INFORMATION TO USERS
This manuscript has been reproduced from the microfihn master. UMI films the t%t directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zed) Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600
EFFECTS OF VEGETATIONAL DIVERSITY ON THE POTATO LEAFHOPPER
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By
Timothy Joseph Miklasiewicz, M. Sc.
The Ohio State University 1997
Dissertation Committee:
Professor Ronald B. Hammond, Adviser Approved by
Professor David J. Hom Professor Benjamin L Stinner Professor Daniel L. Jeffers
Adviser Department of Entomology UMI Number: 9801747
UMI Microform 9801747 Copyright 1997, by UMI Company. Ail rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 Copyright by Timothy J. Miklasiewicz 1997 It is better to travel well than to arrive.
-Old Tibetan saying ABSTRACT
The effects of differences in vegetational diversity upon abundance and behavior of the potato leafhopper, Empoasca fabae (Harris) (Homoptera: Cicadellidae), were examined in four agroecosystems. Each cropping system employed the same planting density and dispersion of a potato leafhopper host plant, soybean {Glycine max (L) Merr.). One of the four cropping systems was a soybean monoculture; the other 3 cropping systems were relay intercropping systems that varied in the number of wheat rows planted between soybean rows. Wheat
{Triticum aestivum L. ) is known to be a non-host plant for the potato leafhopper. Together, the four cropping systems form a gradient of vegetational diversity, that is, of host/non-host concentration. Because of the nature of the relay intercropping systems, a distinction is made between the early part of the season, during which soybean and wheat are co-present in experimental plots, and the period following the wheat harvest, when all four cropping system s are soybean monocultures. Two experiments were conducted to test whether the leafhopper showed an inverse response to increasing amounts of the non-host., as predicted by the resource concentration hypothesis.
In the first experiment, estimates were made of potato leafhopper absolute density on soybean using a portable suction device. Prior to the wheat harvest, potato leafhoppers were substantially and significantly more abundant in the soybean monoculture than in any of the intercropping systems. Mean densities were extremely low in all the intercropping systems.
After the wheat harvest, the disparities between the soybean monoculture and intercropping systems gradually decreased, and fewer collection dates showed significant differences. However, potato leafhopper densities remained greater In the soybean monoculture than In the other systems until late In the season. Females and males showed similar population trends and differences among cropping systems. There were no significant differences In density detected among the intercropping systems.
The second study measured colonization of the four cropping systems using water pan traps located within the soybean canopy and wheat canopy. The presence of wheat In the intercropping systems strongly Inhibited potato leafhopper from making contact with the soybean canopy until after the wheat harvest The great majority of sampling Intervals tested had significantly more potato leafhoppers detected within the soybean canopy In the soybean monoculture than In the Intercropping systems. Within the Intercropping systems, relatively high numbers of leafhoppers trapped within the wheat canopies showed that migrating potato leafhoppers contacted the wheat, but few were trapped within the corresponding soybean canopies of the same cropping systems. There were essentially no differences In colonization among the Intercropping systems. Following the wheat harvest, the disparity In soybean canopy colonization diminished rapidly.
The results Indicate that colonization of soybean. Inhibited by the presence of wheat. Is an Important mechanism In producing the substantial differences In density observed among the cropping systems. Potato leafhopper exhibited a qualitative type of response to the presence of the non-host, rather than quantitative. This was Indicated by the lack of differences In density or colonization among the three Intercropping systems, despite large differences In the amount of wheat present, and differences In host quality.
Ill DEDICATION
This dissertation is dedicated to my family, especially to my wife, parents, brother and sister’s family, and to my friends, in appreciation of their love and support.
IV ACKNOWLEDGEMENTS
I owe a serious and sincere debt to my adviser, Ron Hammond, for his untiring and selfless efforts, ideas, support, and encouragement in seeing this dissertation through to this phase. I don't believe there's really any way I can thank him enough. Secondarily, I'd like to thank Dave Hom for his help and inspiration through the years; for bringing me to OSU, teaching and entertaining me in class, finding niches to fit me into, serving as Co-Adviser to this project (I should have consulted you much more), and for smoothing the way through many difficult interfaces. Thanks to Dan Jeffers for teaching me so much about crop ecology, agronomic procedures, and equipment; and for his great sense of humor and perspective. I'd like to thank
Ben Stinner for serving on my Guidance Committee, for his advice and loaning me equipment and space to work. I express my gratitude to all of the above for their help with this document and for their forbearance of multiple last-minute submissions. I'd also like to thank the families of all of the above, for many favors, calls at unusual hours, and unorthodox requests.
Thanks to members of the soybean lab for so much hard work, especially Judy Smith and Elaine Ressler for their great organizational skills, expertise, and labor. Kaitlin Lucas, Gwen
Ruprecht, Jeremy Wienwille, Laura Richmond, Wendy Riggenbach, and others provided help with collecting and processing samples.
Others among the technical staff at OARDC that were especially helpful were Kevin
Power, Mark Belcher, Mike Dunlap, Dave McCartney, Dan Fickle, Jim Mason, Bill Styer, Roger
Downer, Terry Moore, and Millie Casey.
Others at miscellaneous satellites of the Entomology Department who need to be thanked are the entire Extension Entomology staff, especially Celeste Welty, Dave Shetlar, Hal Wilson, Julie Steele, Jeannette Jansen, Jean Steva, and Lynn Berry. John Flessel helped with references, advice, good stories, assistance with meetings, pig roasts, and loan of equipment
Thanks to Rich Hall for his good advice and the opportunity to TA with him. Roger Williams provided equipment, literature, and many interesting and enjoyable discussions. Bob Treece facilitated my presence at Wooster by providing funds for an associateship. Dick Lindquist and
Skip Nault helped with continued support, good humor, and sound answers to many many forwarded questions from among the public-at-large. Thanks to Roy Rings and George
Shambaugh for many good discussions, help with reference material and for providing some historical perspective.
Mabel Kirchner and Max Johnson helped me in innumerable ways with administrative requirements and with locating things and people through the years. When Mabel told me recently that there were things about the Department that she didn't know, I knew it was time to move on! Betty, Linda, and Rena in the Columbus Departmental Office always found a way to accomplish what I needed to have done.
Chuck Triplehorn, Norm Johnson, and Andrey Sharkov provided access to the Museum of Biodiversity Collections and help with leafhopper systematics. Derrick Blocker was of great help in confirming and correcting identification of leafhoppers; Andy Hamilton provided taxonomic advice and references. John Huber and Serguei Triapitsyn identified mymarids and discussed parasite ecology and host relationships.
Brian Sugarman and Paul McMillen in the Crop & Soil Science Department, Charlie
Mabry of Entomology, and Bill Bardoll of Plant Pathology were invaluable in assisting with planting and harvesting experimental plots. The expertise of the ferm operations staff was
VI responsible for resurrecting dead D-Vacs many times, allowing sampling to continue for another day.
Bert Bishop gave invaluable, but ill-appreciated, advice regarding appropriate statistical analysis of experiments. Jody Lanham, Jim Holman, and Carolyn Britt provided training and advice on the use of computer hardware and software, and miscellaneous services such as slide-making. The fecilities that they helped to develop at OARDC have advanced so far that its difficult to remember how limited things were when I arrived. Connie Britton at the OARDC
Library helped with locating and acquiring local and remote references, and made it alot easier to negotiate both physical collections and various online catalogs.
Many thanks to Dave Heilman, Sherry Ferrell, and Jim Tew at the OARDC Honey Bee
Lab for their efforts in trying to make a beekeeper out of me. Sometime in the not-too-distant future 1 hope to have some of that elusive "free time" to maybe try out some of the many ideas you gave me.
At the University of Maryland, Bill Lamp w as the phantom member of my Guidance
Committee in providing invaluable expertise, advice, and resources to my project. Thanks also to Dave Liewehr for sharing information and insight into mymarid ecology and methology.
