cutworms in particular can decrease plant populations by cutting plants at or below the surface; however, their activity was low in all three rotations. Fewer than 1.5% of maize seedlings were killed by cutworms, below the threshold at which rescue treatment is typically recommended (3% of plants at V2; Penn State Agronomy Guide 2011), and too low to determine the influence of treatments on black cutworm activity. This finding of limited black cutworm damage is consistent with evidence from Ohio that even in no- till environments, economic damage from black cutworms occurs in fewer than 10% of maize fields each year (Willson and Eisley 1992). The use of neonicotinoid seed treatments in our experiment may have also protected against cutworms, although common seed treatments have shown uneven performance against cutworm feeding (Wilde et al. 2007, Kullik et al. 2011). Interestingly, slugs were the most prevalent early season pests in all three rotations. The minimum level of slug defoliation that can result in economic injury in maize varies widely (2 - 59%), depending on weather conditions and inherent variability (Byers and Calvin 1994). Given this wide range of values, it is difficult to interpret the level of damage I observed. While slug feeding was widespread, damage to individual plants averaged less than 25% defoliation in all

63

treatments. In 2010, plants that I tracked over the season did not show a consistent relationship between early slug damage and eventual grain production (Appendix C). This combined with the generally high tolerance of young maize plants to defoliation (Vorst 1986) suggests that the slug damage I observed was mostly sub-economic. Nonetheless, I found that the number of slugs under shelter traps was correlated with damage to maize in both years. Finding this relationship is particularly notable given that refuge traps underestimate small slugs (Archard et al. 2004, Cordoba et al. 2011), which were the predominant size of slug soon after planting (Appendix A). Shelter traps have also shown promise for predicting slug damage in a variety of other cropping systems and locations (Archard et al. 2004, Nash et al. 2007, Cordoba et al. 2011). Although slug monitoring was not the primary objective of this study, the inexpensive and convenient nature of the traps used here suggests that further research to hone their use in North American field crops could be fruitful. Given this generally low level of slug pressure, the responses of slugs to experimental treatments were all the more remarkable. While slug damage and activity-density were similar among treatments in the first year of the study, in 2011 both measures were influenced by rotation. In spring, I found more slug damage and often higher slug activity-density in the Forage rotation than the other two rotations. Because slug damage is strongly associated with crop residue (Hammond and Stinner 1987), this result was likely related to the abundant crop residue and close to 100% ground cover at planting from the red clover or hairy vetch cover crops, versus the generally scant residue in the Grain and Control rotations (E. Snyder, W. Curran, and H. Karsten, unpublished data; also personal observations). The finding of increased slug damage following hairy vetch or red clover agrees with previous studies finding an association between cover crops, especially legumes, and slug damage (Mangan et al. 1995, Glen 2000, Vernava et al. 2004). Interestingly, hairy vetch was associated with greater slug damage than red clover, despite it producing 36% less biomass (E. Snyder, W. Curran, and H. Karsten, unpublished data). Patterns in slug activity-density across the rotations changed in an interesting way over the season in 2011. By late summer, shortly before silage harvest, slug activity-density was no longer highest in the Forage rotation. By early September slug activity-density was highest in the Control rotation, lowest in the low-external-input Grain rotation, and intermediate in the low- external-input Forage rotation. One explanation for this change is that predatory arthropods influenced slug populations. While the early-season predation sample was thwarted by unseasonably cold temperatures (< 10oC), sentinel prey assays in July revealed predation differences across treatments. Predation on sentinel caterpillars under exclusion cages was similar

64

in all rotations during the day, but in the evening invertebrate predators attacked ~80% of caterpillars in the Forage and Grain rotations while predators attacked only ~35% of caterpillars in the Control rotation (Fig. 8). Importantly, pitfall catches of large Carabidae, taken shortly after the sentinel prey sample, were significantly correlated with caterpillar predation (Table 15). Moreover, many of the large carabid species at this site (Pterostichus melanarius, Chlaenius tricolor, Harpalus pensylvanicus) are documented slug predators (Symondson et al. 1996, Eskelson et al. 2011, Chapter 2). Pterostichus melanarius, in particular, tended to be trapped in higher numbers in the Forage and Grain rotations than in the Control rotation, although this difference was not statistically significant. These results suggest that carabid beetle predators were likely responsible for many of the attacks on sentinel caterpillars, and because they are known slug predators, these species may also help limit slug populations in reduced-tillage maize fields. Taken together, these results are consistent with the hypothesis that complex rotations with low-external-input pest management tactics can sustain levels of pest control similar to or superior to those found in the simpler rotations that depend upon a pre-emptive strategy. Multiple factors may have contributed to the pattern of increased nighttime predation in the Forage and Grain rotations in July. Because cover crops preceded maize in the Forage and Grain rotations but not the Control rotation, improved habitat from winter cover and spring residue may have led to enhanced predation services, as has been observed in maize for corn rootworm predation (Lundgren and Fergen 2010). Insecticides may have also contributed to depressed predation in the Control rotation, which was alone in receiving a pre-emptive pyrethroid treatment in spring. In 2011 from mid-June to early August, I found dead ground beetles under the shelter traps, predominantly in the Control rotation (Control: 10 P. melanarius and 14 other carabids; Forage: 3 P. melanarius and 1 other carabid; Grain: 0 P. melanarius and 1 other carabid), although the samples sizes are too low for formal analyses. Nevertheless, these results agree with previous studies associating insecticide applications with reduced predation services (Brust et al. 1985), and burgeoning slug populations (Grant et al. 1982). It is worth noting that pitfall trap data were not entirely consistent with predation on sentinel caterpillars. While in 2011 predation was lowest in the Control rotation, seasonal predator activity-density was lowest in the Grain rotation, intermediate in the Control rotation, and highest in the Forage rotation. The disagreement between pitfall trap catches and sentinel prey data may result in part from differences between numerical abundance and functional significance. For instance, the trend for the Control rotation to have higher catches of predators than the Grain rotation was driven largely by rove beetles (Staphylinidae), most of which were

65

very small (< 5 mm long), potentially limiting their contribution to predation of larger prey items. In contrast, my research suggests that large carabid species may be responsible for much of the predation I documented. These findings point to the importance of including functional assays wherever possible to complement pitfall trapping. Nonetheless, it is interesting that the largest numbers of predatory arthropods were captured in the Forage rotation, particularly since trapping efficiency would tend to be lowest in this treatment given its structural complexity near the soil surface (Greenslade 1964, Melbourne 1999). In particular, seasonal catches of ants were almost twice as high in the Forage rotation versus the other two rotations, and ants were also commonly observed attacking sentinel caterpillars. Despite a few studies documenting the importance of ants to biological control in temperate agroecosystems (Kirk 1981, Brust et al. 1986, Clark et al. 1994), it appears that more research in this direction would be profitable (Way and Khoo 1992, Symondson et al. 2002). Similar to other invertebrate pests, European corn borer injury was relatively low in both years of this study. As expected given the strong efficacy of Bt against this species (Burkness et al. 2002), damage was virtually non-existent on the Bt variety in the Control rotation. Even in the Grain and Forage rotations where non-Bt varieties were planted, maize plants averaged less than one entrance hole per plant. In central Pennsylvania, one larva per plant is associated with between 2 and 6% yield loss, depending on the stage when maize is infested (Bode and Calvin 1990). The low corn borer pressure I observed is consistent with evidence that populations of European corn borer have been in decline since the widespread adoption of Bt crops (Hutchinson et al. 2010, Bohnenblust and Tooker, unpublished data). Given this evidence, it appears that non- Bt varieties can be competitive with Bt varieties under certain conditions in Pennsylvania. This is particularly important given that the Bt seed in the Control rotation cost more than $35/acre more than the conventional seed in the Grain rotation (Glenna Malcolm, personal communication), and seed costs for transgenic seed have generally been rising rapidly since the early 2000’s (Benbrook 2009). While not the focus of this study, one important measure of the competitiveness of the low-external-input approach is grain yields. Yields in the low-external-input Grain rotation were competitive with or even superior to those in the Control rotation (2010, Grain: 204 ± 2.4 bu/acre, Control: 217 ± 4.2 bu/acre; 2011, Grain: 164 ± 4.0 bu/acre, Control: 129 ± 13 bu/acre; G. Malcolm and H. Karsten, unpublished data). These results suggest that pest pressure in the low- external-input Grain rotation did not unduly reduce yield in that rotation relative to the Control rotation.

66

Conclusion

Overall, pest damage to reduced-tillage maize in the two years of this experiment was minor, with slugs, cutworms, and European corn borers failing to materialize in substantial numbers. While there were some trends for the complex, low-external-input rotations to experience greater pest pressure, in general these rotations fared well compared to the Control rotation, which relied on pre-emptive pest management practices. This could be due at least in part to the enhanced predation services in the low-external-input rotations, and the trend for enhanced predator activity-density in one of the two low-external-input rotations in 2011. These results join a growing body of evidence that low-external-input cropping systems can conserve beneficial arthropods, adequately manage invertebrate pests, and be profitable (House and Brust 1989, Settle et al. 1996, Gallandt et al. 1998, Letourneau and Goldstein 2001, Witmer et al. 2003, Liebman et al. 2008, O’Rourke et al. 2008). Future years in this experiment will shed light on the risk to maize from insect and slug pests over a wider range of growing conditions and after treatments have been in place for several years.

67

References

Altieri, M. and C. Nicholls. 2003. Soil fertility management and insect pests: harmonizing soil and plant health in agroecosystems. Soil & Tillage Research 72: 203-211. Archard, G. A., D. A. Bohan, L. Hughes, and C. W. Wiltshire. 2004. Spatial sampling to detect slug abundance in an arable field. Annals of Applied Biology. 145: 165-173. Belaoussoff, S., P. G. Kevan, S. Murphy, and C. Swanton. 2003. Assessing tillage disturbance on assemblages of ground beetles (Coleoptera: Carabidae) by using a range of ecological indices. Biodiversity and Conservation 12: 851-882. Benbrook, C. 2009. The magnitude and impacts of the biotech and organic seed price premium. The Organic Center. Accessed 6/11/12 at www.organic-center.org. Blumberg, A. Y. and D. A. Crossley. 1983. Comparison of soil surface arthropod populations in conventional tillage, no-tillage and old field systems. Agro-Ecosystems 8: 247-253. Bode, W. M. and D. D. Calvin. 1990. Yield-loss relationships and economic injury levels for European corn borer (Lepidoptera: Pyralidae) populations infesting Pennsylvania field corn. Journal of Economic Entomology 83(4): 1595-1603. Bousquet, Y. 2010. Illustrated Identification Guide to Adults and Larvae of Northeastern North American Ground Beetles (Coleoptera: Carabidae). Pensoft Publishers, Sofia-Moscow, Bulgaria. Bowman, G. 2002. Steel in the field: A farmer’s guide to weed management tools. Sustainable Agriculture Network, Beltsville, MD. Brust, G. E. 1990. Direct and indirect effects of four herbicides on the activity of carabid beetles (Coleoptera: Carabidae). Pesticide Science 30 (3): 309-320. Brust, G.E., B.R. Stinner, and D.A. McCartney. 1985. Tillage and soil insecticide effects on predator-black cutworm (Lepidoptera: Noctuidae) interactions in corn agroecosystems. Journal of Economic Entomology 78(6): 1389-1392 Brust, G. E., B. R. Stinner, and D. A. McCartney. 1986. Predator activity and predation in corn agroecosystems. Environmental Entomology 15: 1017-1021. Burkness, E. C., W. D. Hutchinson, R. A. Weinzierl, J. L. Wedberg, S. J. Wold, and J. T. Shaw. 2002. Efficacy and risk efficiency of sweet corn hybrids expressing a Bacillus

68

thuringiensis toxin for Lepidopteran pest management in the Midwestern U.S. Crop Protection 21(2): 157-169. Byers, R. A., B. I. P. Barratt, and D. Calvin. 1989. Comparison between defined-area traps and refuge traps for sampling slugs in conservation tillage crop environments. In: I. F. Henderson, ed. Slugs and Snails in World Agriculture, BCPC Monograph 41, 187-192. Byers, R. A. and D. D. Calvin. 1994. Economic injury levels to field corn from slug (Stylommatophora: Agrolimacidae) feeding. Journal of Economic Entomology 87(5): 1345-1350. Carmona, D. M. and D. A. Landis. 1999. Influence of refuge habitats and cover crops on seasonal activity-density of ground beetles (Coleoptera: Carabidae) in field crops. Environmental Entomology 28(6): 1145-1153. Chaing, H. C. 1970. Effects of manure applications and mite predation on corn rootworm populations in Minnesota. Journal of Economic Entomology 64(3): 934-936. Chichester, L. F. and L. L. Getz. 1973. The terrestrial slugs of northeastern North America. Sterkiana 51: 11-42. Clark, M. S., J. M. Luna, N. D. Stone, and R. R. Youngman. 1994. Generalist predator consumption of armyworm (Lepidoptera: Noctuidae) and effect of predator removal on damage in no-till corn. Environmental Entomology 23(3): 617-622. Cordoba, M., J. Iglesias, P. Ribadulla, and J. Castillejo. 2011. Performance of permanent refuge traps for the assessment of slug populations in pastureland. Annals of Applied Biology 159: 130-140. Davis, A. S. and M. Liebman. 2003. Cropping system effects on giant foxtail (Setaria faberi) demography I: green manure and tillage timing. Weed Science 51(6): 919-929. Dillehay, B. L., G. W. Roth, D. D. Calvin, R. J. Kratochvil, G. A. Kuldau, and J. A. Hyde. 2004. Performance of Bt corn hybrids, their near isolines, and leading corn hybrids in Pennsylvania and Maryland. Agronomy Journal 96: 818-824. Economic Research Service. 2011. Crop Production Practices. Agricultural Resource Management Survey. United States Department of Agriculture. Accessed 6/1/12: http://www.ers.usda.gov/Data/ARMS. Eskelson, M. J., E. G. Chapman, D. D. Archbold, J. J. Obrycki, and J. D. Harwood. 2011. Molecular identification of predation by carabid beetles on exotic and native slugs in a strawberry agroecosystem. Biological Control 56(3): 245-253.