My fellow grad students and other friends deserve thanks for making life fun and interesting amid alot of hard work and stress. Special thanks to Joâo Lopez, Athayde Tonhasca,
Jackie Blackmer, Astri Wayadande, Ty Vaughn, Hannes Busch, Tom Wolf, Charlotte Bedet,
Doug Richmond, LeAnn Beanland, Ed Zaborski, George Keeney, Gustavo Moya-Raygoza,
Isabel Sohn, Debbie Murral, Gloria Mukulu, Alex Stone, and Sandra & Carlos Alcaraz. Mee-Rye
Cha and Chela Vazquez have probably been my closest friends through most of my time at
OSU, and I thank them for their friendship, understanding, and many (but not enough) happy
VII times. I also want to collectively thank my more distant friends from Delaware and California
(many of us have scattered across the landscape) for enduring long periods of my neglect
Finally, I'd like to thank my family: dear wife Cathy, for her love, patience, and support through this long period of work; Jason, Rachel and the Mosleys for welcoming me into their tribe; my parents, especially Mom, who probably didn't know exactly what I was doing all this time, or why, but maintained constant love and encouragement; my brother Greg, for keeping things interesting (a desperately needed and under-appreciated talent); and to my sister Debbie,
Tom, Tyler, Dane, and Taylor, for always being there for everyone whenever they're needed.
Although I may not say it enough, I love you all very much.
Thanks to everyone!
VIII VITA
25 January. 1953 ...... Bom - Wilmington, Delaware
1976 ...... 8. A., Psychology and Sociology University of Delaware Newark, Delaware
1978-1979 ...... Agricultural Research Technician Beneficial Insects Research Lab Agricultural Research Service, USDA Newark, Delaware
1979 ...... B. S., Entomology-Plant Pathology University of Delaware Newark, Delaware
1980-1982 ...... Technical Assistant U.S. Peace Corps/Caribbean Agricultural Research and Development Institute Barbados, West Indies
1982-1984 ...... Biological Technician Boyden Laboratory Agricultural Research Service, USDA Riverside, California
1988 ...... M. S., Entomology University of California Riverside Riverside, California Specialization: Insect Pest Management Thesis Title: Investigations of the influence of high temperature on biological control of the woolly whitefly.
1988-198 9 ...... Staff Research Associate University of California Riverside Riverside, California
1989-199 0...... Graduate Teaching Associate Department of Entomology Ohio State University Columbus, Ohio
ix 1990-199 1 Graduate Research Associate (Extension) Department of Entomology Ohio State University Columbus, Ohio
1991-199 4...... Graduate Research Associate Department of Entomology Ohio State University/OARDO Wooster, Ohio
Publications
Miklasiewicz, T. J., and G. P. Walker. 1990. Population dynamics and biological control of the woolly whitefly (Homoptera: Aleyrodidae) on citrus. Environ. Entomol. 19(5): 1485-1490.
Miklasiewicz, T. J., and 0 . Welty. 1991. Root Maggots. Ohio Cooperative Extension Service, Vegetable Pest Management Circular.
Field of Study
Major Field: Entomology Specialization: Insect Ecology TABLE OF CONTENTS
Section... Page
Abstract...... ii
Dedication ...... iv
Acknowledgements ...... v
Vita...... ix
List of Tables...... xiii
List of Figures...... xvii
Chapters:
1. Vegetational diversity vs. the potato leafhopper...... 1
Hypotheses about vegetational diversity ...... 1
The potato leafhopper...... 3
Soybean-wheat relay intercropping studies ...... 7
2. The influence of varying vegetational diversity upon abundance of the potato leafhopper in four cropping systems...... 10
Introduction ...... 10
Materials and methods ...... 14
Results...... 20
Discussion ...... 30 xi 3. Colonization of four cropping systems by the potato leafhopper, as studied using water pan traps...... 34
Introduction...... 34
Materials and methods ...... 39
Results...... 45
Discussion ...... 70
4. Summary and discussion...... 78
Review of experimental results ...... 78
Analysis of research objectives and hypotheses tested ...... 80
Implications of results ...... 81
Future directions ...... 85
Appendix A. Sex ratios of potato leafhoppers from two field studies...... 88
Appendix B. Leafhoppers and mymarids identified from water pan trap samples 91
Appendix 0. Measurements of light intensity in cropping systems...... 94
Appendix D. Measurements of host plant quantity and quality in cropping
systems...... 102
References...... 113
XII LIST OF TABLES la b ls Eage
Table 2.1 Results of analysis of variance tests and means comparisons for female PLH data. The unit of data used as the dependent variable was a Iog10 (X +1) transformation of the mean number of females collected per D-Vac bag, averaged over the experimental plot (3 bags per plot, 3 plots per cropping system in 1991, 4 plots per cropping system in 1992). The single main effect tested was cropping system (4 levels). A separate analysis was made for each collection date. The P value indicates the probability that an equal or greater F would occur by chance. Each mean shown (based on untransformed data) is the overall average for all plots of the cropping system for the date indicated. Within the sam e date, means followed by the sam e letter were not found significantly different using Tukey’s w procedure. Upper-case letters indicate means comparison results when a=0.01, lower-case letters indicate results when a=0.05 ...... 24
Table 2.2 Results of analysis of variance tests and means comparisons for male PLH data. The unit of data used as the dependent variable was a loglO (X +1) transformation of the mean number of males collected per D-Vac bag, averaged over the experimental plot (3 bags per plot, 3 plots per cropping system in 1991, 4 plots per cropping system in 1992). The single main effect tested was cropping system (4 levels). A separate analysis was made for each collection date. The P value indicates the probability that an equal or greater F would occur by chance. Each mean shown (based on untransformed data) is the overall average for all plots of the cropping system for the date indicated. Within the sam e date, means followed by the same letter were not found significantly different using Tukey’s w procedure. Upper-case letters indicate means comparison results when a=0.01, lower-case letters indicate results when a=0.05 ...... 25
Table 3.1 Measurements of the relative frequency of PLH recovery within soybean- centered and wheat-centered traps. The number of sampling intervals which had PLH present in traps can be compared among cropping systems and to the total possible number of intervals for which recovery might have occurred. Female and male data are presented separately. This table is calculated on a presence/absence basis for each cropping system; i.e. recovery of any PLH in a single trap (3 traps/plot, 3 plots for each cropping system) gives a positive score for the sampling interval...... 53
XIII lafele
Table 3.2 The cumulative number of PLH recovered over the series of sampling intervals before and after the wheat harvest Separate sums are listed for sex, year, trap placement, and cropping system. NA- not applicable as cropping system was not sampled in the time period or with the trap placement referenced ...... 54
Table 3.3 Summary of results of one-way analysis of variance tests and means comparison tests for female data from soybean-centered traps preceding wheat harvest MODEL: fixed-effect model with number of female PLH recovered per plot per day (transformed, see text) as the dependent variable, with cropping system as the single independent variable (4 levels). Sampling intervals were analyzed separately. Only sampling intervals which had PLH in more than one cropping system and produced non-significant values for Levene's Test were submitted to analyses of variance and means comparisons. The P value indicates the probability that an equal or greater value of F (from one-way test) would occur by chance. For means comparison, means followed by the same letter were not significantly different from one another at least at the a=0.05 level. Upper-case letters indicate that the differences between the marked mean and all other means were significant at the a=0.01 level. NS = no significant differences. Cropping systems listed are ranked in order of descending value of means (left to right); parentheses indicate groups of equal m eans...... 55
Table 3.4 Summary of results of two-way analyses of variance tests for female PLH data within specified sampling intervals. MODEL: fixed-effect model with number of female PLH recovered per day as the dependent variable (transformed, see text), with cropping system and trap placement as main effects, and with the interaction of cropping system and trap placement as the interaction effect. The P value indicates the probability that an equal or greater value of F (from two-way test) would occur by chance. NS = no significant differences were found for the effect indicated (a=0.05). The column at the far right lists the combinations of cropping system and trap placement that had any female PLH present during the sampling interval specified. Designations for the cropping systems (see text) are followed by s for soybean-centered traps or w for wheat-centered traps. The numerical relationships among means for the cropping systems listed are provided by greater than (>) or equal to (=) signs between pairs of means. Rankings do not imply significant differences among means ...... 56
XIV lablÊ Page
Table 3.5 Summary of results of one-way analysis of variance tests and means comparison tests for male data from soybean-centered traps preceding the wheat harvest MODEL: fixed-effect model with number of male PLH recovered per plot per day (transformed, see text) as the dependent variable, with cropping system as the single independent variable (4 levels). Sampling intervals were analyzed separately. Only sampling intervals which had PLH in more than one cropping system and produced non-significant values for Levene’s Test were submitted to analyses of variance and means comparisons. The P value indicates the probability that an equal or greater value of F (from one-way test) would occur by chance. For m eans comparison, means followed by the sam e letter were not significantly different from one another at least at the a=0.05 level. Upper-case letters indicate that the differences between the upper-case marked mean and all other means were significant at the a=0.01 level. NS = no significant differences. Cropping systems listed are ranked in order of descending value of means (left to right): parentheses indicate groups of equal means...... 58
Table 3.6 Summary of results of analysis of variance tests and means comparison tests for male data from wheat-centered traps. MODEL: fixed-effect model with number of male PLH recovered per plot per day (transformed, see text) as the dependent variable, with cropping system as the single independent variable (4 levels). Sampling intervals were analyzed separately. Only sampling intervals which had PLH in more than one cropping system and produced non-significant values for Levene’s Test were submitted to analyses of variance and means comparisons. For means comparison, means followed by the same letter w ere not significantly different from one another at least at the a=0.05 level. Upper-case letters indicate that the differences between the upper-case marked mean and all other means were significant at the a=0.01 level. NS = no significant differences. Cropping systems listed are ranked in order of descending value of means (left to right); parentheses indicate groups of equal means...... 59
Table 3.7 Summary of results of two-way analyses of variance tests for male PLH data within specified sampling intervals. MODEL: fixed-effect model with number of male PLH recovered per plot per day (transformed, see text) as the dependent variable, with cropping system and trap placement as main effects, and with the interaction of cropping system and trap placement as the interaction effect. The P value indicates the probability that an equal or greater value of F (from two-way test) would occur by chance. NS = no significant differences were found for the effect indicated (a=0.05). The column at the far right lists the combinations of cropping system and trap placement that had any female PLH present during the sampling interval specified. Designations for the cropping systems (see text) are followed by s for soybean-centered traps or w for wheat-centered traps. The numerical relationships among means for the cropping system s listed are provided by greater than (>) or equal to (=) signs between pairs of means. Rankings do not imply significant differences among means ...... 60 XV lablê Page
Table A.1 Sex ratio of potato leafhoppers captured during the abundance study. The grand mean (all cropping systems) for each sample collection date is shown for males, females, and total potato leafhoppers (males + females). The sex ratio is expressed as the proportion of total potato leafhoppers that were female. A small number of potato leafhoppers were not sexed due to specimen damage, these were excluded from the calculations ...... 88
Table D.1 Soybean canopy height in four cropping systems, measured in the field, 1991 and 1993. Summary statistics for each cropping system by replication and overall for each year. Units of height measurement are cm...... 104
Table D.2 Height of soybean in four cropping systems, distance between cotyledonary node and terminal node, 1991 and 1993. Summary statistics for each cropping system by replication and overall for each year. Units of height measurement are cm ...... 106
Table D.3 Leaf area for soybean in four cropping systems, 1991 and 1993. Statistics shown here are for aggregate leaf area of plants obtained from a sample corresponding to 0.25 m^ ground area. Units of area measurement are cm^ ...... 108
Table D.4 Biomass for soybean in four cropping systems, 1993. Statistics shown here are for aggregate dry epigial biomass of plants obtained from a sample corresponding to 0.25 m^ ground area. Units of biomass measurement are grams...... 110
Table D.5 Vegetative stage for soybean in four cropping systems, 1991 and 1993. Statistics shown here are based upon stage ratings for individual plants. 111
XVI LIST OF FIGURES
Figure Page
Fig. 2.1 Graphic representing the relationship of soybean to wheat rows in the 4 cropping systems studied. Only a small portion of each cropping system is indicated, that is, the width of a single planter pass through the field, approximately 2.4 m (8 feet). Wheat rows are represented by dashed lines; soybean rows are represented by solid lines. Within each planter pass, 3 soybean rows and a variable number of wheat rows (depending on treatment) are contained, a) 14 wheat rows: 3 soybean rows, b) 7 wheat rows: 3 soybean rows, c) 3 wheat rows: 3 soybean rows, d) 0 wheat rows: 3 soybean rows (soybean monoculture) ...... 15
Fig. 2.2 Mean adult female PLH captured per D-Vac sample by cropping system, 1991. Each mean represents 3 samples per plot, 3 plots per cropping system (see text). Arrows indicate the beginning and end of the wheat harvest...... 26
Fig. 2.3 Mean adult female PLH captured per D-Vac sample by cropping system, 1992. Each mean represents 3 samples per plot, 4 plots per cropping system (see text). Arrows indicate the beginning and end of the wheat harvest...... 27
Fig. 2.4 Mean adult male PLH captured per D-Vac sample by cropping system, 1991. Each mean represents 3 samples per plot, 3 plots per cropping system (see text). Arrows indicate the beginning and end of the wheat harvest...... 28
Fig. 2.5 Mean adult male PLH captured per D-Vac sample by cropping system, 1992. Each mean represents 3 samples per plot, 4 plots per cropping system (see text). Arrows indicate the beginning and end of the wheat harvest...... 29
Fig. 3.1 Comparison of female PLH recovered from soybean-centered traps in 4 cropping systems during 1991. The height of the ribbon at each articulation point indicates the mean of females recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot). The break in the ribbons and date axis labelling shows the period when sampling was interrupted by the wheat harvest ...... 62
XVII Figure Page
Fig. 3.2 Comparison of female PLH recovered from wheat-centered traps in 3 cropping systems during 1991. The height of the ribbon at each articulation point indicates the mean of females recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot)...... 63
Fig. 3.3 Comparison of female PLH recovered from soybean-centered traps in 4 cropping systems during 1992. The height of the ribbon at each articulation point indicates the mean of females recovered per plot per day of the sampling interval (3 plots per cropping system. 3 traps per plot)...... 64
Fig. 3.4 Comparison of female PLH recovered from wheat-centered traps in 3 cropping systems during 1992. The height of the ribbon at each articulation point indicates the mean of females recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot)...... 65
Fig. 3.5 Comparison of male PLH recovered from soybean-centered traps in 4 cropping systems during 1991. The height of the ribbon at each articulation point indicates the mean of males recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot). The break in the ribbons and date axis labelling shows the period when sampling was interrupted by the wheat harvest ...... 66
Fig. 3.6 Comparison of male PLH recovered from wheat-centered traps in 3 cropping systems during 1991. The height of the ribbon at each articulation point indicates the mean of males recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot)...... 67
Fig. 3.7 Comparison of male PLH recovered from soybean-centered traps in 4 cropping systems during 1992. The height of the ribbon at each articulation point indicates the mean of males recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot)...... 68