69

Gallandt, E. R., E. B. Mallory, A. R. Alford, F. A. Drummond, E. Groden, M. Liebman, M. C. Marra, J. C. McBurnie, and G. A. Porter. 1998. Comparison of alternative pest and soil management strategies for Maine potato production systems. American Journal of Alternative Agriculture 13: 146-161. Glen, D. M. 2000. The effects of cultural measures on cereal pests and their role in integrated pest management. Integrated Pest Management Reviews 5: 25-40. Glen, D. M. and W. O. C. Symondson. 2003. Influence of soil tillage on slugs and their natural enemies. pp. 207-227in: A. El Titi (ed.) Soil Tillage in Agroecosystems. CRC Press, Boca Raton, FL. Grant, J. F., K. V. Yeargan, B. C. Pass, and J. C. Parr. 1982. Invertebrate organisms associated with alfalfa seedling loss in complete-tillage and no-tillage plantings. Journal of Economic Entomology 75: 822-826. Gray, M. E., T. W. Sappington, N. J. Miller, J. Moeser, and M. O. Bohn. 2009. Adaptation and invasiveness of Western Corn Rootworm: Intensifying research on a worsening pest. Annual Review of Entomology 54: 303-321. Greenslade, P. J. M. 1964. Pitfall trapping as a method for studying populations of Carabidae (Coleoptera). Journal of Animal Ecology 33(2): 301-310. Hammond, R. B. and B. R. Stinner. 1987. Seedcorn maggots (Diptera: Anthomyiidae) and slugs in conservation tillage systems in Ohio. Journal of Economic Entomology 80(3): 680- 684. Hatten, T. D., N. A. Bosque-Pérez, J. R. Labonte, S. O. Guy, and S. D. Eigenbrode. 2007. Effects of tillage on the activity density and biological diversity of carabid beetles in spring and winter crops. Environmental Entomology 36(2): 356-368. Heap, I. 2012. International Survey of Herbicide Resistant Weeds. Accessed 6/1/12: www.weedscience.org Hendrix, P.F., R.W. Parmelee, D.A. Crossley Jr., D.C. Coleman, E.P. Odum, and P.M. Groffman. 1986. Detritus food webs in conventional and no-tillage agroecosystems. BioScience 36(6): 374-380. Holland, J. M. and M. L. Luff. 2000. The effects of agricultural practices on Carabidae in temperate agroecosystems. Integrated Pest Management Reviews 5: 109-129. Hommay, G., J. C. Kienlen, F. Jacky, and C. Gertz. 2003. Daily variation in the number of slugs under refuge traps. Annals of Applied Biology 142: 333-339.

70

House, G. J. and M. D. R. Alzugaray. 1989. Influence of cover cropping and no-tillage practices on community composition of soil arthropods in a North Carolina agroecosystem. Environmental Entomology 18(2): 302-307. House, G. J. and G. E. Brust. 1989. Ecology of low-input, no-tillage agroecosystems. Agriculture, Ecosystems and Environment 27: 331-345. House, G. J. and R. W. Parmelee. 1985. Comparison of soil arthropods and earthworms from conventional and no-tillage agroecosystems. Soil & Tillage Research 5: 351-360. Hutchinson, W. D., E. C. Burkness, P. D. Mitchell, R. D. Moon, T. W. Leslie, S. J. Fleischer, M. Abrahamson, K. L. Hamilton, K. L. Steffey, M. E. Gray, R. L. Hellmich, L. V. Kaster, T. E. Hunt, R. J. Wright, K. Pecinovsky, T. L. Rabaey, B. R. Flood, and E. S. Raun. 2010. Areawide suppression of European corn borer with Bt maize reaps savings to non-Bt maize growers. Science 330(6001): 222-225. Iglesias, J., J. Castillejo, and R. Castro. 2001. Mini-plot field experiments on slug control using biological and chemical control agents. Annals of Applied Biology 139: 285-292. Johnson, T. B., F. T. Turpin, M. M. Schreiber, and D. R. Griffith. 1984. Effects of crop rotation, tillage, and weed management systems on black cutworm (Lepidoptera: Noctuidae) infestations in corn. Journal of Economic Entomology 77: 919-921. Kaston, B. J. 1972. How To Know The Spiders, 2nd ed. W. C. Brown Company, Dubuque, Iowa. Kirk, V. M. 1981. Corn rootworm: Population reduction associated with the ant, Lasius neoniger. Environmental Entomology 10(6): 966-967. Kleinman, P. J. A., A. N. Sharpley, L. S. Saporito, A. R. Buda, and R. B. Bryant. 2009. Application of manure to no-till soils: phosphorous losses by sub-surface and surface pathways. Nutrient Cycles in Agroecosystems 84: 215-227. Knisley, C. B. and T. D. Schultz. 1997. The Biology of Tiger Beetles and a Guide to the Species of the South Atlantic States. Virginia Museum of Natural History, Special Publication Number 5, Martinsville, VA. Kromp, B. 1999. Carabid beetles in sustainable agriculture: a review on pest control efficacy, cultivation impacts and enhancement. Agriculture, Ecosystems, and Environment 74(1- 3): 187-228. Kullik, S. A., M. K. Sears, and A. W. Schaafsma. 2011. Sublethal effects of Cry 1F Bt corn and clothianidin on black cutworm (Lepidoptera: Noctuidae) larval development. Journal of Economic Entomology 104(2): 484-493.

71

Leslie, T. W., G. A. Hoheisel, D. J. Biddinger, J. R. Rohr, and S. J. Fleischer. 2007. Transgenes sustain epigeal insect biodiversity in diversified vegetable farm systems. Environmental Entomology 36(1): 234-244. Leslie, T. W., D. J. Biddinger, C. A. Mullin, and S. J. Fleischer. 2009. Carabidae population dynamics and temporal partitioning: response to coupled neonicotinoid-transgenic technologies in maize. Environmental Entomology 38(3): 935-943. Letourneau, D. K. and B. Goldstein. 2001. Pest damage and arthropod community structure in organic vs. conventional tomato production in California. Journal of Applied Ecology 38: 557-570. Liebman, M., L. R. Gibson, D. N. Sundberg, A. H. Heggenstaller, P. R. Westerman, C. A. Chase, R. G. Hartzler, F. D. Menalled, A. S. Davis, and P. M. Dixon. 2008. Agronomic and economic performance charateristics of conventional and low-external-input cropping systems in the central Corn Belt. Agronomy Journal 100(3): 600-610. Littell, R. C., G. A. Milliken, W. W. Stroup, R. D. Wolfinger, and O. Schabenberger. 2006. SAS® for Mixed Models, Second Edition. SAS Institute Inc., Cary, NC. Lopez, M. D., J. R. Prasifka, D. J. Bruck, and L. C. Lewis. 2005. Utility of ground beetle species in field tests of potential nontarget effects of Bt crops. Environmental Entomology 34(5): 1317-1324. Lundgren, J. G., J. T. Shaw, E. R. Zaborski, and C. E. Eastman. 2006. The influence of organic transition systems on beneficial ground-dwelling arthropods and predation of insects and weed seeds. Renewable Agriculture and Food Systems 21(4): 227-237. Lundgren, J. G., and J. K. Fergen. 2010. The effects of a winter cover crop on Diabrotica virgifera (Coleoptera: Chrysomelidae) populations and beneficial arthropod communities in no-till maize. Environmental Entomology 39(6): 1816-1828. Mangan, F., R. DeGregorio, M. Schonbeck, S. Herbert, K. Guillard, R. Hazzard, E. Sideman, and G. Litchfield. 1995. Cover cropping systems for brassicas in the Northeastern United States. 2. Weed, insect, and slug incidence. Journal of Sustainable Agriculture 5(3): 15- 36. McDonnell, R. J., T. D. Paine, and M. J. Gormally. 2009. Slugs: A Guide to the Invasive and Native Fauna of California. University of California, Division of Agriculture and Natural Resources, Publication 8336.

72

Meisinger, J. J. and W. E. Jokela. 2000. Ammonia volatilization from dairy and poultry manure. In: Managing Nutrient and Pathogens from Animal Agriculture (NRAES-130). Natural Resource, Agriculture, and Engineering Service, Ithaca, NY. Melbourne, B. A. 1999. Bias in the effect of habitat structure on pitfall traps: An experimental evaluation. Australian Journal of Ecology 24: 228-239. Mortensen, D. A., J. F. Egan, B. D. Maxwell, M. R. Ryan, and R. G. Smith. 2012. Navigating a critical juncture for sustainable weed management. BioScience 62(1): 75-84. Nash, M. A., L. J. Thomson, and A. A. Hoffmann. 2007. Slug control in Australian canola: monitoring, molluscicidal baits and economic thresholds. Pest Management Science 63: 851-859. National Agricultural Statistics Service. 2011. Quick Stats 2.0 database. United States Department of Agriculture. Accessed 6/1/12: http://www.nass.usda.gov/Quick_Stats. Natural Resources Conservation Service. 2011. Assessment of the effects of conservation practices on cultivated cropland in the Chesapeake Bay Region. United States Department of Agriculture. Norris, R. F. and M. Kogan. 2005. Ecology of interactions between weeds and arthropods. Annual Review of Entomology 50: 479-503. O’Rourke, M. E., M. Liebman, and M. E. Rice. 2008. Ground beetle (Coleoptera: Carabidae) assemblages in conventional and diversified crop rotation systems. Environmental Entomology 37(1): 121-130. The Pennsylvania State University. 2011. Penn State Agronomy Guide. Accessed 6/1/12: http://extension.psu.edu/agronomy-guide. The Pennsylvania State University. 2012. Sustainable Dairy Cropping Systems. Accessed 7/2/12: http://plantscience.psu.edu/research/areas/crop-ecology-and-management/cropping- systems Phelan, P. L., J. F. Mason, and B. R Stinner. 1995. Soil-fertility management and host preference by European corn borer, Ostrinia nubilalis (Hübner), on Zea mays L.: A comparison of organic and conventional chemical farming. Agriculture, Ecosystems, & Environment 56(1): 1-8. Port, C. M. and G. R. Port. 1986. The biology and behavior of slugs in relation to crop damage and control. Agricultural Zoology Reviews 1: 255-297.

73

Purvis, G. and J. P. Curry. 1984. The influence of weeds and farmyard manure on the activity of Carabidae and other ground-dwelling arthropods in a sugar beet crop. Journal of Applied Ecology 21: 271-283. Rotz, C. A., P. J. Kleinman, C. J. Dell, T. L. Veith, and D. B. Beegle. 2011. Environmental and economic comparisons of manure application methods in farming systems. Journal of Environmental Quality 40(2): 438-448. SAS Institute Inc. 2009. SAS/STAT® 9.2 User’s Guide, Second Edition. SAS Institute Inc., Cary, NC. Settle, W. H., H. Ariawan, E. T. Astuti, W. Cahyana, A. L. Hakim, D. Hindayana, and A. S. Lestari. 1996. Managing tropical rice pests through conservation of generalist natural enemies and alternative prey. Ecology 77(7): 1975-1988. Stinner, B.R. and G. J. House. 1990. Arthropods and other invertebrates in conservation-tillage agriculture. Annual Review of Entomology 35: 299-318. Symondson, W. O. C., D. M. Glen, C. W. Wiltshire, C. J. Langdon, and J. E. Liddell. 1996. Effects of cultivation techniques and methods of straw disposal on predation by Pterostichus melanarius (Coleoptera: Carabidae) upon slugs (Gastropoda: Pulmonata) in an arable field. Journal of Applied Ecology 33(4): 741-753. Symondson, W. O. C., K. D. Sunderland, and M. H. Greenstone. 2002. Can generalist predators be effective biocontrol agents? Annual Review of Entomology 47: 561-594. Thomas, C. F. G., N. J. Brown, and D. A. Kendall. 2006. Carabid movement and vegetation density: implications for interpreting pitfall trap data from split-field trials. Agriculture, Ecosystems and Environment 113: 51-61. Triplehorn, C. A. and N. F. Johnson. 2005. Borror and Delong’s Introduction to the Study of Insects, 7th edition. Thomson Brooks/Cole, Belmont, CA. Ubick, D., P. Paquin, P. E. Cushing, and V. Roth (eds.) 2005. Spiders of North America: An Identification Manual. American Arachnological Society. Vernava, M. N., P. M. Phillips-Aalten, L. A. Hughes, H. Rowcliffe, C. W. Wiltshire, and D. M. Glen. 2004. Influences of preceding cover crops on slug damage and biological control using Phasmarhabditis hermaphrodita. Annals of Applied Biology 145: 279-284. Vorst, J. J. 1986. Assessing hail damage to corn. Institute of Agriculture and Natural Resources, Cooperative Extension Service. University of Nebraska NebGuide 803. Wang, Z. and L. A. Goonewardene. 2004. The use of MIXED models in the analysis of animal experiments with repeated measures data. Canadian Journal of Animal Science 84: 1-11.

74

Way, M. J. and K. C. Khoo. 1992. Role of ants in pest management. Annual Review of Entomology 37: 479-503. Wilde, G., K. Roozeboom, A. Ahmad, M. Claassen, B. Gordon, W. Heer, L. Maddux, V. Martin, P. Evans, K. Kofoid, J. Long, A. Schlegel, and M. Witt. 2007. Seed treatment effects on early-season pests of corn and on corn growth and yield in the absence of insect pests. Journal of Agricultural and Urban Entomology 24(4): 177-193. Willson, H. R. and J. B. Eisley. 1992. Effects of tillage and prior crop on the incidence of five key pests on Ohio corn. Journal of Economic Entomology 85(3): 853-859. Willson, H. R. and B. R. Stinner. 1994. Recovery of field corn following insecticide treatment to arrest defoliation by Pseudaletia unipuncta (Haworth) (Lepidoptera: Noctuidae). Journal of Agricultural Entomology 11(4): 383-392. Witmer, J. E., J. A. Hough-Goldstein, and J. D. Pesek. 2003. Ground-dwelling and foliar arthropods in four cropping systems. Environmental Entomology 32(2): 366-376. Zehnder, G., G. M. Gurr, S. Kühne, M. R. Wade, S. D. Wratten, and E. Wyss. 2007. Arthropod pest management in organic crops. Annual Review of Entomology 52: 57-80.

75

Tables

Table 3-1. Management practices in maize plots in the three crop rotations.

Year Management practice Control Forage Grain 2010 Maize variety Pioneer 35F44 Pioneer 38B84 Pioneer 35F38 Seed treatment THX/CLO THX THX Transgenic traits HXX, LL, RR2 RR2 - Soil insecticide 23 d after planting (PYR) - - Previous crop * * * Previous cover crop ** ** ** Planting date 5/26 5/26 5/25 2011 Maize variety Pioneer 35F48AM TA290-08 Pioneer 35F38 Seed treatment THX/CLO CLO THX Transgenic traits HXX, LL, RR2 LL - Soil insecticide At planting (PYR) - - Previous crop Soybean Wheat Soybean Previous cover crop - RC or HV Rye Planting date 6/1 6/3 5/26

Key: PYR = Pyrethroid THX = Thiamethoxam CLO = Clothianidin HXX = Herculex extra (lepidopteran and coleopteran Bt traits) LL = Liberty link (Glufosinate tolerant) RR2 = Roundup Ready 2 (Glyphosate tolerant) RC = Red clover HV = Hairy vetch

*2010 was the first year of the experiment and previous crops varied across the site.