Fig. 3.8 Comparison of male PLH recovered from wheat-centered traps in 3 cropping systems during 1992. The height of the ribbon at each articulation point indicates the mean of males recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot) ...... 69
Fig. A.1 Sex ratio of potato leafhoppers recovered from water pan traps, 1991. Sex ratio was expressed a s the proportion of total potato leafhoppers recovered that were female. A small number of potato leafhoppers were not sexed due to damage, and were not included. Labels near datapoints show the number of specimens upon which the sex ratio was based ...... 89
XVIII Figure Page
Fig. A.2 Sex ratio of potato leafhoppers recovered from water pan traps, 1992. Sex ratio was expressed as the proportion of total potato leaôioppers recovered that were female. A small number of potato leafhoppers were not sexed due to damage, and were not included. Labels near datapoints show the number of specimens upon which the sex ratio was based. Gaps in the plot indicate sampling intervals during which no potato leafhoppers were recovered ...... 90
Fig. C.1 Variation in light intensity throughout the day, measured above the top canopy in the cropping system s studied. Units of light intensity are pmoles*s'’*m'\ Time of the day is expressed in hours of the 24 hour clock...... 96
Fig. 0.2 Comparison of light intensity at the top of the soybean canopy and at ground level for the soybean monoculture (0:3). Heights of bars show m eans (N=5) with standard error caps. Units of light intensity are pmoles*s’’*m'’ ...... 97
Fig. 0.3 Comparison of light intensity at the top of the wheat canopy, at the top of the soybean canopy, and at ground level for the 3:3 intercropping system. Heights of bars show m eans (N=5) with standard error caps. Units of light intensity are pmoles*s'’*m'’ ...... 98
Fig. 0.4 Comparison of light intensity at the top of the wheat canopy, at the top of the soybean canopy, and at ground level for the 7:3 intercropping system. Heights of bars show means (N=5) with standard error caps. Units of light intensity are pmoles*s"’*m'’ ...... 99
Fig. 0.5 Comparison of light intensity at the top of the wheat canopy, at the top of the soybean canopy, and at ground level for the 14:3 intercropping system. Heights of bars show means (N=5) with standard error caps. Units of light intensity are pmoles*s'’*m'’ ...... 100
Fig. 0.6 Light intensity at the top of the soybean canopy expressed as a proportion of the amount above the wheat canopy. Heights of the bars provide means (N=5), with standard error caps. Five periods of the day are shown for 3 intercropping system s ...... 101
XIX CHAPTER 1
VEGETATIONAL DIVERSITY VS. THE POTATO LEAFHOPPER
Hypotheses about Vegetational Diversity
Among the most durable and popular themes In community ecology is the notion that abundance of individual herbivore species and the damage they cause to plants is suppressed as the vegetational complexity of the plant environment increases. This idea has frequently been offered as an explanation for the apparent discrepency between the often devastating losses to crops in artificial systems and less severe or less frequent injuries observed for wild plants in natural system s (van Emden & Williams 1974). Root (1973) formalized two hypotheses, the resource concentration hypothesis and the enemies hypothesis, which provided mechanisms that might be able to account for these differences. The resource concentration hypothesis states that diet-specialized herbivores are more likely to find and remain in habitats where their host plants are concentrated. Reduction in the density of host plants, unfavorable alteration of their spatial arrangement, or increase in the frequency of non-host plants is believed to inhibit host location or exploitation. Herbivores may demonstrate a greater propensity to leave a habitat with less concentrated host resources, or exhibit reduced rates of population increase. The enemies hypothesis holds that the abundance of herbivores and the frequency of outbreaks among herbivorous pests is suppressed because natural enemies are more abundant or effective in producing herbivore mortality in vegetationally complex habitats. Diverse habitats are deemed to provide better conditions to predators and parasites; these allow their populations to be sustained or increased, inhibit emigration, and reduce the probability of local extinction. These conditions include greater availability of resources such as prey or hosts, pollen, and nectar, more shelter, and improved microclimate.
In addition to these two major hypotheses, several minor ones have been introduced that have not received much attention. In systems with more plant diversity, greater levels of interspecific competition among herbivores might be important in limiting their densities (Kareiva
1982); increased plant competition might slow plant development, altering the temporal pattern of susceptibility of plants to damage; or differences in host plant quality might change the rate of development or reproduction of the herbivore (Andow 1983a, 1983b).
While the idea that vegetational diversity can suppress herbivore density or pest outbreaks has been uncritically accepted by many entomologists, several reviewers have attempted to quantify whether such a conclusion is warranted. Andow (1983b, 1988), Kareiva
(1983), Risch et al. (1983), and Stanton (1983) indicated that for studies which they considered as acceptable tests of the resource concentration hypothesis, about half of herbivore species tested were reduced in abundance as vegetational diversity increased. For example, Andow
(1983b) determined that 53% of 198 herbivore species were less abundant on a population per plant basis in polycultures or weedy habitats than in monocultures; 18% were more abundant,
9% showed no change, and 20% showed a varied response. In accordance with predictions of the resource concentration hypothesis, polyphagous herbivores increased more frequently in polycultures than monophagous species (44% vs. 10%), and herbivores increased more often in polycultures with perenniai crops than those with annuals (29% vs. 13%) (Risch etal. 1983).
Few studies were able to provide much insight into mechanisms responsible for differences because they inadequately accounted for herbivore behavioral responses to host density and arrangement, actual host part utilization, variation in host quality resulting from plant competition, and effects of abiotic factors. The intention of this research is to investigate the effects of vegetational diversity on a particular herbivore, the potato leafhopper. The characteristics of this species, described below, were well suited to provide a productive and interesting exploration of the mechanisms underlying abundance differences within a range of diverse habitats. Of the several hypotheses noted above, the resource concentration hypothesis appeared to be the most appropriate for application to the circumstances here, because of its prominence and the relatively greater emphasis placed on herbivore movement and habitat colonization. An experimental combination of host (soybean) monoculture and host/non-host (soybean/wheat) bicultures, introduced later in this chapter, provided the vegetational context for this investigation.
The Potato Leafhopper
The potato leafhopper, Empoasca fabae (Harris), is a pest of a wide variety of crops over much of the eastern and central United States, including farms and home gardens. Plants that are especially affected include legumes, such as alfelfa, red and white clover, dry bean, snap bean, peanut, cowpea, and soybean; vegetables and fruit, such as potato, eggplant, beet, strawberry, apple, brambles, and grape; and ornamentals, such as Nonway maple, wisteria, dahlia, and rose (Washburn 1908; Fenton & Hartzell 1923; Ackerman 1931; Poos & Haenseler
1931; Poos & Johnson 1936; Poos & Wheeler 1943; DeLong 1938,1965; Batten & Poos 1938;
Ohio State University Extension Bulletin #861 1997). The type and extent of injury differs somewhat according to plant. On alfalfa, where extensive research has been conducted, feeding can cause distorted veins and leaf margins, color change, reduced photosynthesis, increased respiration and transpiration, increased non-structural carbohydrate in leaves, stunting, necrosis, and chemical changes (Smith & Poos 1931, Poos & Johnson 1936, DeLong
1971, Ogunlana & Pedigo 1974, Womack 1984, Hower 1989, Nielsen etal. 1990). On an economic level, yield loss, stand reduction, and reduction in forage value through loss of crude protein may result (Poos & Johnson 1936, Helmet a/. 1980, Flinn etal. 1990a). While this leafhopper infrequently causes economic damage to commercial varieties of soybean, it is among the m ost abundant herbivores inhabiting the crop in the Midwest (Helm et el. 1980).
Until now, there has been a consensus belief that the potato leafhopper (PLH) does not have the ability to withstand long periods of cold (DeLong & Caldwell 1935; Decker & Maddox
1967; Decker & Cunningham 1967,1968; Speaker et a/. 1990; Shields & Sher 1992) and that usually its winter range in the United States is confined to areas near the Gulf of Mexico that have no more than brief periods of sub-freezing weather. Recent recoveries of PLH overwintering adults in pine habitats in areas as far north as coastal Maryland and southern
Tennesee may require a revision of the widely accepted overwintering scenario (NC-193
Technical Committee 1993). Each year, as temperatures increase in the spring and the geographic distribution of acceptable hosts increases, PLH expands its range to include most of the lower-elevation areas of the eastern U. S. (DeLong 1938). This expansion in range is assisted by flights that take advantage of rapidly moving, northward-directed air m asses typically associated with storm fronts (Medlar 1957, Pienkowski & Medlar 1964, Decker & Cunningham
1967, Taylor & Reling 1986, NC-193 Technical Committee 1990). The coincidence of the development of sufficiently warm temperatures and availability of suitable host material for feeding in the northern target areas, persistence of weather phenomena that lead to the formation of north-bound storms, and decline in weather and host plant conditions favoring large population increases in southern source areas, create a temporal window lasting (typically) from mid-May through early July (Huff 1963, Kieckheffer & Medler 1964, Pienkowski & Medler 1964,
Kouskolekas & Decker 1966, Deckeref a/. 1971, Hogg 1985, Taylor 1989, Carlson etal. 1991).
During this window most long-distance migration of PLH occurs.
Following long-distance flight, migrant PLH are believed to undergo a three-step process leading to colonization of new habitats (Flinn ef a/. 1990b). The first step involves a reacquisition of attraction to stimuli (such as cues from host plants) that was inhibited in order to make migration possible (Kennedy 1985). Flying migrants come to ground, and begin searching for hosts in a series of short flights. Landing on hosts induces cessation of flight and feeding begins (Flinn etal. 1990b). Landing on non-hosts, such as grasses, promotes continued searching behavior (short flights). PLH may directly colonize crops, or feed and breed on wild hosts, such as maple, locust, oak, and honeysuckle (Gyrisco 1958). Adults appear to remain very mobile throughout their lives, so that theoretically redistribution among local hosts may occur at any time.
PLH has a relatively short immature developmental period. DeLong (1938) determined the length of the egg stage to be 9.8-10.4 d, the nymphal stage to last 12.3-15.4 d, and the preoviposition period for adults to be 5.6-6.6 d (ranges of means over 3 years). In Ohio, four generations can occur (DeLong 1938). Flinn etal. (1986) estimated egg to adult survivorship as
16-60% (mean=40%), but Hogg & Hoffman (1989) considered that estimate to be excessively low. Adult longevity under field conditions averages 30-35 d, and females can lay 2-3 eggs per day (DeLong 1938, Decker ef a/. 1971, Flinn etal. 1986). Hogg & Hoffman (1989) estimated that for temperature conditions prevalent in southem Wisconsin, PLH populations can increase
100-fold during the growing season. This high reproductive potential, combined with the mobility and unpredictability of pest occurrence in fields, means that crop managers must frequently monitor their fields of damage-susceptible crops and be prepared to take prompt remedial action.
A variety of studies have demonstrated that PLH density and damage can be reduced through manipulating vegetational diversity. Gentsch (1982), Lamp etal. (1984a), Kingsley etal.
1986, and Oloumi-Sadeghi etal. (1987,1989) found that density of this pest was reduced when grassy weeds were present in alfelfe plots. Andow (1992) determined that PLH density per bean {Phaseolus vulgaris L.) leaflet was higher in a bean monoculture than when weeds were present, although the density in relation to ground area did not differ. In cropping systems with soybean, Hammond & Stinner (1987) found fewer PLH in no-tillage systems than under conventional tillage, related to the presence of grassy weeds; Smith et al. (1988) reported lower populations of this leafhopper in no-tillage systems with a killed rye cover crop, compared with cropping systems without rye remaining above ground.. Lamp (1991) showed that oat-alfelfa mixtures had 80% less adults than alfelfe monocultures. Smith (1987) and Smith etal. (1992) assayed PLH adult behavior in the laboratory and concluded that volatiles emanating from crabgrass (Digitana sanguinalis L.) acted as locomotory stimulants and inhibited oviposition and feeding upon alfalfa. Coggins (1991) showed in laboratory studies that adult PLH emigrated from alfalfa-grass mixtures significantly more than from alfalfa monocultures. Lamp & Zhao
(1993) argued that as a polyphagous herbivore, PLH did not possess specialized sensory capabilities for host finding (characteristic of true diet-specialized herbivores), and could therefore be inhibited from finding hosts when they were present in a mixture with non-hosts.
In a study that employed a gradient of vegetational diversity consisting of five treatments with fixed bean (Phaseolus vulgaris, a PLH host) and variable tomato (a PLH non-host) density,
Roltsch & Gage (1990b) reported a clear inverse relationship between density of the non-host and densities of PLH nymphs and eggs (both indicators of oviposition). Bean plants were found to have reduced quality (total nitrogen, yield, and leaf area) as the tomato density increased.
Adult females increased oviposition in response to improved host quality (total plant nitrogen), however the differences in host plant quality were unable to account for the differences in PLH density observed. In a related study (Roltsch & Gage 1990a) conducted in the laboratory and greenhouse, the presence of non-host vegetation (tomato or cabbage) suppressed oviposition on bean. The authors concluded that reduced oviposition resulted from decreased residency time on hosts when non-hosts were also present. The same relationship between abundance and vegetational diversity has been shown with several other Empoasca species. In California, Letoumeau (1990) found that Empoasca complex (£ abrupta DeLong and E. recurvata DeLong) adults were more abundant in squash
monocultures than in polycultures of squash, cowpea, and com. Altieri etal. (1978) and
Schoonhoven etal. (1981) demonstrated that the neotropical leafhopper E kraemen Ross &
Moore was suppressed in dry bean plantings by the presence of grassy weeds or maize in polyculture.
Soybean-Wheat Relay Intercropping Studies.
The studies presented in this dissertation examine the ecology of PLH in soybean-wheat relay intercropping systems and a soybean monoculture. Relay intercropping soybean into winter wheat ("full relay intercropping") is an economically and ecologically advantageous cropping system in the northern Midwest, which has been recognized as a viable altemative to conventional sequential double cropping of winter grain and soybean (Jeffers etal. 1977; Jeffers and Triplett 1979; Chan etal. 1980; Jeffers 1987,1990). Soybean grown in monoculture
(various tillage practices) is the standard commercial production system for the crop, and it is used over an extensive area in Ohio (Acker 1987).
Hammond & Jeffers (1990) employed relay intercropping in an experiment that is in many respects a parent study to the current work. Their experimental design used two isolines of soybean, one highly susceptible and one resistant to PLH injury. Each isoline was intercropped with wheat and grown in monoculture. D-Vac suction sampling during June through August showed that for both isolines in the presence of wheat, mean adult and mean nymphal densities were significantly and substantially less than in corresponding monocultures.
Following the wheat harvest during July, leafhopper numbers increased on the susceptible line, especially in 1989. PLH density was usually more than 100-fold greater in monoculture than in polyculture prior to the wheat harvest. Both isolines yielded similarly in polyculture, however in
7 1989 leafhopper damage in monoculture reduced the mean yield of the susceptible line to 21% of that of the resistant line.
The experiments discussed in this dissertation utilized a single pubescent variety of soybean, which provided a good quality host for maintenance and development of PLH populations, but resisted damage that might contributed to variability in host quality among the cropping systems studied. This soybean was grown in monoculture and in three soybean-wheat relay intercropping systems that differed in the number of rows of wheat present. The four cropping systems formed a gradient of vegetational diversity. The purpose of the current research was to examine the effects of vegetational diversity on PLH, with emphasis on identification of mechanisms responsible for the pattern of population suppression that has been commonly reported. This purpose was addressed through two field studies.
The objective of the first study (Chapter 2) was to determine whether there was a differential response in abundance of PLH to the gradient of vegetational diversity represented by the cropping systems. The null hypothesis stated that there was no difference among cropping systems in PLH abundance. An altemative hypothesis, based upon the resource concentration hypothesis, stated that PLH would be distributed among cropping systems in numbers inversely proportionate to the amount of wheat present.