**In 2010, all treatments in blocks 1, 2, and most of 3 were preceded by a rye cover crop, while part of block 3 and all of block 4 had no cover crop.

76

Table 3-2. Management practices and timing of management activities in maize split-plots across the three crop rotations. IM = injected manure; BM = broadcasted manure; RH = reduced herbicide; SH = standard herbicide.

Control Forage Grain IM BM IM BM RH SH 2010 Manure application Inject Broadcast Inject Broadcast Inject Inject Manure rate (T/A) 16.1 16.1 14.8 17.0 16.8 16.8 Cover crop herb. 5/19 5/19 5/19 5/19 5/19 5/19 Pre-emergence herb. 5/28 5/28 - - 5/25 (B) 5/25 Post-emergence herb. 6/18 6/18 6/18 6/18 - 6/18 High-residue cultiv. - - - - 6/18, 6/25 - 2011 Manure application Inject Broadcast Inject Broadcast Inject Inject Manure rate (T/A) 18.5 20.3 20.7 20.4 18.6 18.6 Cover crop herb. - - 5/12 5/12 5/6 5/6 Pre-emergence herb. 6/4 6/4 - - 6/29 (B) 6/29 Post-emergence herb. 6/29 6/29 6/29 6/29 - 6/24 High-residue cultiv. - - - - 6/23, 6/27 -

(B) = Herbicides were banded over the row, rather than being spraying over the entire field.

77

Table 3-3. Maize establishment (mean ± SE, n = 4) at V5 as a percentage of the target number of seeds planted, by year and rotation. Values marked with different letters are statistically different at P ≤ 0.05 based on Tukey’s post-hoc comparisons.

Seeds planted Stand as % of Year Rotation (#/acre) seeding rate 2010 Control 32,000 94 ± 0.3 Forage 35,000 90 ± 1.8 Grain 32,000 94 ± 0.7 2011 Control 32,000 100 ± 1.5a Forage 35,000 88 ± 1.2b Grain 32,000 90 ± 1.6b

Table 3-4. Slug damage (mean ± SE, n = 4) to maize seedlings at V2 and V5 by rotation. Values marked with different letters are statistically different at P ≤ 0.05 based on ESTIMATE comparisons within stages.

% of plants damaged Damage rating* Year Rotation V2 V5 V2 V5 2010 Control 30 ± 18 48 ± 21 0.40 ± 0.27 0.71 ± 0.39 Forage 19 ± 6 55 ± 16 0.20 ± 0.06 0.63 ± 0.20 Grain 23 ± 7 48 ± 14 0.28 ± 0.10 0.53 ± 0.17 2011 Control 18 ± 9b 32 ± 10b 0.18 ± 0.09b 0.33 ± 0.11b Forage 53 ± 7a 72 ± 6a 0.56 ± 0.07a 0.73 ± 0.06a Grain 30 ± 4b 32 ± 6b 0.31 ± 0.04b 0.32 ± 0.07b *Damage rating scale: 0: no defoliation; 1: < 25% defoliation; 2: 25-50% defoliation; 3: 50-75% defoliation; 4: > 75% defoliation.

Table 3-5. Slug damage to maize seedlings (mean ± SE, n = 4) at V2 and V5 by cover crop treatment in the Forage rotation. Values marked with different letters are statistically different at P ≤ 0.05 based on ESTIMATE comparisons within stages.

Previous % of plants damaged Damage rating* Year cover crop V2 V5 V2 V5 2011 Hairy vetch 63 ± 6a 82 ± 6a 0.67 ± 0.07a 0.83 ± 0.06a Red clover 44 ± 8b 62 ± 7b 0.45 ± 0.09b 0.62 ± 0.07b *Damage rating scale: 0: no defoliation; 1: < 25% defoliation; 2: 25-50% defoliation; 3: 50-75% defoliation; 4: > 75% defoliation.

78

Table 3-6. Percentage of maize seedlings severed by cutworms (mean ± SE, n = 4) by year and rotation.

Year Rotation V2 V5 2010 Control 0.0 ± 0.0 0.0 ± 0.0 Forage 0.2 ± 0.2 0.0 ± 0.0 Grain 0.6 ± 0.4 0.0 ± 0.0 2011 Control 0.2 ± 0.2 0.3 ± 0.2 Forage 0.5 ± 0.3 0.3 ± 0.3 Grain 0.9 ± 0.5 1.1 ± 0.2

Table 3-7. European corn borer damage (mean ± SE, n = 4) by year and rotation. Values were square root transformed for analysis, but untransformed values are shown here. Values marked with different letters are statistically different at P ≤ 0.05 based on Tukey’s post-hoc comparisons.

Year Rotation Holes/plant 2010 Control 0.0 ± 0.0c Forage 0.3 ± 0.1b Grain 0.8 ± 0.2a 2011 Control 0.0 ± 0.0b Forage 0.4 ± 0.1a Grain 0.3 ± 0.1a

79

Table 3-8. Adult carabid species collected via pitfall trapping, 2010 and 2011.

2010 2011 Combined Species Total % Total % Total % Pterostichus melanarius (Illiger) 1011 76.4 669 47.2 1680 61.3 Bembidion quadrimaculatum oppositum Say 88 6.7 134 9.5 222 8.1 Amara spp. 19 1.4 126 8.9 145 5.3 Harpalus pensylvanicus (DeGeer) 26 2 101 7.1 127 4.6 Cicindella punctulata punctulata Olivier 2 0.2 103 7.3 105 3.8 Pterostichus mutus (Say) 46 3.5 43 3 89 3.2 Chlaenius tricolor Dejean 14 1.1 54 3.8 68 2.5 Harpalus affinis (Schrank) 6 0.5 45 3.2 51 1.9 Other Harpalus spp. 16 1.2 18 1.3 34 1.2 Poecilus lucublandus (Say) 11 0.8 19 1.3 30 1.1 Bembidion rapidum (LeConte) - - 18 1.3 18 0.7 Notiophilis sp. 1 9 0.7 7 0.5 16 0.6 Anisodactylus sanctaecrucis (Fabricius) 4 0.3 10 0.7 14 0.5 Dyschirius sp. 1 4 0.3 6 0.4 10 0.4 Scarites subterraneus Fabricius 8 0.6 2 0.1 10 0.4 Bembidion sp. 1 - - 9 0.6 9 0.3 Paracliniva bipustulata (Fabricius) 2 0.2 5 0.4 7 0.3 Diplochelia obtusa (LeConte) 5 0.4 1 0.1 6 0.2 Anisodactylus rusticus (Say) - - 5 0.4 5 0.2 Cicindella sexguttata Fabricius 2 0.2 3 0.2 5 0.2 Poecilus chalcites (Say) - - 5 0.4 5 0.2 Stenolophus comma (Fabricius) 4 0.3 1 0.1 5 0.2 Scarites quadriceps Chaudoir 3 0.2 1 0.1 4 0.1 Badister notatus Haldeman 1 0.1 2 0.1 3 0.1 Harpalus rubripes (Duftschmid) - - 3 0.2 3 0.1 Agonum cupripenne (Say) 1 0.1 1 0.1 2 0.1 Chlaenius emarginatus Say - - 1 0.1 1 < 0.1 Chlaenius sp. 1 1 0.1 - - 1 < 0.1 Colliuris pensylvanica (Linné) 1 0.1 - - 1 < 0.1 Dicaelus elongatus Bonelli - - 1 0.1 1 < 0.1 Sphaeroderus sp. 1 1 0.1 - - 1 < 0.1 Other Carabidae 38 2.9 23 1.6 61 2.2 Total Carabidae 1323 100 1416 100 2739 100.0

80

Table 3-9. Seasonal totals of predatory arthropods captured in pitfall traps in 2010 (mean ± SE, n = 4). Data were square root transformed for analysis but raw values are shown here. Values marked with different letters are significantly different based on Tukey’s post-hoc tests at P ≤ 0.05. Lowercase letters signify split-plot comparisons. IM = inject manure; BM = broadcast manure; RH = reduced herbicide; SH = standard herbicide.

Control Forage Grain Rotation Split Taxon (Life stage)a BM IM BM IM RH SH P P Formicidae (A) 45.1 ± 9.6 56.1 ± 14.3 65.1 ± 29.5 69.6 ± 16.8 38.9 ± 11.7 50.6 ± 14 0.38 0.65 Lycosidae (J,A) 40.9 ± 13.9 24.5 ± 5.8 53.3 ± 6.5 30.3 ± 9.1 16.4 ± 2.2 33.6 ± 10.1 0.23 0.08 Linyphiidae (J,A) 24.4 ± 6.8ab 18.9 ± 3.9b 29.8 ± 3.0a 21.4 ± 4.9ab 15.3 ± 3.2b 18.0 ± 3.9b 0.02 0.04 Staphylinidae (A) 24.5 ± 5.6 18.9 ± 1.8 25.1 ± 6.0 18 ± 2.8 16.9 ± 2.5 18.9 ± 1.7 0.69 0.27 Carabidae (A) 22.5 ± 5.7 26.3 ± 11.0 25.9 ± 9.0 25.8 ± 4.7 37.1 ± 9.2 27.9 ± 7.7 0.64 0.48 Pterostichus melanarius 14.8 ± 6.0 20.1 ± 10.4 21.1 ± 9.6 21.1 ± 5.8 30 ± 10.4 19.3 ± 8.9 0.79 0.10 Bembidion quad. opp. 2.8 ± 0.9 1.4 ± 0.4 1.9 ± 0.1 1.6 ± 0.6 1.4 ± 0.5 2 ± 0.7 0.89 0.51 Pterostichus mutus 0.4 ± 0.4 0.1 ± 0.1 0.1 ± 0.1 0.0 ± 0.0 1.4 ± 1.4 3.8 ± 3.8 0.47 0.44 Harpalus pensylvanicus 0.6 ± 0.2 0.8 ± 0.3 0.4 ± 0.4 0.4 ± 0.1 0.6 ± 0.5 0.5 ± 0.4 0.67 0.96 Amara spp. 0.5 ± 0.5 0.9 ± 0.9 0.0 ± 0.0 0.3 ± 0.1 0.3 ± 0.1 0.5 ± 0.4 0.69 0.44 Other Harpalus spp. 0.4 ± 0.4 0.1 ± 0.1 0.6 ± 0.5 0.4 ± 0.2 0.4 ± 0.2 0.1 ± 0.1 0.74 0.68 Chlaenius tricolor 0.5 ± 0.4 0.6 ± 0.5 0.1 ± 0.1 0.1 ± 0.1 0.3 ± 0.1 0.1 ± 0.1 0.50 0.74 Carabidae (L) 0.1 ± 0.1 0.9 ± 0.7 0.5 ± 0.0 0.1 ± 0.1 0.1 ± 0.1 0.5 ± 0.3 1.00 0.20 Coccinellidae (L,A) 4.5 ± 0.8 4.0 ± 0.5 6.8 ± 2.7 4.0 ± 1.0 4.3 ± 1.4 5.4 ± 1.6 0.97 0.45 Opiliones (J,A) 3.1 ± 0.5 4.1 ± 2.0 3.5 ± 1.2 5.3 ± 1.5 6.0 ± 2.6 5.6 ± 1.8 0.51 0.68 Gryllidae (J,A) 1.4 ± 0.5 1.5 ± 1.0 2.0 ± 1.1 1.6 ± 0.8 3.0 ± 0.9 1.3 ± 0.8 0.59 0.09 Other Araneae (J,A) 1.0 ± 0.5 1.6 ± 0.4 1.8 ± 0.4 1.8 ± 0.1 1.4 ± 0.5 1.4 ± 0.6 0.62 0.48 Lampyridae (L,A) 1.0 ± 0.5 0.3 ± 0.1 0.8 ± 0.4 0.6 ± 0.2 0.3 ± 0.1 0.4 ± 0.2 0.46 0.42 Chrysopidae (L) 0.3 ± 0.3 0.4 ± 0.2 0.1 ± 0.1 0.1 ± 0.1 0.1 ± 0.1 0.3 ± 0.1 0.66 0.78 Total predators 168.8 ± 10.5 157.4 ± 5.8 214.5 ± 32.8 178.5 ± 18.4 139.6 ± 22.5 163.8 ± 29.0 0.15 0.48 aLife stage notation: A = adult; J = juvenile; L = larva

81

Table 3-10. Seasonal totals of predatory arthropods captured in pitfall traps in 2011 (mean ± SE, n = 4). Data were square root transformed for analysis but raw values are shown here. Values marked with different letters are significantly different based on Tukey’s post-hoc tests at P ≤ 0.05. Uppercase letters signify rotation comparisons. IM = inject manure; BM = broadcast manure; RH = reduced herbicide; SH = standard herbicide.

Control Forage Grain Taxon (Life stage) Rotation Split BM IM BM IM RH SH P P Formicidae (A) 37.1 ± 8.3B 32.0 ± 7.2B 58.9 ± 13.7A 66.0 ± 12.4A 31.8 ± 9.7B 28.3 ± 6.9B 0.00 0.69 Staphylinidae (A) 39.6 ± 7.3A 36.5 ± 8.0A 36.9 ± 7.2A 42.9 ± 3.7A 23.8 ± 3.1B 24.3 ± 4.4B 0.02 0.73 Lycosidae (J,A) 22.1 ± 7.1 22.4 ± 3.9 10.1 ± 1.2 11.3 ± 1.6 12.8 ± 1.6 18.3 ± 4.1 0.08 0.48 Linyphiidae (J,A) 9.1 ± 1.9B 10.0 ± 2.8B 17.5 ± 2.4A 21.0 ± 1.8A 13.9 ± 1.1A 16.0 ± 2.7A 0.00 0.35 Carabidae (A) 29.3 ± 2.6 30.8 ± 6.3 33 ± 5.6 30.6 ± 3.4 27.5 ± 4.6 25.9 ± 9.6 0.67 0.92 Pterostichus melanarius 8.0 ± 3.0 10.4 ± 4.7 18.3 ± 4.9 15.5 ± 0.7 14.6 ± 7.1 16.9 ± 10.7 0.50 0.96 Bembidion quad. opp. 2.6 ± 0.7 1.9 ± 1.2 3.1 ± 1.0 2.3 ± 0.4 4.5 ± 0.9 2.4 ± 0.9 0.24 0.12 Amara spp. 4.5 ± 2.8 5.3 ± 3.9 2.3 ± 1.3 2.9 ± 1.7 0.4 ± 0.1 0.5 ± 0.4 0.24 0.67 Cicindella punctulata 5.3 ± 3.3A 4.6 ± 1.9A 1.0 ± 1.0AB 1.8 ± 0.7AB 0.0 ± 0.0B 0.3 ± 0.1B 0.04 0.48 Harpalus pensylvanicus 3.3 ± 1.0 1.5 ± 0.4 1.4 ± 0.1 2.6 ± 0.9 2.1 ± 0.6 1.8 ± 0.3 0.94 0.33 Chlaenius tricolor 1.6 ± 0.5AB 0.3 ± 0.3AB 2.9 ± 1.4A 1.6 ± 0.3A 0.1 ± 0.1B 0.3 ± 0.1B 0.02 0.10 Harpalus affinis 1.8 ± 0.8 2.8 ± 1.8 0.4 ± 0.2 0.4 ± 0.2 0.3 ± 0.3 0.1 ± 0.1 0.08 0.54 Pterostichus mutus 0.0 ± 0.0 0.3 ± 0.1 0.6 ± 0.6 0.5 ± 0.5 2.3 ± 2.3 1.8 ± 1.3 0.55 0.64 Poecilus lucublandus 0.3 ± 0.1 0.3 ± 0.1 0.0 ± 0.0 0.6 ± 0.4 1.0 ± 0.4 0.3 ± 0.1 0.34 0.03* Bembidion rapidum 0.0 ± 0.0B 0.1 ± 0.1B 0.9 ± 0.4A 0.8 ± 0.3A 0.4 ± 0.2B 0.1 ± 0.1B 0.02 0.81 Other Harpalus spp. 0.5 ± 0.2 0.6 ± 0.5 0.0 ± 0.0 0.5 ± 0.4 0.6 ± 0.4 0.0 ± 0.0 0.59 0.20 Carabidae (L) 1.9 ± 1.0 1.0 ± 0.5 0.9 ± 0.2 1.0 ± 0.2 0.6 ± 0.1 0.3 ± 0.3 0.09 0.34 Opiliones (J,A) 4.9 ± 0.4 8.6 ± 0.8 7.4 ± 1.2 6.3 ± 0.6 4.3 ± 1.2 4.1 ± 0.8 0.08 0.10 Gryllidae (J,A) 1.0 ± 0.7B 1.6 ± 0.2B 4.3 ± 0.9A 4.1 ± 0.7A 2.9 ± 0.6AB 2.1 ± 0.5AB 0.01 0.20 Coccinellidae (L,A) 1.3 ± 0.4 1.9 ± 0.3 2.3 ± 0.8 1.9 ± 0.6 2.1 ± 0.5 1.5 ± 0.7 0.78 0.44 Other Araneae (J,A) 1.4 ± 0.5 1.5 ± 0.5 1.3 ± 0.5 0.8 ± 0.4 0.9 ± 0.6 0.8 ± 0.3 0.61 0.87 Lampyridae (L,A) 0.5 ± 0.2 0.5 ± 0.3 0.4 ± 0.2 0.3 ± 0.3 0.1 ± 0.1 0.0 ± 0.0 0.20 0.71 Chrysopidae (L,A) 0.0 ± 0.0 0.1 ± 0.1 0.0 ± 0.0 0.1 ± 0.1 0.1 ± 0.1 0.0 ± 0.0 1.00 0.42 Total predators 148 ± 26AB 147 ± 18AB 173 ± 21A 186 ± 21A 121 ± 13B 121 ± 14B 0.02 0.85 *Although this overall p-value was significant, Tukey’s test did not identify any means as significantly different.

82

Table 3-11. Repeated measures analysis of predation on sentinel caterpillars in June 2011.

Num. d.f. Den. d.f. F P Rotation 2 6 2.14 0.198 Split-plot(Rotation) 3 9 0.32 0.811 Cage 1 3 53.45 0.005 Rotation*Cage 2 6 2.59 0.155 Split-plot(Rotation)*Cage 3 9 0.43 0.736 Time 1 36 0.54 0.467 Time*Rotation 2 36 2.65 0.084 Time*Split-plot(Rotation) 3 36 0.02 0.997 Time*Cage 1 36 2.88 0.098 Time*Rotation*Cage 2 36 0.49 0.618 Time*Split-plot(Rotation)*Cage 3 36 0.13 0.945

Table 3-12. Predators observed during sentinel prey experiment, June 2011, three hours after sentinel caterpillars were deployed.

3 h time point Predator taxa AM PM Formicidae (Ants) 3 10 Carabidae (Ground beetles) - 8 Opiliones (Harvestmen) - 13 Lycosidae (Wolf spiders) - 2 Agriolimacidae (Slugs) - 2 Total 3 35

Table 3-13. Repeated measures analysis of predation on sentinel caterpillars in July 2011.

Num. d.f. Den. d.f. F P Rotation 2 6 3.22 0.112 Split-plot(Rotation) 3 9 2.83 0.099 Caged 1 3 12.81 0.037 Rotation*Cage 2 6 2.88 0.133 Split-plot(Rotation)*Cage 3 9 0.29 0.835 Time 1 36 56.91 < 0.001 Time*Rotation 2 36 8.82 0.001 Time*Split-plot(Rotation) 3 36 0.28 0.840 Time*Cage 1 36 8.84 0.005 Time*Rotation*Cage 2 36 7.58 0.002 Time*Split-plot(Rotation)*Cage 3 36 0.22 0.884

83 Table 3-14. Predators observed in sentinel prey experiment, July 2011, three hours after sentinel caterpillars were deployed.

3 h time point Predator taxa AM PM Formicidae (Ants) 22 68 Carabidae (Ground beetles) 2 31 Opiliones (Harvestmen) - 4 Lycosidae (Wolf spiders) - 8 Gryllidae (Crickets) - 7 Chilopoda (Centipedes) - 1 Dermaptera (Earwigs) - 1 Coccinellidae (Lady beetles) 1 - Total 25 120

Table 3-15. Correlations between predation on sentinel caterpillars on 7/20/11 and numbers of arthropods in pitfall samples from 7/21/11 to 7/23/11. Predator activity-densities were square root transformed for analysis.

3 hours post 12 hours post r P r P Day Formicidae -0.02 0.95 0.15 0.64 All predators -0.32 0.32 0.11 0.73 Night Formicidae 0.15 0.64 0.18 0.58 Pterostichus melanarius 0.48 0.10 0.34 0.28 Large (>9mm) Carabidae* 0.70 0.01 0.57 0.05 All predators -0.41 0.18 -0.41 0.18

*Cicindella punctulata punctulata was excluded because tiger beetles are primarily diurnal (Knisley and Schultz 1997).

84

Figures

Figure 3-1. Schematic of maize comparisons in the Sustainable Dairy Cropping Systems project. Cover crop treatments were in place in 2011 but not 2010.

85

Figure 3-2. Slug activity-density through time by rotation (mean ± SE, n = 4). Values were square root transformed for analysis but here I show the untransformed values. The vertical line in September marks maize silage harvest. In 2010, management practices in spring precluded slug sampling for most of the early season, hence the few data points in spring 2010. Asterisks indicate dates on which there was a significant difference between rotations at P ≤ 0.05 based on ESTIMATE comparisons within dates.

86

Figure 3-3. Slug activity-density (means ± SE, n = 4) after maize silage harvest in 2011 by rotation and split-plot treatment. Values were square root transformed for analysis but here I show the untransformed values. Bars with different letters are significantly different based on Tukey’s post-hoc comparisons at P ≤ 0.05. C-BM: Control rotation, broadcast manure; C-IM: Control rotation, injected manure; G-RH: Grain rotation, reduced herbicide; G-SH: Grain rotation, standard herbicide.

87

Figure 3-4. Correlations between number of slugs per trap during spring and damage to corn seedlings at V2 and V5 (n = 12 plots). Slug activity-density was square root transformed for analysis but here I show the untransformed values.

88

Figure 3-5. Seasonal patterns of activity-density of predatory arthropods captured in pitfall traps in 2010 and 2011, pooled across treatments (mean ± SE, n = 12 plots).

89

Figure 3-6. Seasonal patterns of activity-density of predatory arthropods captured in pitfall traps in 2010 and 2011, pooled across treatments (mean ± SE, n = 12 plots).

90

Figure 3-7. Predation on sentinel caterpillars in June over 12 hours as a function of time of day, caging treatment, and rotation.

Figure 3-8. Predation on sentinel caterpillars in July over 12 hours as a function of time of day, caging treatment, and rotation.

Chapter 4

A laboratory-based assessment of the influence of thiamethoxam seed treatment on interactions among soybeans (Glycine max), slugs (Deroceras reticulatum) and ground beetles (Chlaenius tricolor)

Introduction

Over a third of U.S. cropland is managed without tillage (Horowitz et al. 2010), changing relative to tilled fields important aspects of agroecosystems including pest and predator assemblages (Stinner and House 1990). No-till management reduces soil disturbance and allows crop residue to accumulate, creating conditions that were historically thought to promote invertebrate pests as a group (Gregory and Musick 1976). Subsequent studies have shown that pests respond to no-till in a species-specific manner, with positive, negative, and neutral responses possible (Musick 1985, Stinner et al. 1988, House and Alzugaray 1989, Stinner and House 1990, Hammond 1997). In those cases where no-till reduces pest pressure, it often does so by conserving natural enemies, increasing top-down suppression relative to conventionally tilled systems (House and Stinner 1983, Brust et al. 1985, Brust et al. 1986, Landis et al. 1987, Brust and House 1990, Wardle 1995). It is therefore especially important in no-till environments to evaluate the non-target risks of insecticides, in order to balance the benefits of primary pest control with the potential costs of generating secondary pests via natural enemy disruption (e.g. Brust et al. 1985). Neonicotinoids are a relatively new class of insecticides, introduced in the 1990s and rapidly becoming one of the most widely used classes of insecticides in the world (Jeschke and Nauen 2008, Jeschke et al. 2011). They are highly systemic, allowing for novel application techniques such as soil drenches and seed treatments (Jeschke et al. 2011), which are hypothesized to reduce exposure of non-target organisms relative to more traditional spray applications (e.g., Epperlein et al. 2001, Elbert et al. 2008). In field crops such as maize and soybeans, seed treatments are the predominant method of neonicotinoid application, targeting early season pests such as seed corn maggot, (Delia platura Meigen), wireworms (Agriotes spp.), and white grubs (Scarabaeidae spp.) among others (Elbert et al. 2008). The neonicotinoids

92 contained in these seed treatments are taken up by crop roots and translocated to shoots and leaves, where they persist for weeks to months after planting (Bonmatin et al. 2003, Laurent and Rathahao 2003, Bonmatin et al. 2005). While use data are not publicly available, extension entomologists report that neonicotinoid seed treatments (NSTs) have become the norm in transgenic maize and are increasingly common in soybeans and small grains (Gray and Onstad 2008, Steffey 2008, J. F. Tooker, personal communication). Shortly after crop emergence, when concentrations of neonicotinoids in treated plant tissues are high, crop seedlings are susceptible to slugs, one of the most challenging groups of pests for no-till growers in the Northeast and mid-Atlantic U.S. (Hammond and Byers 2002, Chapter 1). Nonetheless, few studies have examined whether NSTs have activity against slugs. Given that neonicotinoids bind fairly specifically to insect acetylcholine receptors (Tomizawa and Casida 2005), and that slugs and insects are distantly related (arthropods vs. mollusks), it is difficult to predict how slugs might respond to these compounds. Previous research indicates modest activity of imidacloprid against slugs in wheat (Rose and Oades 2001, Simms et al. 2006), but not in canola (Simms et al. 2006). Furthermore, imidacloprid at usual doses is apparently not lethal to slugs or their eggs (Rose and Oades 2001, Iglesias et al. 2002, Simms et al. 2006). The use of second generation neonicotinoids (e.g. thiamethoxam, clothianidin) is now widespread in field crops, but their activity against slugs is poorly characterized. To improve IPM recommendations for growers in slug-prone regions, there is a need to assess whether these new compounds and application technologies protect seedlings against slugs and have potential non- target effects. Biological control of slugs in North America has been little explored (Thomas et al. 2011), but in Europe generalist ground beetles have been shown to aggregate in areas of high slug abundance, suggesting a contribution to slug suppression (Symondson et al. 1996, Bohan et al. 2000, Symondson et al. 2002). Ground beetle species such as Chlaenius tricolor and Pterostichus melanarius are common in many parts of the Northeast and mid-Atlantic U.S. (Bousquet 2010, Leslie et al. 2010), and appear to consume pest slug species (Symondson et al. 1996, Eskelson et al. 2011, Chapter 2). These ground beetles are highly susceptible to neonicotinoids with almost 100% mortality in a four-day exposure assay to corn seedlings treated with imidacloprid, clothianidin, or thiamethoxam (Mullin et al. 2005). Transfer of neonicotinoids from crops to ground beetles via slugs could represent a novel route of insecticide exposure to natural enemies, with the potential to relax top-down pressure on slugs and other pest populations.

93 My objective was to examine whether thiamethoxam alters relationships among soybeans, slugs, and natural enemies. Thiamethoxam is frequently applied to soybean seed, usually in conjunction with fungicides such as mefenoxam and fludioxonil (i.e., CruiserMaxx®, Syngenta Crop Protection, Inc., Greensboro, NC). I focused on soybeans because they are the most widely planted no-till crop in Pennsylvania, with over 60% of all soybean acres managed without tillage (USDA NASS 2010). Slugs can cause economic damage to soybeans when they graze below the apical meristem, killing the plant and reducing plant stands (Barratt et al. 1994, Hammond 2000, Hammond and Byers 2002). The most significant pest slug on soybean is the gray garden slug (Deroceras reticulatum), a pest of many field crop species not only in North America but also in its native range in Europe, as well as in many other parts of the world where this species has been introduced (South 1992). In Pennsylvania, this species appears to have one generation per year, with small juveniles present in spring that grow throughout the season, then mature and lay eggs in fall (Appendix A). Chlaenius tricolor is a native ground beetle species that proved to be an aggressive slug predator in preliminary experiments (Chapter 2). In addition, field-collected C. tricolor in Kentucky were more likely to have slug remains in their gut than twelve other ground beetle species tested (Eskelson et al. 2011). I investigated 1) whether thiamethoxam protects soybean seedlings from slug damage, and 2) whether slugs that consume thiamethoxam-treated soybean seedlings can pass the insecticide on to their beetle predators. On the basis of previous studies and preliminary experiments, I predicted that thiamethoxam seed treatment would have negligible influence on slug feeding behavior, but that toxins would be transferred up the food chain to harm beetle predators. Moreover, I predicted that toxicity to beetles would increase with the dose of thiamethoxam.