The objective of the second study (Chapter 3) was to determine whether PLH colonized host plants (soybean) in the habitats of the different cropping system s differentially, in response to varying levels of vegetational diversity. The null hypothesis stated that there was no difference in habitat colonization. An altemative hypothesis, in accordance with the resource concentration hypothesis, was that there would be reductions in habitat colonization as the amount of wheat (non-host) increased.
In association with the experiments reported above, additional measurements were made of in order to better characterize the habitats represented by the cropping systems used in
8 the experiments. Factors were selected for measurement because they were believed to have potential to influence PLH abundance or behavior, based on reports in the literature or information obtained during the course of current studies. The specific factors examined were shading, vegetative characteristics of soybean plants, the presence of parasitoids (Mymaridae, some of which are known parasitoids of PLH; Huber 1986) and other species of leafhoppers that might serve as altemative hosts or prey for PLH natural enemies. It was beyond the scope of the present research to fully explore the relationships of these factors to PLH abundance or behavior. Representative profiles of light intensity for different locations within the habitat of the cropping systems are provided in Appendix C. Quantification of soybean vegetative characteristics in the cropping systems is provided in Appendix D. Species lists of mymarids and leafhoppers recovered from sampling are contained in Appendix B. This information may be of use in stimulating additional investigation using similar or other cropping systems.
Both experiments were conducted concurrently using the same experimental plots. The information on PLH abundance (Chapter 2) and behavior (Chapter 3) is therefore expected to be complementary, and of greater value than similar experiments run separately. CHAPTER 2
THE INFLUENCE OF VARYING VEGETATIONAL DIVERSITY UPON ABUNDANCE OF THE POTATO LEAFHOPPER IN FOUR CROPPING SYSTEMS
INTRODUCTION
Relay intercropping describes a subclass of mixed cropping systems being examined as an innovative alternative for efficient use of land and favorable growing conditions. In the northern Midwest, relay intercropping of soybean and winter wheat has been shown to have economic and agronomic advantages over monocropping (i.e. solitary monoculture) and conventional double cropping (i.e. sequential monocultures) (Jeffers & Triplett 1979; Chan et al.
1980; Jeffers 1987,1990). Advantages include higher profitability, better soil and water conservation, and potentially, reductions in herbicide use. A fortuitous discovery is that soybean-wheat relay intercropping systems exhibit substantially reduced density of the potato leafhopper, Empoasca fabae (Harris), relative to soybean monocultures (Hammond & Jeffers
1990). Similar suppression of abundance of this insect has been found when alfalfe has been grown in the presence of grassy weeds (Lamp etal. 1984a) or oat (Lamp 1991) compared with alfalfe monocultures. Although the potato leafhopper (PLH) rarely causes economic damage to commercial varieties of soybean, it is among the most abundant herbivores inhabiting the crop in the Midwest (Helm e( at. 1980). Soybean in monoculture is a good host for this leafhopper during spring to mid-summer, facilitating reproduction of enormous numbers of the insects because of the extensive areas planted to the crop; PLH that develop on soybean may disperse to infest and damage other, more susceptible crops. The species has a wide host range,
10 including more than 200 known species (Poos & Wheeler 1943,1949; Lamp etal. 1984b; Ranne
& Lamp 1990; NC-193 Technical Committee 1993). It is a principal pest of leguminous field and
horticultural crops, potatoes, a variety of vegetables, and ornamentals (see Chapter 1 ). The frequency, extent, and type of damage produced by PLH feeding differs depending on host
plant. For alfalfa, where feeding behavior has been studied most extensively, damage results in yield loss, stand reduction, and reduction in forage value through loss of crude protein (Poos &
Johnson 1936. Helm etal. 1980, Flinn etal. 1990a).
PLH is believed to be unable to withstand severe cold weather, such as prevails in the
northern Midwest (DeLong & Caldwell 1935; Decker & Maddox 1967; Decker & Cunningham
1967, 1968; Specker et al. 1990; Shields & Sher 1992) and its winter range in the United States appears to be mainly restricted to southern states bordering the Gulf of Mexico and Atlantic
Ocean (NC-193 Technical Committee 1993). Each year, as temperatures increase in the spring. PLH expands its range to include most of the lower-elevation areas of the eastern U. S.
(DeLong 1938). This expansion in range is assisted by flights that take advantage of rapidly moving, northward-directed air masses typically associated with storm fronts (Medler 1957.
1962; Pienkowski & Medler 1964; Decker & Cunningham 1967; Taylor & Reling 1986; NC-193
Technical Committee 1990). Because PLH must reestablish itself throughout much of the northern portion of its range in spring or early summer of each year, it is believed to be vulnerable to manipulation if the factors that influence its movement and selection of habitats were better understood (Lamp & Zhao 1993). Indeed, suppression of PLH abundance has been reported from a variety of agroecosystems when vegetational diversity has been increased
(Lament 1981. Lamp etal. 1984a . Oloumi-Sadeghi etal. 1987. Hammond & Jeffers 1990. Lamp
1991, Roltsch & Gage 1990b. Coggins 1991. Andow 1992)
The resource concentration hypothesis has frequently been invoked in attempting to explain why certain herbivores exhibit decreased abundance when in vegetationally complex
11 habitats then in simple (usually monospecific) ones. This extremely durable and popular idea, formalized by Root (1973), proposed that herbivores respond to visual and chemical stimuli from host and non-host plants, as well as to microclimate and other environmental modifications produced by plants. Herbivore responses include movement and changes in rates of survival and reproduction. The complex of responses to environmental stimuli determines the relative attractiveness of the habitat, the evaluation of resource concentration made by individual herbivores. Factors believed to strongly influence this evaluation are the number, density, spatial arrangement, and relative attractiveness of hosts, mediated by interference from non hosts (Tahvanainen & Root 1972, Root 1973).
Hammond & Jeffers (1990) found that the presence of wheat in a soybean-wheat relay intercropping system was sufficient to protect from damage even glabrous soybean which is highly susceptible to injury by PLH. They employed a commercially viable version of an intercropping system (“full relay intercropping") that contained a relatively large amount of wheat.
The present series of field studies included a very similar “full relay intercropping" system
(different cultivars and a somewhat different seeding rate for wheat than used by Hammond &
Jeffers) and a soybean monoculture, and added two cropping systems that had amounts of wheat (numbers of rows) intermediate between the extremes. These cropping systems
■■epresent a gradient of vegetational diversity, produced by manipulating the amount of non-host present while attempting to minimize differences in host density and dispersion. In a study that also employed a gradient of vegetational diversity, consisting of five treatments with fixed bean
{Phaseolus vulgaris, a PLH host) and variable tomato {Lycopersicon esculentum, a PLH non host) density, Roltsch & Gage (1990b) reported an inverse relationship between density of the non-host and densities of PLH nymphs and eggs.
The objective of this study was to determine whether there was a differential response in abundance of PLH to the gradient of vegetational diversity represented by the cropping systems.
12 The null hypothesis stated that there was no difference among cropping systems in PLH abundance. An altemative hypothesis, based upon the resource concentration hypothesis, stated that PLH would be distributed within cropping systems in numbers inversely proportionate to the amount of wheat present.
13 MATERIALS AND METHODS
experimental design, agronomic procedures.The experiment was conducted on a Wooster
silt loam at the Ohio Agricultural Research and Development Center in Wooster, Ohio. Adjacent
fields were used during the two years of the study. The experimental design employed four
treatments replicated three times (1990-91) or four times (1991-92) in a randomized complete
block design. Three treatments were relay intercropping systems that combined different
amounts of soft red winter wheat {Triticum aestivum cv. Becker) with soybean {Glycine max cv.
Flyer); one treatment was a soybean monoculture. The seeding rate for wheat within rows was
the same for all cropping systems. The amount of wheat in cropping systems was determined
by controlling the number and spacing of wheat rows planted. Wheat was drilled (Moore Unidrill,
Moore Uni-Drill Ltd., Bally money, N. Ireland, U. K.) on 8 October 1990 and 15 October 1991 in
patterns that produced 14:3, 7:3, 3:3, and 0:3 ratios of wheat to soybean rows (Fig. 2.1). The
numerator of each ratio is the sum of wheat rows planted within the distance defined by a single
planter p ass through the field (8 feet wide, ca. 2.4 m); similarly, the denominator of these ratios
is the sum of the soybean rows planted in that same space. Because soybean was planted at
the same density and row spacing in all cropping systems, the denominator is always the same.
These ratio-designations will be used to refer to the cropping systems throughout this
dissertation. The cropping systems that have wheat and soybean co-present for part of the year
are also called intercropping systems. I have already referred to the cropping system without wheat for all of the year as a soybean monoculture, and will continue this practice. The seeding
rates for wheat were ca. 106, 53, and 23 kg/ha for the 14:3, 7:3, and 3:3 cropping systems,
respectively. Soybean (inoculated with Bradyrhizobium japonicum) was planted into gaps in the wheat planting on 9 May 1991 and 14 May 1992, using a no-till planter (Model 71 Flexi-planter,
John Deere Corporation, Moline, IL, U. S. A.). The estimated plant populations for soybean at germination were 305,100 plants/ha in 1991 and 319,800 plants/ha in 1992.
14 Fig. 2.1. Graphic representing the relationship of wheat to soybean rows in the 4 cropping system s studied. Only a small portion of each cropping system is indicated, that is, the width of a single planter pass through the field, approximately 2.4 m (8 feet). Wheat rows are represented by dashed lines; soybean rows are represented by solid lines. Within each planter pass, 3 soybean rows and a variable number of wheat rows (depending on treatment) are contained. a) 14 wheat rows; 3 soybean rows b) 7 w heat rows: 3 soybean rows c) 3 wheat rows: 3 soybean rows d) 0 wheat rows: 3 soybean rows (soybean monoculture).
15 Each plot (unique combination of treatment and block, N=12) m easured 30.5 m by 19.4 m, and contained 24 soybean rows. Rows were oriented north-south. Blocks were oriented east-west in 1990-91 and north-south in 1991-92. Within the study field, mowed grass driveways, 4.9 m wide and oriented east-west, were superimposed between blocks in 1990-91 and cut across blocks (separating cropping systems within blocks) in 1991-92. The driveways were planted to wheat at the time of wheat planting, but during the course of the experiment were colonized by a variety of plant species common in turf. The fields in which the experimental plots were located were surrounded by a driveway/buffer area 5.5-15.2 m wide, that was inhabited by perennial grasses, predominantly fescues {Festuca spp.) and ryegrass {Lolium spp.). Driveways and buffers within and adjacent to the experimental plots were kept mowed to <10 cm throughout the experiment. Adjacent fields surrounding the experiment were occupied by alfalfa, field com, soybean, wheat, or were left fellow.
Research plots and driveways were fertilized in mid-April of each year with 448 kg/ha 0-
26-26; 224 kg/ha 34-0-0 (NH4NO3) was applied to driveways and research plots containing wheat. Plots containing wheat were treated with 2,4-0 ester at 560.4 g a.i./ha on 22 April 1991 and 6 May 1992; during 1992, thifensulfuron also was applied to these plots at 17.5 g. a.i./ha.
Soybean monoculture plots were treated with 840.5 g a.i. linuron/ha and 1,681.0 g a.i. alachlor/ha within 2 d after planting. Weed escapes were spot-treated with a solution of glyphosate (1% in 1991, 3% in 1992) and 0.5% non-ionic surfectant (George II, Central
Petroleum Company, Cleveland, OH, U. S. A.) during May of each year.
sample collection. The sampling method for this study utilized a backpack-mounted D-Vac® suction device (D-Vac Vacuum Insect Net, D-Vac Company, Ventura, CA, U. S. A.) to capture arthropods on and near soybean plants (Dietrick 1961). Helm et al. (1980) believed this to be the best relative method for sampling populations of adult leafhoppers on soybean, and some
16 (Ogunlana & Pedigo 1974, Marston etal. 1976, SImonet ef a/. 1979, Fleischer ef a/. 1982) consider it to approach an absolute method of sampling. The collection cone of the D-Vac was lowered over the plants for ca. 1 sec at each of 10 locations within a plot, >1.5 m from the plot border. For each such sample, collection locations were regularly distributed over at least two soybean rows. The area under the cone at each location was 856 cm^, therefore each sample
(“10 suctions” - Lamp & Smith 1989) was equivalent to 0.856 m^ ground area. The cone was lowered to the greatest extent possible without damaging plants: if possible the lip of the cone w as placed in contact with the ground briefly. Fine mesh net-bags which intercepted the arthropods that were sucked into the cone were removed from the D-Vac and placed immediately into 0.95 liter capacity plastic freezer containers. The contents of the net-bags were killed by freezing at <-10°C. There were 3 samples per plot per collection date. Collections were made from 3:30 PM to dusk. Temperature conditions during sample collection ranged from 15-34°C; sampling was not conducted when wind velocity substantially disturbed plants, or if condensation or precipitation rendered plant surfaces moist.
During each year, a full complement of D-Vac sampling was initiated shortly after the initiation of substantial PLH immigration into the area. The decision regarding the appropriate date was based upon the emergence and growth of soybean plants to be sampled, and on early season preliminary sampling (using a D-Vac, sweep net, water pan trap, and/or visual observation) by this author and others, reports of leafhopper immigration by the Cooperative
Extension Service in Ohio and other states, and the occurrence of weather conditions favorable to migration (Medler 1957, Huff 1963, Pienkowski & Medler 1964, Decker & Cunningham 1967,
Taylor & Reling 1986, NC-193 Technical Committee 1990). The first samples were collected 4
June 1991 and 15 June 1992. Thereafter, samples were collected at 6-10 d intervals (mean =
7.8 d) in the period prior to the wheat harvest (5-12 July 1991, 9-16 July 1992). Following the harvest, the interval was 7-18 d (mean = 11.1 d) until soybean senescence. The order of block
17 sampling w as random. All samples were collected from within a block before proceeding to
sample from other blocks.
sample examination Samples were examined using a Wild 5A stereomicroscope or a Bausch
& Lomb Stereo Zoom 5 microscope. Adult PLH were sexed and counted. Identification of this
species was based upon a careful examination of external morphological characters, including
the size and shape of sclerites and setation, and upon coloration. The total number of samples
examined was 802.
analysis. In this study, the experimental unit is a single research plot. The basic unit of data is
a combination of all of the samples derived from the same plot. For convenience, I have used
the mean number of PLH per sample, rather than the sum of PLH from the 3 samples collected
in each plot. This unit of data was used to calculate means for cropping systems that are
presented in figures and tables, and it was the raw datum used to produce transformations for
statistical analysis. Data were transformed by adding 1.00000 to each datum, then converting the result to a base 10 logarithm. In almost all cases this succeeded in reducing the
heterogeneity of variances and rendering residuals sufficiently normal to permit analysis of variance testing. Data were analyzed separately for each collection date, and for each sex.
Datasets of transformed data were screened using Levene’s Test (Levene 1960; %VARTEST
Macro, Minitab 11, Minitab Inc., State College, PA, U. S. A.). Those datasets with non
significant scores (a=0.05) from this test were subjected to further analysis. In order to determine whether differences observed among cropping systems were statistically significant, a one-way analysis of variance test was performed (ONEWAY Procedure, Minitab 11 ). Residuals from analysis of variance tests were examined graphically in order to detect excessive deviation from a normal distribution (ONEWAY Procedure, GNORMAL Option, Minitab 11). If significant differences were found using the one-way test, means were compared using Tu key's w
18 Procedure (Zar 1974, Steel & Torrie 1980). This test used is a relatively conservative test, applicable to both planned and unplanned comparisons, which yields an experimentwise error rate, rather than a comparisonwise error rate (Steel & Torrie 1980, Sokal & Rohlf 1981).