Materials and methods

Seeds, slugs, and beetles

All soybean seeds in the experiment were a single variety, A1016495 (FS HiSOY® RR2) from Growmark, Inc. (Bloomington, IL). Seeds were commercially treated in one of four ways, representing the range of what is commercially available in soybeans: 1) untreated control, U; 2) fungicide-alone (ApronMaxx®, Syngenta), F; 3) fungicide plus low rate insecticide

94 (CruiserMaxx®, Syngenta: thiamethoxam at 0.0756 mg ai/seed), F + L; and 4) fungicide plus high rate insecticide (CruiserMaxx®, Syngenta: thiamethoxam at 0.152 mg ai/seed), F + H. I collected gray garden slugs (D. reticulatum) in areas free from insecticide use in the vicinity of State College, PA, primarily a grassy old field and my backyard. Slugs were maintained in the laboratory at room temperature, in covered plastic boxes, which were lined with a layer of moist potting soil, and fed organic cabbage until used in experiments. These slug colonies were checked several times weekly to remove dead or moribund slugs and to change food as needed. Using dry pitfall traps or hand collection, I collected adults of C. tricolor from corn and alfalfa fields at the Russell E. Larson Agricultural Research Farm at Rock Springs (Pennsylvania Furnace, PA). Beetles were housed individually in 16-oz Reynolds Del-Pak® plastic containers (Reynolds Metal Company, Richmond, VA) with a layer of moist potting soil. Similar to previous studies (e.g., Lundgren et al. 2009), beetles were fed kitten food that had been softened in water (Purina® ProPlan® Selects®; Nestlé Purina PetCare, St. Louis, MO), and maintained in a growth chamber (21oC, 14:10 L:D) until used in experiments. Beetles were in closed containers with moist soil and so I assume that humidity was high. Beetles were checked several times a week to change food. In the following text, I report mean ± SD unless otherwise noted.

Soybean-slug interactions

A. Small slugs

To determine whether seed treatments alter slug feeding, I conducted laboratory experiments with soybean seedlings and small gray garden slugs in June 2011. The experiment was a four by two factorial design with four seed treatments crossed with presence or absence of slugs (n = 8 containers per seed-slug treatment). I included the no-slug control to account for possible direct effects of seed treatments on plant growth (Ford et al. 2010). Dry soil (Premier ® Pro-Mix® BX) was sifted through a 2mm screen and moistened with water at a ratio of approximately two parts water to one part soil by mass. Moist potting soil was added to each 16- oz clear plastic container (Reynolds Del-Pak®) to a depth of ~2.5cm, and four soybean seeds

95 were planted, evenly spaced. One day after seeds were planted, four juvenile slugs (0.045 ± 0.003 g) were randomly assigned to each container in the slug treatment. All containers were then placed in a growth chamber (21oC, 14:10 L:D) for seven days, with status of slugs (alive/dead) and seedlings (undamaged, damaged, killed) recorded each day. After one week, I recovered slugs and soybean seedlings and weighed them. Whole soybean seedlings were separated into living and dead components, with a seedling scored as dead if slugs had eaten completely through the stem. Live and dead soybean components were rinsed in water to remove soil particles, dried in a drying oven at 65oC, and weighed to the nearest 0.01g.

B. Medium slugs

The soybean-slug experiment was repeated with larger slugs in fall 2011. Bioassays were as described, except that instead of four small slugs per container, the slug treatment consisted of a single, medium-sized slug (0.22 ± 0.09 g) per container. The experiment was a two-way design with seed treatment crossed with presence of slugs (n = 34 containers per seed treatment with slugs; n = 24 containers per seed treatment without slugs). Due to logistical constraints, this experiment was blocked into three consecutive trials occurring during September and October 2011. Seed treatments were represented equally within each trial (Table 1). At the end of the experiment, I separated soybean seedlings into living and dead components, rinsed them in water, dried them in a drying oven at 65oC, and weighed them to the nearest 0.01g. Slugs were weighed and then retained for use in predator bioassays, described below.

Slug-ground beetle interactions

To determine whether slugs can transmit systemic, seed-applied insecticides from soybean seedlings to ground beetle predators, medium slugs from the soybean-slug experiments were used in an additional experiment. After slugs were weighed at the end of seven days of feeding, I transferred them to new 16-oz plastic containers with ~1 cm of moist potting soil (one slug per container). Due to limited numbers of ground beetles, not all slugs were carried over into this experiment; slugs were chosen to keep slug mass similar across treatments. The sides of the

96 containers were coated with Fluon (BioQuip Products, Inc., Rancho Dominguez, CA) to keep slugs in the arena where beetles could attack them (Symondson 1993). Beetles were starved for one week prior to the experiment, and then randomly assigned to containers in the lab and introduced in the early evening (n = 17 to 19 beetles per seed-slug treatment). Because these experiments used medium-sized slugs from above, they were initiated in three trials during September and October 2011 (Table 2). I observed interactions in containers under low light conditions for the first 3.5 hours after beetles were introduced, recording the status of slugs and beetles every 15 to 30 minutes. Then, I moved containers to a growth chamber (21oC, 14:10 L:D). For seven days after slugs were introduced, I checked containers daily and recorded the status of slugs and beetles. I classified slugs as alive or dead; dead slugs were inspected under a microscope to confirm predation. Beetle flip-time was recorded daily as a quantitative measure of beetle coordination (Smith and Krischik 1999, Lundgren and Wiedenmann 2002, Eisenback et al. 2010). For each trial, I flipped a beetle on its back using forceps and used a stopwatch to record to the closest half second the time necessary for the beetle to right itself, ending a trial after 30 seconds if the beetle failed to flip over. To reduce variability, I flipped each beetle over four times in succession and averaged those four values as the flip time for that day (Lundgren and Wiedenmann 2002). Following these assays, I classified beetles as dead, impaired, or normal. Dead beetles showed no signs of movement when gently prodded with forceps. Impaired beetles showed symptoms of poisoning such as twitching, difficulty walking due to partial paralysis, slow flip-time (> 1 second), or inability to right themselves at all. Normal beetles walked quickly when prodded with forceps, and consistently righted themselves almost instantaneously (flip-time ≤ 1 second). After one week, I removed all remaining slugs or remnants from containers, and maintained beetles with kitten food in a growth chamber as described above. I checked them daily for another week, and then checked twice per week to change food as needed and record beetle mortality.

Statistical analyses

In the experiment with small slugs, I assessed the influence of seed treatment on slug survival by a non-parametric Kruskal-Wallis test in Minitab version 16 (Minitab, Inc., State College, PA). The influence of seed treatment on change in slug mass (as % of starting mass) was

97 tested by one-way analysis of variance (ANOVA). Soybean live biomass after seven days was assessed by two-way ANOVA, with slugs, seed treatment, and their interaction as fixed effects. Both ANOVAs were conducted using the General Linear Model (SPSS 2008). The influence of seed treatments on patterns of slug damage to plants over time was assessed with a mixed model repeated measures analysis (SPSS 2005). Several possible covariance structures for the within- subject correlation were identified, and then evaluated using finite-sample corrected Akaike Information Criteria (AICC) (Wang and Goonewardene 2004, SPSS 2005). On this basis a first- order ante dependence covariance structure was chosen. In the experiment with medium slugs, the analysis differed slightly given that each container had only one slug, and the experiment was divided into three trials. To test whether seed treatment or trial influenced slug survival, I conducted G-tests of independence (Gotelli and Ellison 2004). Because 50% of cells had expected values less than 5, the G statistic was corrected with the Williams adjustment (Cochran 1954, Williams 1976). Slug mass change (as % of starting mass) was analyzed by ANOVA, with seed treatment as a fixed effect and trial included as a random effect to account for variation between trials. Soybean biomass after seven days was cubed to meet distributional assumptions, and then analyzed by ANOVA, with slugs, seed treatments, and their interaction as fixed effects, and trial and its interactions as random effects (GLM, SPSS 2008). I compared patterns of slug damage to plants over time with a mixed model repeated measures analysis (SPSS 2005). Again, the first-order ante dependence covariance structure was chosen to model the within-subject correlation. While the factor “trial” and its interactions with treatment and day were originally included in the model as random factors, this approach over-fit the model because “trial” explained little variation in plant damage; therefore, results are presented with the factor “trial” excluded from the analysis. To test whether slug survival in the presence of C. tricolor differed between treatments, I performed Kaplan-Meier survival analysis (SPSS 2008). The survival distributions for slugs from the four treatments were compared using the non-parametric Mantel-Cox test, stratified by trial to control for variation in predation among trials (similar to Gotthard 2000). To examine the influence of slug consumption on beetles, I pooled into a single “poisoned” category the numbers of beetles impaired or dead in each treatment during the first seven days of the experiment. For those beetles that ate slug tissue during the study, I compared the numbers of poisoned versus normal beetles across trials and treatments using G-tests of independence as above. For those beetles that did not eat slug tissue, sample sizes were too small for statistical comparisons and I only report qualitative results. While flip-time was intended as a

98 continuous measure, its distribution was strongly bimodal with most beetles flipping either in less than one second or not at all; only a few beetles on any given day showed an intermediate response. Flip-time therefore was not analyzed as a continuous variable. For those beetles that eventually recovered from impairment, “days to recovery” was defined as the number of days for a beetle to regain an average flip-time less than or equal to one second. Using a two-sample t-test, I compared the days to recovery for beetles in the low and high thiamethoxam treatments. The effect of slug treatment on beetle survival over time was examined using Kaplan- Meier survival analysis (SPSS 2008). The survival distributions for beetles from the four slug treatments were compared using a Mantel-Cox test, stratified by trial. Many beetles were still alive at the time of writing, so February 15th, 2012 was chosen as an end date for this analysis.

Results

Soybean-slug experiments

A. Small slugs

Survival of small slugs was high across seed treatments (Figure 1A), with no significant differences in slug survival by seed treatment (Kruskal-Wallis H = 6.46, d.f. = 3, P = 0.09). Although the P-value for this test was marginal, there was no discernible pattern in slug survival across treatments (mean rank of slug survival: F+H > F > U > F+L). Slugs gained mass across the study but there were no significant differences in slug mass change by seed treatment (Figure 2A;

Seed treatment: F3,28 = 0.36, P = 0.78). Slugs readily attacked soybean seedlings with all four seed treatments, resulting in no significant differences in the number of seedlings damaged over time

(Figure 3A; Seed treatment: F3,32.2 = 1.21, P = 0.32; Day: F6,42.8 = 92.72, P < 0.001; Day*Seed treatment: F18,42.8 = 1.28, P = 0.25). Slugs reduced soybean biomass relative to the no-slug control, but seed treatments did not influence biomass, with or without slugs present (Figure 4A; Table 3).

99 B. Medium slugs

Survival of medium slugs was high across seed treatments (Figure 1B). I found no significant differences in survival by trial (Gadj = 4.05, d.f. = 2, P = 0.13), so I pooled data across trials to test for the effect of seed treatment. Seed treatments did not significantly influence slug survival (Gadj = 2.95, d.f. = 2, P = 0.40), and although change in slug mass varied across trials (Table 4), there was little evidence that seed treatment influenced change in slug mass within or across trials (Figure 2B; Table 4). Patterns of slug damage to soybean seedlings did not differ significantly by seed treatment (Figure 3B; Table 5). Soybean biomass did not differ significantly by seed treatment, with or without slug feeding (Figure 4B; Table 6).

Slug-ground beetle experiments

Overall, 75% of C. tricolor adults killed the slug with which they were confined, with most predation occurring by the morning after slugs were introduced (Figure 5). The rate of beetle predation on slugs did not differ significantly between slug treatments (Mantel-Cox, stratified by trial, χ2 = 2.22, d.f. = 3, P = 0.53). In contrast, a majority of beetles that fed upon slugs from the low and high thiamethoxam treatments appeared poisoned (Figure 6). Symptoms ranged from twitching and mild motor difficulty, to partial paralysis (especially of hind legs), to extensive paralysis and death. Trial did not significantly influence the likelihood of beetle poisoning (G = 0.053, d.f. = 2, P = 0.97), so data were pooled across trials to test for the effect of slug treatment. Slug feeding history significantly influenced beetle poisoning (Gadj = 29.02, d.f. = 3, P < 0.001). The proportions of beetles impaired in high and low thiamethoxam treatments were identical (8 impaired / 13 total = 0.62; Figure 6), while no beetles were impaired in the untreated and fungicide-only treatments. Post-hoc tests were therefore redundant and were not conducted. Among the 25% of beetles that did not consume slugs during the study, symptoms of intoxication were rare (Figure 6). The single exception was a beetle that died after attacking, but not killing, a slug from the low thiamethoxam treatment. This beetle displayed symptoms of poisoning before it died including labored walking and inability to right itself. Five beetles in the high thiamethoxam treatment and four beetles in the low thiamethoxam treatment eventually recovered after initial impairment. Days to recovery did not differ significantly in these two groups (F+H: 4.8 ± 1.1 days; F+L: 3.8 ± 1.2 days; t = 1.34, df = 7, P = 0.22). Beetle survival curves differed

100 significantly depending on the feeding history of their slug prey (Figure 7; Mantel-Cox, stratified by trial: χ2 = 8.3, d.f. = 3, P = 0.04). Pairwise comparisons revealed that beetles in the low thiamethoxam treatment died significantly faster than beetles in the untreated and fungicide-only controls (F+L v. U: P = 0.03, F+L v. F: P = 0.02), while beetle survival in the high thiamethoxam treatment did not differ significantly from any other treatment.

Discussion

Thiamethoxam seed treatment at low (0.0756 mg ai/seed) or high (0.152 mg ai/seed) commercial rates, combined with common fungicides, did not influence the feeding behavior of D. reticulatum on soybean seedlings. Plant damage, plant biomass, slug survival, and change in slug mass after one week of feeding were all similar across the seed treatments. These results are in general agreement with previous evidence that neonicotinoids have low acute toxicity to slugs (Iglesias et al. 2002, Simms et al. 2006). Nonetheless, imidicloprid appeared to reduce slug pressure to wheat, especially at high doses (Rose and Oades 2001, Simms et al. 2006). The failure of thiamethoxam to deter slugs in this study could be a result of several factors, including insecticide dose and differences between imidacloprid and thiamethoxam. Alternatively, variability in slug response to NSTs may emerge from interactions between neonicotinoids and particular crop species. For instance, despite its apparent deterrent effects in wheat, imidacloprid seed treatment failed to protect canola seedlings from slug damage (Simms et al. 2006). Neonicotinoids have recently been shown to induce salicylic acid-associated plant responses, leading to widespread changes in plant biochemistry (Ford et al. 2010). These changes in plant quality can in turn mediate the response of non-target herbivores to neonicotinoids (Szczepaniec 2009). Further research is needed to understand the changes in the physiology of various crop plants as a result of NSTs, and the consequences for non-target herbivores including slugs. While seed treatments did not influence slug feeding behavior, slugs that fed for one week on thiamethoxam-treated soybean seedlings were toxic to the majority of C. tricolor individuals that consumed them. I did not directly quantify the amount of insecticides or their metabolites in plants, slugs, or beetles; yet, the symptoms I observed—uncoordinated movement, twitching, paralysis, and death—were consistent with neurotoxicity from neonicotinoid exposure. Similar symptoms have been reported for ground beetles sprayed directly with imidacloprid (Kunkel et al. 2001), coccinellids confined on imidacloprid-treated sunflowers (Smith and

101 Krischik 1999), coccinellids feeding on thiamethoxam- or clothianidin-treated corn seedlings (Moser and Obrycki 2009), and predatory beetles foraging on imidacloprid-treated hemlock branches (Eisenback et al. 2010). In this study, beetle intoxication occurred at both insecticide doses, and survivorship was reduced in the low thiamethoxam treatment, with a similar but non- significant trend in the high thiamethoxam treatment. Fungicides alone did not lead to toxicity in beetles, consistent with earlier results for mefenoxam/fludioxonil (Mullin et al. 2005). I cannot rule out the possibility that fungicides synergized thiamethoxam to increase its toxicity, as fungicides have been shown to do in other systems (Pedersen et al. 2003, Schmuck et al. 2003). For those beetles that survived impairment, recovery of normal locomotory function took several days. In the field, beetles with poor coordination or paralysis would likely be at risk from their own predators. For example, the ground beetle, Harpalus pensylvanicus, became highly susceptible to ant predation following sublethal exposure to imidacloprid (Kunkel et al. 2001). In addition to increasing the vulnerability of natural enemies to predation, sublethal effects of insecticides can disrupt behaviors that are important to biocontrol (Desneux et al. 2007). Parasitoid wasps, for instance, that usually rely on volatile cues to locate their caterpillar hosts showed reduced attraction to these cues after feeding on extrafloral nectar from imidacloprid- treated cotton (Stapel et al. 2000). The foraging behavior of ground beetles on slugs is poorly studied, but at least one species appears to use olfactory cues to locate slug prey both in the adult (McKemey et al. 2004) and larval (Thomas et al. 2008) stages. Understanding the sublethal effects of NSTs on predator foraging behavior is an important area for further study. Contrary to my hypothesis, beetle poisoning, rates of recovery, and longevity were similar regardless of whether slug prey were subject to low or high thiamethoxam treatments. The equivalent effect of these two doses on C. tricolor suggests that the concentration of insecticide beetles received via slugs is either equivalent for the two doses or that the low dose is sufficient to cause harm and higher doses cause no greater effects. This latter possibility would be consistent with the low dose/high toxicity nature of neonicotinoids that provides some advantages relative to other classes of insecticides (Elbert et al. 2008). The absence of a dose effect was observed also with H. pensylvanicus after consuming food pellets sprayed at two rates of imidacloprid (0.168 kg ai/ha and 0.336 kg ai/ha; Kunkel et al. 2001). Where dose effects have been observed, they involved a much larger range of doses (1 to 100 ppm imidacloprid applied to hemlock branches; Eisenback et al. 2010).

102 It is also possible that potential dose effects were diluted by several sources of variability. Slugs ate a varying amount of soybean tissue, beetles ate varying amounts and locations of slug tissue, and beetles varied in the time between slug introduction and attack. We do not yet know how long neonicotinoids or their residues persist in slug tissues, but in general neonicotinoids are highly water soluble and are rapidly excreted in vertebrates (EPA 2002, Tomizawa and Casida 2005). Anecdotally my results suggest that neonicotinoids may also be rapidly excreted by slugs; beetles that ate slugs from the high thiamethoxam treatment showed high levels of poisoning if they ate slugs within the first 13 hours (8/10 beetles intoxicated), but not if they ate slugs several days after introduction (0/3 beetles intoxicated). Studies that quantify thiamethoxam and its metabolites (especially clothianidin, Nauen et al. 2003) as it travels through the plant – slug – beetle food chain will help to resolve this question. Beetles preyed on slugs at roughly equal rates regardless of the seed treatments on which the slugs were fed. This suggests that C. tricolor lacks a behavioral mechanism to avoid slugs that have fed upon neonicotionid-treated plant tissue. Similarly, the ground beetle, H. pensylvanicus, failed to avoid imidacloprid-treated food pellets (Kunkel et al. 2001), and several ground beetle species failed to discriminate between normal prey and prey treated with dimethoate (Mauchline et al. 2004). Coccinellid larvae also spent similar amounts of time on neonicotinoid-treated corn seedlings versus controls (Moser and Obrycki 2009). In contrast, ground beetles took longer to begin feeding on neonicotinoid-treated corn seeds versus seeds treated only with fungicides (Mullin et al. 2005). Treated seeds with the insecticide in high concentration on the surface may present a more apparent set of sensory cues than a prey item in which the insecticide or its residues may be present on the interior of the animal, or in lower concentration. The failure of C. tricolor to avoid contaminated slugs would tend to increase the risk of toxicity through this route of exposure. In addition to documented routes of direct contact exposure to neonicotinoid insecticides through soil or dust (Girolami et al. 2012, Krupke et al. 2012), the systemic nature of neonicotinoids would seem to make ingestion an important mode of non-target exposure. Indeed, my results join a growing body of evidence that natural enemies can be exposed to neonicotinoids through their diet. Omnivorous natural enemies may be endangered when they feed upon leaves, nectar, or pollen of treated plants (Lundgren 2009, Moser and Obrycki 2009, Prabhaker et al. 2011, Seagraves and Lundgren 2011), or on the treated seeds themselves (Mullin et al. 2005). Natural enemies may also imbibe neonicotinoids if they obtain water through guttation drops (Girolami et al. 2009). In some cases, concentrations of neonicotinoids that are sublethal to the

103 target pest are sufficient to impair natural enemies that feed upon these pests. The coccinellid beetle, Hippodamia undecimnotata, showed reductions in several fitness parameters when it fed upon aphids reared on imidacloprid-treated faba bean (Papachristos and Milonas 2008). Similarly, toxicity of neonicotinoid-treated citrus branches to the vedalia beetle Rodolia cardinalis was enhanced by cottony cushion scale prey (Grafton-Cardwell and Gu 2003). Parasitism rates of the whitefly, Bemisia tabaci, were reduced by imidacloprid and thiamethoxam seed treatment in cotton (Naveed et al. 2010). In other cases, as in this study, neonicotinoids or their metabolites are transmitted through herbivores that are not targets of the insecticide. Spider mites, for instance, are not susceptible to neonicotinoids and their populations on diverse plant hosts have been enhanced following treatment with imidacloprid (Sclar et al. 1998, James and Price 2002, Zeng and Wang 2010, Szczepaniec et al. 2011). It appears that multiple mechanisms are responsible for this effect, including transmission of toxins from spider mites to their predators (James 2003, Szczepaniec et al. 2011), along with changes in plant quality (Szczepaniec 2009) and direct enhancement of mite fecundity (James and Price 2002). Although natural enemies may encounter neonicotinoids through their diet, the likelihood of these exposures under field conditions is not well characterized, especially for NSTs. The few field studies to examine the non-target effects of NSTs on ground-dwelling predators have reached mixed conclusions across studies, years, and cropping systems. In Catalonia, imidacloprid seed treatment in maize did not influence pitfall catches of ground beetles or most other natural enemies (Albajes et al. 2003, de la Poza et al. 2005). In contrast, imidacloprid reduced the abundance of ground beetles in soil samples in one out of three years in Illinois maize plots, although not in pitfall traps (Bhatti et al. 2005), and NSTs reduced ground beetle activity- density in one out of two years in minimum-tilled maize in Pennsylvania (Leslie et al. 2010). In the latter study, ground beetle communities in plots with NSTs and control communities diverged shortly after planting, and then became more similar later in the season, suggesting that NSTs had a greater influence on ground beetles early in the season (Leslie et al. 2010). In sugar beets in Germany, imidicloprid seed treatments similarly depressed early-season ground beetle activity (Epperlein et al. 2001). In contrast, several NSTs in the U.K. failed to significantly decrease ground beetle catches, although the authors noted a numerical trend for fewer beetles in the treated plots (Baker et al. 2002). These field studies lead to the conclusion that NSTs can negatively influence ground beetles, but do not always do so. One factor contributing to this variability could be differences in the composition of prey taxa in different agroecosystems, varying in their ability to withstand and

104 transmit neonicotinoids to higher trophic levels. For this reason, generalizing the risks of systemic insecticides across cropping systems and regions may be misleading or unproductive. Given the potential for slugs to transmit toxins from treated seedlings to ground beetles as demonstrated here, field studies are warranted in slug-prone, no-till environments to examine the influence of NSTs on the community of ground-dwelling predators.

Conclusion

The results of this and previous studies suggest that NSTs have little utility against slugs; furthermore, NSTs may actually exacerbate slugs and other pests through transfer of toxins to generalist predators. Field studies are needed to better characterize this risk, especially in no-till systems. While NSTs are undoubtedly a powerful pest management tool in some circumstances, I agree with several authors that their reflexive use in soybeans deserves re-evaluation (McCornack and Ragsdale 2006, Johnson et al. 2008, Steffey 2008, Johnson et al. 2009, Magalhaes et al. 2009, Seagraves and Lundgren 2011, Tinsley et al. 2011). Furthermore, mollusks are an important part of the pest fauna around the world in a wide array of vegetable, fruit, and field crops (South 1992, Barker 2002), many of which are treated with neonicotinoids. Additional lab and field studies are needed to assess the compatibility of neonicotinoids and mollusk biocontrol in these systems.

105

References

Albajes, R., C. López, and X. Pons. 2003. Predatory fauna in cornfields and response to imidacloprid seed treatment. Journal of Economic Entomology 96(6): 1805-1813. Baker, P., L. A. Haylock, B. H. Garner, R. J. N. Sands, and A. M. Dewar. 2002. The effects of insecticide seed treatments on beneficial invertebrates in sugar beet. pp. 653-658 in: Proceedings of the BCPC Conference – Pests & Diseases 2002, Brighton, UK. Barker, G. M. (ed.) 2002. Molluscs as Crop Pests. CABI Publishing, Wallingford, UK. Barratt, B. I. P., R. A. Byers, and D. L. Bierlein. 1994. Conservation tillage crop yields in relation to grey garden slug [Deroceras reticulatum (Müller)] (Mollusca: Agriolimacidae) density during establishment. Crop Protection 13(1): 49-52. Bhatti, M. A., J. Duan, G. P. Head, C. Jiang, M. J. McKee, T. E. Nickson, C. L. Pilcher, and C. D. Pilcher. 2005a. Field evaluation of the impact of corn rootworm (Coleoptera: Chrysomelidae)-protected Bt corn on ground-dwelling invertebrates. Environmental Entomology 34(5): 1325-1335. Bohan, D. A., A. C. Bohan, D. M. Glen, W.O.C. Symondson, C. W. Wiltshire, and L. Hughes. 2000. Spatial dynamics of predation by carabid beetles on slugs. Journal of Animal Ecology 69(3): 367-379. Bonmatin, J. M., I. Moineau, R. Charvet, C. Fleche, M. E. Colin, and E. R. Bengsch. 2003. A LC/APCI-MS/MS method for analysis of imidacloprid in soils, in plants, and in pollen. Analytical Chemistry 75: 2027-2033. Bonmatin, J. M., P. A. Marchand, R. Charvet, I. Moineau, E. R. Bengsch, and M. E. Colin. 2005. Quantification of imidacloprid uptake in maize crops. Journal of Agricultural and Food Chemistry 53: 5336-5341. Bousquet, Y. 2010. Illustrated Identification Guide to Adults and Larvae of Northeastern North American Ground Beetles (Coleoptera: Carabidae). Pensoft Series Faunistica No. 90 Pensoft Publishers, Sofia-Moscow, Bulgaria. Brust, G. E. and G. J. House. 1990. Effects of soil moisture, no-tillage and predators on southern corn rootworm (Diabrotica undecimpunctata howardi) survival in corn agroecosystems. Agriculture, Ecosystems and Environment 31: 199-216.

106 Brust, G. E., B. R. Stinner, and D. A. McCartney. 1985. Tillage and soil insecticide effects on predator-black cutworm (Lepidoptera: Noctuidae) interactions in corn agroecosystems. Journal of Economic Entomology 78(6): 1389-1392. Brust, G. E., B. R. Stinner, and D. A. McCartney. 1986. Predator activity and predation in corn agroecosystems. Environmental Entomology 15(5): 1017-1021. Cochran, W. G. 1954. Some methods for strengthening the common χ2 tests. Biometrics 10(4): 417-451. de la Poza, M., X. Pons, G. P. Farinos, C. Lopez, F. Ortego, M. Eizaguirre, P. Castanera, and R. Albajes. 2005. Impact of farm-scale Bt maize on abundance of predatory arthropods in Spain. Crop Protection 24: 677-684. Desneaux, N., A. Decourtye, and J.-M. Delpuech. 2007. The sublethal effects of pesticides on beneficial arthropods. Annual Review of Entomology 52: 81-106. Eisenback, B. M., S. M. Salom, L. T. Kok, and A. F. Lagalante. 2010. Lethal and sublethal effects of imidacloprid on hemlock woolly adelgid (Hemiptera: Adelgidae) and two introduced predator species. Journal of Economic Entomology 103(4): 1222-1234. Elbert, A. M. Haas, B. Springer, W. Thielert, and R. Nauen. 2008. Applied aspects of neonicotinoid uses in crop protection. Pest Management Science 64: 1099-1105. Environmental Protection Agency. 2002. Thiamethoxam: pesticide tolerance. Federal Register 67: 66561-66571. Epperlein, K. and H.-W. Schmidt. 2001. Effects of pelleting sugar-beet seed with Gaucho® (imidacloprid) on associated fauna in the agricultural ecosystem. Pflanzenschutz- Nachrichten Bayer 54(3): 369-398. Eskelson, M. J., E. G. Chapman, D. D. Archbold, J. J. Obrycki, and J. D. Harwood. 2011. Molecular identification of predation by carabid beetles on exotic and native slugs in a strawberry agroecosystem. Biological Control 56(3): 245 – 253. Ford, K. A., J. E. Casida, D. Chandran, A. G. Guelvich, R. A. Okrent, K. A. Durkin, R. Sarpong, E. M. Bunnelle, and M. C. Wildemuth. 2010. Neonicotinoid insecticides induce salicylate-associated plant defense responses. Proceedings of the National Academy of Sciences USA 107(41): 17527-17532. Girolami, V., L. Mazzon, A. Squartini, N. Mori, M. Marzaro, A. di Bernardo, M. Greatti, C. Giorio, and A. Tapparo. 2009. Translocation of neonicotinoid insecticides from coated seeds to seedling guttation drops: A novel way of intoxication for bees. Journal of Economic Entomology 102(5): 1808-1815.

107 Girolami, V., M. Marzaro, L. Vivan, L. Mazzon, M. Greatti, C. Giorio, D. Marton, and A. Tapparo. 2012. Fatal powdering of bees in flight with particulates of neonicotinoids seed coating and humidity implication. Journal of Applied Entomology 136: 17-26. Gotelli, N. J. and A. M. Ellison. 2004. The analysis of categorical data. pp. 350-382 in: A Primer of Ecological Statistics. Sinauer Associates, Inc.: Sunderland, MA. Gotthard, K. 2000. Increased risk of predation as a cost of high growth rate: An experimental test in a butterfly. Journal of Animal Ecology 69(5): 896-902. Grafton-Cardwell, E. E. and P. Gu. 2003. Conserving vedalia beetle, Rodolia cardinalis (Mulsant) (Coleoptera: Coccinellidae), in citrus: A continuing challenge as new insecticides gain registration. Journal of Economic Entomology 96(5): 1388-1398. Gray, M. E. and D. W. Onstad. 2008. Increasing corn acres and prophylactic use of Bt hybrids: Implications for IPM and IRM. In: Proceedings of the 2008 Illinois Crop Protection Technology Conference, University of Illinois Extension. Gregory, W. W. and G. J. Musick. 1976. Insect management in reduced tillage systems. Bulletin of the Entomological Society of America 22(3): 302-304. Hammond, R. B. 1997. Long-term conservation tillage studies: Impact of no-till on seedcorn maggot (Diptera: Anthomyiidae). Crop Protection 16(3): 221-225. Hammond, R. B. 2000. Simulation of moderate levels of slug injury to soybean. Crop Protection 19: 113-120. Hammond, R. B. and R. A. Byers. 2002. Agriolimacidae and Arionidae as pests in conservation- tillage soybean and maize cropping in North America. pp. 301-314 in: Barker, ed. Molluscs as Crop Pests. CAB International, Wallingford, UK. Horowitz, J., R. Ebel, and K. Ueda. 2010. “No-till” farming is a growing practice. U.S. Department of Agriculture, Economic Research Service, Bulletin #70. House, G. J. and B. R. Stinner. 1983. Arthropods in no-tillage soybean agroecosystems: Community composition and ecosystem interactions. Environmental Management 7(1): 23-28. House, G. J. and M. D. R. Alzugaray. 1989. Influence of cover cropping and no-tillage practices on community composition of soil arthropods in a North Carolina Agroecosystem. Environmental Entomology 18(2): 302-307. Iglesias, J., J. Castillejo, A. Ester, R. Castro, and M. J. Lombardia. 2002. Susceptibility of the eggs of the field slug Deroceras reticulatum to contact with pesticides and substances of biological origin on artificial soil. Annals of Applied Biology 140: 53 – 59.

108 James, D. G. 2003. Toxicity of imidacloprid to Galendromus occidentalis, Neoseiulus fallacies and Amblyseius andersoni (Acari: Phytoseiidae) from hops in Washington State, USA. Experimental and Applied Acarology 31: 275-281. James, D. G. and T. S. Price. 2002. Fecundity in twospotted spider mite (Acari: Tetranychidae) is increased by direct and systemic exposure to imidacloprid. Journal of Economic Entomology 95(4): 729-732. Jeschke, P. and R. Nauen. 2008. Neonicotinoids – from zero to hero in insecticide chemistry. Pest Management Science 64: 1084-1098. Jeschke, P., R. Nauen, M. Schindler, and A. Elbert. 2011. Overview of the status and global strategy for neonicotinoids. Journal of Agricultural and Food Chemistry 59: 2897 – 2908. Johnson, K. D., M. E. O’Neal, J. D. Bradshaw, and M. E. Rice. 2008. Is preventative, concurrent management of the soybean aphid (Hemiptera: Aphididae) and bean leaf beetle (Coleoptera: Chrysomelidae) possible? Journal of Economic Entomology 101(3): 801- 809. Johnson, K. D., M. E. O’Neal, D. W. Ragsdale, C. D. Difonzo, S. M. Swinton, P. M. Dixon, B. D. Potter, E. W. Hodgson, and A. C. Costamagna. 2009. Probability of cost-effective management of soybean aphid (Hemiptera: Aphididae) in North America. Journal of Economic Entomology 102(6): 2101-2108. Krupke, C. H., G. J. Hunt, B. D. Eitzer, G. Andino, and K. Given. 2012. Multiple routes of pesticide exposure for honey bees living near agricultural fields. PLoS One 7(1): e29268. doi:10.1371/journal.pone.0029268. Kunkel, B. A., D. W. Held, and D. A. Potter. 2001. Lethal and sublethal effects of bendiocarb, halofenozide, and imidacloprid on Harpalus pensylvanicus (Coleoptera: Carabidae) following different modes of exposure in turfgrass. Journal of Economic Entomology 94(1): 60-67. Landis, D. A., J. R. Bradley, and F. Gould. 1987. Behavior and survival of Heliothis zea (Lepidoptera: Noctuidae) prepupae in no-tillage and conventional-tillage corn. Environmental Entomology 16(1): 94-99. 14 Laurent, F. M. and E. Rathahao. 2003. Distribution of [ C]Imidacloprid in sunflowers (Helianthus annuus L.) following seed treatment. Journal of Agricultural and Food Chemistry 51: 8005-8010.

109 Leslie, T. W., D. J. Biddinger, J. R. Rohr, and S. J. Fleischer. 2010. Conventional and seed-based insect management strategies similarly influence nontarget coleopteran communities in maize. Environmental Entomology 39(6): 2045-2055. Lundgren, J. G. 2009. Plant-incorporated pest resistance and natural enemies. pp. 309-331 in: Relationships of Natural Enemies and Non-Prey Foods. Progress in Biological Control 7. Springer Science + Business Media B. V. Lundgren, J. G. and R. N. Wiedenmann. 2002. Coleopteran-specific Cry3Bb toxin from transgenic corn pollen does not affect the fitness of a nontarget species, Coleomegilla maculata DeGeer (Coleoptera: Coccinellidae). Environmental Entomology 31(6): 1213- 1218. Lundgren, J. G., M. E. Ellsbury, and D. A. Prischmann. 2009. Analysis of the predator community of a subterranean herbivorous insect based on polymerase chain reaction. Ecological Applications 19(8): 2157-2166. Magalhaes, L. C., T. E. Hunt, and B. D. Siegfried. 2009. Efficacy of neonicotinoid seed treatments to reduce soybean aphid populations under field and controlled conditions in Nebraska. Journal of Economic Entomology 102(1): 187-195. Mauchline, A. L., J. L. Osborne, and W. Powell. 2004. Feeding responses of carabid beetles to dimethoate-contaminated prey. Agricultural and Forest Entomology 6: 99-104. McCormack, B. P. and D. W. Ragsdale. 2006. Efficacy of thiamethoxam to suppress soybean aphid populations in Minnesota soybean. Crop Management DOI: 10.1094/CM-2006- 0915-01-RS. McKemey, A. R., D. M. Glen, and W. O. C. Symondson. 2004. How does a carabid predator find aggregations of slugs in the field? Electroantennograms and behavioral assays suggest chemical cues. Bulletin of Entomological Research 94: 235-244. Moser, S. E. and J. J. Obrycki. 2009. Non-target effects of neonicotinoid seed treatments; mortality of coccinellid larvae related to zoophytophagy. Biological Control 51: 487-492. Mullin, C. A., M. C. Saunders, II, T. W. Leslie, D. J. Biddinger, and S. J. Fleischer. 2005. Toxic and behavioral effects to Carabidae of seed treatments used on Cry3Bb1 and Cry1Ab/c protected corn. Environmental Entomology 34(6): 1626 – 1636. Musick, G. J. 1985. Management of arthropod pests in conservation-tillage systems in the Southeastern U.S. pp. 191-204 In: W. L. Hargrove, F. C. Boswell, and G. W. Langdale (eds.) Proceedings of the Southern Region No-Till Conference. July 16-17, 1985. Griffin, Georgia.

110 Nauen, R., U. Ebbinghaus-Kintscher, V. L. Salgado, and M. Kaussmann. 2003. Thiamethoxam is a neonicotinoid precursor converted to clothianidin in insects and plants. Pesticide Biochemistry and Physiology 76: 55-69. Naveed, M., A. Salam, M. A. Saleem, M. Rafiq, and A. Hamza. 2010. Toxicity of thiamethoxam and imidacloprid as seed treatments to parasitoids associated to control Bemisia tabaci. Pakistan Journal of Zoology 42(5): 559-565. Papachristos, D. P. and P. G. Milonas. 2008. Adverse effects of soil applied insecticides on the predatory coccinellid Hippodamia unidecimnotata (Coleoptera: Coccinellidae). Biological Control 47: 77-81. Pedersen, W. L., J. D. Kline, C. A. Bradley, and D. S. Mueller. 2003. Influence of mealaxyl fungicide seed treatment on severity of rootworm (Diabrotica spp.) damage to corn (Zea mays) under no-tillage conditions. Crop Protection 22: 647-652. Prabhaker, N., S. J. Castle, S. E. Naranjo, N. C. Toscano, and J. G. Morse. 2011. Compatibility of two systemic neonicotinoids, imidacloprid and thiamethoxam, with various natural enemies of agricultural pests. Journal of Economic Entomology 104(3): 773-781. Rose, P. W. and L. Oades. 2001. Effects of imidacloprid cereal seed treatment against wireworms and slugs. In: Proceedings No. 76, Seed Treatment: Challenges and Opportunities. British Crop Protection Council, Farnham, pp. 191 – 196. Schmuck, R., R. Stadler, and H. W. Schmidt. 2003. Field relevance of a synergistic effect observed in the laboratory between an EBI fungicide and a choronicotinyl insecticide in the honeybee (Apis mellifera L., Hymenoptera). Pest Management Science 59(3): 279- 286. Sclar, C. D., D. Gerace, and W. S. Crenshaw. 1998. Observations of population increases and injury by spider mites (Acari: Tetranychidae) on ornamental plants treated with imidacloprid. Journal of Economic Entomology 91(1): 250-255. Seagraves, M. P. and J. G. Lundgren. 2011. Effects of neonicotinoid seed treatments on soybean aphid and its natural enemies. Journal of Pest Science [early online publication]. Simms, L. C., A. Ester, and M. J. Wilson. 2006. Control of slugs damage to oilseed rape and wheat with imidacloprid seed dressings in laboratory and field experiments. Crop Protection 25: 549 – 555. Smith, S. F. and V. A Krischik. 1999. Effects of systemic imidacloprid on Coleomegilla maculata (Coleoptera: Coccinellidae). Environmental Entomology 28(6): 1189-1195.

111 South, A. 1992. Terrestrial slugs: Biology, ecology, and control. Chapman and Hall: New York, NY. SPSS. 2005. Linear mixed-effects modeling in SPSS: An introduction to the MIXED procedure. Chicago, IL: SPSS, Inc. SPSS. 2008. SPSS Statistics Base 17.0 User’s Guide and SPSS Advanced Statistics 17.0. Chicago, IL: SPSS, Inc. Stapel, J. O., A. M. Cortesero, and W. J. Lewis. 2000. Disruptive sublethal effects of insecticides on biological control: Altered foraging ability and life span of a parasitoid after feeding on extrafloral nectar of cotton treated with systemic insecticides. Biological Control 17: 243-249. Steffey, K. L. 2008. Managing insects in high-production soybeans: Forethought or afterthought? In: Proceedings of the 2008 Illinois Crop Protection Technology Conference, University of Illinois Extension. Stinner, B. R. and G. J. House. 1990. Arthropods and other invertebrates in conservation-tillage agriculture. Annual Review of Entomology 35: 299-318. Stinner, B. R., D. A. McCartney, and D. M. Van Doren, Jr. 1988. Soil and foliage arthropod communities in conventional, reduced and no-tillage corn (Maize, Zea mays L.) systems: A comparison after 20 years of continuous cropping. Soil & Tillage Research 11: 147- 158. Symondson, W.O.C. 1993. Chemical confinement of slugs: an alternative to electric fences. Journal of Molluscan Studies 59: 259-261. Symondson, W.O.C., D. M. Glen, C. W. Wiltshire, C. J. Langdon, and J. E. Liddell. 1996. Effects of cultivation techniques and methods of straw disposal on predation by Pterostichus melanarius (Coleoptera: Carabidae) upon slugs (Gastropoda: Pulmonata) in an arable field. Journal of Applied Ecology 33(4): 741 – 753. Symondson, W.O.C., D. M. Glen, A. R. Ives, C. J. Langdon, and C. W. Wiltshire. 2002. Dynamics of the relationship between a generalist predator and slugs over five years. Ecology 83(1): 137-147. Szczepaniec, A. 2009. Mechanisms underlying outbreaks of spider mites following applications of imidaceloprid. Ph.D. dissertation. University of Maryland, Department of Entomology. Szczepaniec, A., S. F. Creary, K. L. Laskowski, J. P. Nyrop, and M. J. Raupp. 2011. Neonicotionid insecticide imidacloprid causes outbreaks of spider mites on elm trees in urban landscapes. PLoS One 6(5): e20018. doi:10.1371/journal.pone.0020018.

112 Thomas, R. S., D. M. Glen, and W. O. C. Symondson. 2008. Prey detection through olfaction by the soil-dwelling larvae of the carabid predator Pterostichus melanarius. Soil Biology & Biochemistry 40: 207-216. Thomas, A.K., R.J. McDonnell, and J.D. Harwood. 2011. Slugs from the Nearctic: what we need to learn from the Western Palearctic. Proceedings of the IOBC/WPRS workgroup on Slugs and Snails: Slugs and Snail Control in the 21st Century. Tinsley, N. A., K. L. Steffey, R. E. Estes, J. R. Heeren, M. E. Gray, and B. W. Diers. 2011. Field- level effects of preventative management tactics on soybean aphids (Aphis glycines Matsumura) and their predators. Journal of Applied Entomology [early view, first published online 1 Aug. 2011]. Tomizawa, M. and J. E. Casida. 2005. Neonicotinoid insecticide toxicology: Mechanisms of selective action. Annual Review of Pharmacology and Toxicology 45: 247-268. U. S. Department of Agriculture - National Agricultural Statistics Service. 2010. News Release: Tillage Practices with Updated Alfalfa Seedings. Accessed 1/24/12 at: http://www.nass.usda.gov/Statistics_by_State/Pennsylvania/Publications/Survey_Results/ index.asp Wang, Z. and L. A. Goonewardene. 2004. The use of MIXED models in the analysis of animal experiments with repeated measures data. Canadian Journal of Animal Science 84: 1-11. Wardle, D. A. 1995. Impacts of disturbance on detritus food webs in agro-ecosystems of contrasting tillage and weed management practices. Advances in Ecological Research 26: 105-185. Williams, D. A. 1976. Improved likelihood ratio tests for complete contingency tables. Biometrika 63(1): 33-37. Zeng, C.-X. and J.-J. Wang. 2010. Influence of exposure to imidacloprid on survivorship, reproduction and vitellin content of the carmine spider mite, Tetranychus cinnabarinus. Journal of Insect Science 10(20): available online: insectsicence.org/10.20.

113

Tables

Table 4-1. Sample sizes for the soybean-slug study with medium D. reticulatum and soybean seedlings, carried out in 3 trials.

Soybean seedlings Soybean seedlings + medium slug – no slug control Trial number 1 2 3 Total 1 2 3 Total Untreated 12 12 10 34 12 6 6 24 Fungicide-only 12 12 10 34 12 6 6 24 Fungicide + low rate thiamethoxam 12 12 10 34 12 6 6 24 Fungicide + high rate thiamethoxam 12 12 10 34 12 6 6 24

Table 4-2. Sample sizes for the slug-beetle study with medium D. reticulatum and C. tricolor, carried out in 3 trials.

Medium slug + C. tricolor Trial number 1 2 3 Total Untreated 8 6 5 19 Fungicide-only 8 6 3 17 Fungicide + low rate thiamethoxam 8 6 4 18 Fungicide + high rate thiamethoxam 9 5 4 18

Table 4-3. ANOVA table for tests of the effects of slugs and seed treatment on soybean biomass at the end of seven days (small slug experiment).

Source of variation d.f. SS MS F P Slug 1 0.22 0.215 29.65 <0.001 Seed treatment 3 0.01 0.003 0.42 0.74 Slug*Seed treatment 3 0.01 0.005 0.68 0.57 Error 56 0.41 0.007

114 Table 4-4. ANOVA table for tests of the effects of seed treatment and trial on slug mass change (medium slug experiment).

Source of variation d.f. SS MS F P Trial 2 3766.9 1883.5 6.44 0.03 Seed treatment 3 456.2 152.1 0.52 0.67 Trial*Seed treatment 6 1743.2 290.5 0.53 0.78 Error 110 60405.1 549.1

Table 4-5. Linear mixed model repeated measures analysis to test for the effect of seed treatment on soybean damage over seven days of slug feeding (medium slug experiment).

Source of variation Num. d.f. Denom. d.f.* F P Seed treatment 3 123.6 0.36 0.78 Day 6 179.9 264.8 <0.001 Day*Seed treatment 18 179.9 0.74 0.77

*calculated via Satterthwaite approximation

Table 4-6. ANOVA table for tests of the effects of slugs and seed treatment on soybean biomass at the end of seven days of slug feeding (medium slug experiment).

Source of variation d.f. SS MS F P Slug 1 0.22 0.222 30.16 0.03 Seed treatment 3 0.01 0.003 0.45 0.73 Slug*Seed treatment 3 0.003 0.001 0.45 0.73 Trial 2 0.07 0.033 2.91 0.17 Trial*Slug 2 0.01 0.007 3.21 0.11 Trial*Seed treatment 6 0.04 0.006 2.69 0.13 Trial*Slug*Seed treatment 6 0.01 0.002 0.95 0.46 Error 194 0.47 0.002

115

Figures

Figure 4-1. Mean numbers (A) and proportion (B) of slugs surviving by seed treatment: U = Untreated; F = Fungicide only; F+L = Fungicide + low rate thiamethoxam; F+H = Fungicide + high rate thiamethoxam. Data were collected in (A) spring, with 4 small slugs per container (n = 8) and, (B) fall, with one medium slug per container (n = 34). Error bars show one standard error.

Figure 4-2. Slug mass gain (mean % ± SE) by seed treatment: U = Untreated; F = Fungicide only; F+L = Fungicide + low rate thiamethoxam; F+H = Fungicide + high rate thiamethoxam. Data were collected in (A) spring, with 4 small slugs per container (n = 8) and, (B) fall, with one medium slug per container (n = 29-32).

116

Figure 4-3. Number of soybean plants (out of 4) damaged by slugs (mean ± SE) as a function of time and seed treatment: U = Untreated; F = Fungicide only; F+L = Fungicide + low rate thiamethoxam; F+H = Fungicide + high rate thiamethoxam. Data were collected in (A) spring, with 4 small slugs per container (n = 8) and, (B) fall, with one medium slug per container (n = 29- 32).

Figure 4-4. Soybean biomass (mean ± SE) at the end of seven days, by slugs and seed treatments: U = Untreated; F = Fungicide only; F+L = Fungicide + low rate thiamethoxam; F+H = Fungicide + high rate thiamethoxam. Data were collected in (A) spring, with four small slugs per container (n = 8) and, (B) fall, with one medium slug per container (n = 24 for no slug controls; n = 29-32 for slug treatments). Means that do not share a letter are different at P ≤ 0.05.

117

Figure 4-5. Proportion of slugs surviving in containers with C. tricolor, as a function of time and slug treatment: U = Untreated (n = 19); F = Fungicide only (n = 17); F+L = Fungicide + low rate thiamethoxam (n = 18); F+H = Fungicide + high rate thiamethoxam (n = 18).

Figure 4-6. Symptoms of beetle posioning seen during seven days after exposure to slugs that had fed on soybeans with various seed treatments: U = Untreated (n = 19); F = Fungicide only (n = 17); F+L = Fungicide+ low rate thiamethoxam (n = 18); F+H = Fungicide + high rate thiamethoxam (n = 18).

118

Figure 4-7. Proportion of C. tricolor surviving by treatment of its slug prey: U = Untreated (n = 19); F = Fungicide only (n = 17); F+L = Fungicide + low rate thiamethoxam (n = 18); F+H = Fungicide + high rate thiamethoxam (n = 18). Survival curves that do not share a letter in the key are different at P ≤ 0.05.

Appendix A

Supplemental data on the natural history of slugs in Pennsylvania field crops

To build knowledge about slug natural history in Pennsylvania field crops, I monitored slugs under shelter traps in maize and alfalfa plots over two years, in the experiment described in Chapter 3. The plots were managed using a range of practices, but for the purposes of summarizing slug activity I have pooled the data from all the plots. I sampled in eight alfalfa plots and twelve maize plots, with eight shelter traps per plot for a total of 160 shelter traps. Shelter traps and the protocol for checking them are described in detail in Chapter 3. Briefly, in 2010, I sampled from mid-April to mid-June, and again from mid-August to mid-November. In 2011, I sampled from mid-April to mid-December. When field activities precluded sampling (e.g., alfalfa harvest, manure application), I removed the traps and then replaced them as soon as possible after the field activity. Otherwise the traps stayed in the field continuously. In 2011, I collected an additional piece of information to shed light on slug life cycles. At the beginning of the season, I randomly selected half of the traps in each plot and collected slugs from under those traps nine times over the season. Slugs were transferred into containers with moist paper towels in the field, kept cool in transit, and then returned to the lab and weighed. The gray garden slug (Deroceras reticulatum) was the most prevalent species found under shelter traps (Figure 1). This species had peaks in activity corresponding to the times of crop establishment in spring and fall, indicating its importance as a crop pest. The other two slug species were generally found less frequently, although the marsh slug (D. laeve) was the most prevalent species in winter of 2011 (Figure 1). Data on slug mass over the seasons provided insights into slug life cycles in this region. The gray garden slug appeared to be an annual species, with predominantly small juveniles in spring that matured through the summer and reached maturity in fall (Figure 2). The marsh slug had quite a different pattern, with larger individuals found in spring (likely overwintered juveniles or adults) and most of the smaller juveniles present in fall. These complementary life history patterns could be a result of temporal niche partitioning in the two congeneric slugs. The banded slug (Arion fasciatus) was collected in smaller numbers and had a more complex pattern in biomass over the season, potentially indicating a multi-year life cycle.

120

Figures

Figure A-1. Activity-density (mean ± SE) of three pest slug species observed under shelter traps during a field study in central Pennsylvania. Traps were located in both alfalfa and maize plots (n = 4 to 20 plots per date).

121

2.0 Gray garden

1.5

)

g

(

s 1.0

s

a M

0.5

0.0

1 1 1 1 1 1 1 1 1 /1 /1 /1 /1 /1 /1 /1 /1 /1 9 0 7 8 9 9 7 5 1 /0 /3 /1 /0 /1 /0 /0 /2 /0 5 5 6 7 8 9 0 0 2 1 1 1

Figure A-2. Biomass of gray garden slugs (Deroceras reticulatum) collected from shelter traps in maize and alfalfa plots in 2011. The black symbol signifies the median biomass on each date.

0.4 Marsh

0.3

)

g

(

s

s 0.2

a M

0.1

0.0

1 1 1 1 1 1 1 1 1 1 /1 /1 /1 /1 /1 /1 /1 /1 /1 /1 9 0 7 8 9 9 9 7 5 1 /0 /3 /1 /0 /2 /1 /0 /0 /2 /0 5 5 6 7 7 8 9 10 10 12

Figure A-3. Biomass of marsh slugs (Deroceras laeve) collected from shelter traps in maize and alfalfa plots in 2011. The black symbol signifies the median biomass on each date.

122

1.6 Banded 1.4

1.2

1.0

)

g

(

s 0.8

s a

M 0.6

0.4

0.2

0.0

1 1 1 1 1 1 1 /1 /1 /1 /1 /1 /1 /1 9 0 7 9 7 5 1 /0 /3 /1 /0 /0 /2 /0 5 5 6 9 10 10 12

Figure A-4. Biomass of banded slugs (Arion fasciatus) collected from shelter traps in maize and alfalfa plots in 2011. The black symbol signifies the median biomass on each date.

Appendix B

Supplemental tables describing statistical methods in Chapter 3

Table B-1. Mixed model analysis for cross-rotation comparisons, including split-plot effects.

Source of variation df equation df Random Error terms Block r-1 3 X Rotation a-1 2 Block X Rotation (r-1)(a-1) 6 X Error A Split-plot(Rotation) (a)(b-1) 3 Block X Split-plot(Rotation) (r-1)(b-1)(a) 9 X Error B Total adjusted (r*a*b) - 1 23

Table B-2. Mixed model analysis for the Forage rotation, including split-plot and split-split-plot effects (2011 only):

Source of variation df equation df Random Error terms Block r-1 3 X Split-plot a-1 1 Split-split-plot b-1 1 Split-plot X Split-split-plot (a-1)(b-1) 1 Block X Split-plot (r-1)(a-1) 3 X Error A Block X Split-split-plot (r-1)(b-1) 3 X Error B Block X Split-plot X Split-split-plot (r-1)(a-1)(b-1) 3 X Error C Total adjusted (r*a*b) – 1 15

124 Table B-3. Repeated measures mixed model analysis for cross-rotation comparisons, including split-plot effects (e.g. with two time points, t = 2).

Source of variation df equation df Random Error terms Block r-1 3 X Rotation a-1 2 Block X Rotation (r-1)(a-1) 6 X Error A Split-plot(Rotation) (a)(b-1) 3 Block X Split-plot(Rotation) (a)(b-1)(r-1) 9 X Error B Time t-1 1 Time X Rotation (t-1)(a-1) 2 Time X Split-plot(Rotation) (t-1)(b-1)(a) 3 Time X Block (t-1)(r-1) 3 X Error C Time X Block X Rotation (t-1)(r-1)(a-1) 6 X Error C Time X Block X Split-plot(Rotation) (t-1)(r-1)(b-1)(a) 9 X Error C Total adjusted (r*a*b*t) – 1 47

Appendix C

Supplemental data on the influence of slug damage on grain yield in maize

In 2010, I followed maize plants over the season to better understand the influence of early-season slug damage on grain yield. This effort focused on maize plants in the broadcast manure treatment within the Control rotation from the field experiment that is described in detail in Chapter 3. In each split-plot (n = 4), I selected twelve maize plants in early July (7/2/10) in a stratified random pattern and marked them with flagging tape around their bases. All maize plants were within the center eight rows of each split-split plot to avoid edge effects, and were at growth stage V7. Seven independent observers rated each maize plant for slug damage. Because the maize seedlings were beyond the stage where slug damage is typically most severe, we rated only the bottom four leaves to capture damage relative to an earlier growth stage of the plant. Those leaves were rated on a damage scale as follows: 0 = no damage; 0.4 = < 10% defoliation, 1 = 10- 25% defoliation, 2 = 25-50% defoliation, 3 = 50-75% defoliation, and 4 = 75-100% defoliation. In October, when maize plants had dried down and were ready for grain harvest (10/23/10), I collected the marked maize plants and weighed their whole ears. I assume that ear weight is strongly correlated to grain yield. To examine whether early season slug damage was related to eventual grain yield of individual plants, I used regression analysis (Minitab v. 16, Minitab, Inc., State College, PA). Similar to results in dry years from a previous study (Byers and Calvin 1994), I found that the relationship between slug damage and ear weight was better described by a quadratic rather than a linear function. Despite the overall significance of the regression (F2,45 = 6.11, P = 0.004), the function including slug damage and its square explained relatively little variation in ear weight of individual plants (R2 = 0.21, Figure 1). Furthermore, the quadratic nature of the relationship indicated that slug damage was not related to ear weight in a simple way.

126

Reference

Byers, R. A. and D. D. Calvin. 1994. Economic injury levels to field corn from slug (Stylommatophora: Agrolimacidae) feeding. Journal of Economic Entomology 87(5): 1345-1350.

127

Figure

Figure C-1. The relationship between slug damage (% defoliation of the lowest four leaves at V7) and ear weight of individual maize plants at harvest (n = 48 plants, split among 4 plots).