19 RESULTS
As described so far, each study year spanned an autumn to autumn period, and they have been referred to as 1990-91 and 1991-92. However, for the sake of brevity and in order to reduce confusion, in the results and discussion sections each study year will be referred to by the portion during which samples and obsen/ations were made, 1991 and 1992.
Figures 2.2 and 2.3 present graphically the changes in abundance for adult female PLH for 1991 and 1992, respectively. Figures 2.4 and 2.5 provide the equivalent information for males. Results of analyses of variance and tables of means with means comparison test results are given in Table 2.1 (females) and Table 2.2 (males).
Females. On the first collection date in 1991, females were found to be present in very low abundance in all except one (7:3) cropping system (Fig. 2.2). While significant differences were detected in the analysis of variance test even at this early date, the means comparison test indicated that none of the cropping systems which had females present differed significantly
(Table 2.1). By the second sampling date, a consistent pattern of differences was established that remained in effect until after the wheat harvest. The mean number of females was very low, less than one per sample, and similar (no significant differences) among all three intercropping systems. Females showed little or no increase in abundance in the intercropping systems until after the wheat harvest. For the 7:3 and 14:3 cropping systems, very few plots had any females detected prior to the wheat harvest. In contrast, virtually every soybean monoculture plot had detectable levels of females on every sampling date throughout the study. In the second through fourth sampling dates, which preceded the wheat harvest, females were on average 16-
133 times more abundant in monoculture plots than in intercropping plots (Table 2.1). Analyses of variance showed these differences to be highly significant (one date with P<0.01, two with
P<0.001).
20 Immediately following the wheat harvest in 1991 (fifth sampling date), the pattern of PLH abundance was found to have changed. Females were more widely distributed within the intercropping plots, and the magnitude of differences between the soybean monoculture and intercropping systems was substantially reduced (Fig. 2.2). The source of the new colonists in the intercropping plots is not known, but it is probable that a high proportion were derived from plots of the soybean monoculture, or from adjacent fields (which included alfalfa). PLH of both sexes remain mobile throughout their adult lives, and based upon behavioral observations of
PLH in culture, newly developed adult females appear to be especially active. Following the wheat harvest, while density of females remained greater in the soybean monoculture than in the intercropping systems until the final sampling date, significant differences were found on only one of four dates. In 1991, the highest recorded mean density for females in monoculture occurred in mid-July, soon after the wheat harvest. The intercropping system which was most similar to the monoculture, 3:3, also reached its yearly maximum at that time. The cropping systems which had substantially greater amounts of wheat, 7:3 and 14:3, reached their highest levels during the second week of August. Female densities in all cropping systems declined thereafter.
In a typical year, PLH begin to accumulate in soybean plantings with emerged plants beginning in late May and early June; however in 1992 the major migration of PLH into the area was later than usual (Chapter 3). On the initial sampling date of 1992 in mid-June, small numbers of PLH females, similar to those detected at the beginning of June 1991, were collected from the soybean monoculture, but from no other cropping system (Fig. 2.3 and Table
2.1). By the second sampling date, 10 d later, soybean monoculture plots were inundated with
PLH. The mean number of females was considerably greater than the maximum collected during the previous year. Yet the differences in female abundance among the cropping systems studied were so great that the datasets did not qualify for analysis of variance (Levene’s test).
21 because even the transformed data could not be rendered sufficiently normal. Mean densities of females fluctuated at relatively high levels (11.9-26.8 females/sample) in the soybean monoculture during late June to early July until the wheat harvest, but remained near or below one per sample in the intercropping systems. Analysis of variance and means comparison showed that females were significantly more abundant (P<0.001) in the soybean monoculture than in the intercropping systems (Table 2.1).
As occurred during 1991, female densities in 1992 increased following the wheat harvest in all cropping systems, reaching their yearly maxima in late July for the soybean monoculture and early August for the intercropping systems. The monoculture had substantially greater densities of females until after mid-August. Of seven sampling dates following the wheat harvest, three had significant analysis of variance results (two with P<0.05, one with P<0.001).
For one additional date (16 July 1992), the significance level of the analysis of variance result was probably compromised by having a reduced number of blocks sampled (two and three, instead of four) because rain interrupted sampling and prevented completion of sampling for several days. During the two years of the study, no significant differences were found among intercropping systems in abundance of females (Table 2.1).
Males. The pattem of abundance and of differences among cropping systems was similar for males to that observed for females. The major differences were that the mean densities were less for males than females through most of the portion of the year studied. Males were especially scarce early in each year (Fig. 2.4 and 2.5, and Table 2.2). PLH engaged in long distance migration and early-season colonists of crops have been shown to be predominantly female (Click 1960, Medler ef a/. 1966, Decker ef a/. 1971, Taylor & Reling 1986, Flinn etal.
1990b). In 1991, observations of PLH nymphal development on soybean in the experimental plots and on nearby snap beans showed that some of the first mature nymphs were developing
22 into adults on 24 June (TJM unpublished). This coincided with a substantial decrease in the proportion of total PLH collected that were female, from 86% to 52% between 14 and 27 June
(Appendix A). Male abundance exceeded that of females in late August and September. This may be an indication of differential sexual emigration from these cropping systems (Taylor &
Reling 1986, Flinn et al. 1990b). These exceptions aside, the dynamics of males and females were substantially the same.
In 1991 and 1992, except for the initial sampling date each year, when males were very scarce in all cropping systems, all sampling dates preceding the wheat harvests had significant analysis of variance results (two with P<0.05, one with P<0.01, three with P In each of these cases, males reached significantly greater densities in the soybean monoculture than in any intercropping systems, at least at the P<0.05 level. Means comparisons did not detect any significant differences among intercropping systems during the period preceding the wheat harvest. Following the wheat harvest, significant differences were found through analysis of variance tests in two of four dates in 1991 and two of seven dates in 1992 (one with P<0.05, two with P<0.01, one with P<0.001; Table 2.2). For half of these dates, the soybean monoculture was significantly distinct from all the intercropping systems; in the other cases it was not significantly distinguishable from the 3:3 cropping system. As was the case with females, means comparison tests detected no significant differences in male abundance among the intercropping systems (Table 2.2). 23 Table 2.1 Results of Analysis of Variance Tests and Means Comparisons for Female PLH Collection Date Result of Cropping System Analysis of Variance Test 0:3 3:3 7:3 14:3 4 June 1991 P=0.044 1.444a 0.006ab 0.000b 0.333ab 14 June 1991 P<0.001 14.778A 0.500B 0.111B 0.333B 20 June 1991 P=0.008 10.667a 0.667b 0.000b 0.000b 27 June 1991 P<0.001 9.000A 0.111B O.OOOB O.OOOB 17 July 1991 P=0.203 15.111 5.556 5.778 6.222 26 July 1991 P=0.009 13.000a 2.778ab 4.333b 2.333b 10 August 1991 P=0.760 8.000 5.111 8.889 6.333 28 August 1991 P=0.574 1.333 0.889 2.556 1.889 15 June 1992 not tested 1.417 0.000 0.000 0.000 25 June 1992 not tested 26.833 0.000 0.250 0.167 2 July 1992 P<0.001 11.917A O.OOOB 0.583B 0.250B 9 July 1992 P<0.001 23.333A 0.167B 1.083B G.273B 16 July 1992 P=0.054 13.556 0.778 2.167 1.000 23 July 1992 P<0.001 61.364A 4.091 B 4.583B 2.250B 1 August 1992 P=0.011 54.333a 13.364ab 7.714b 5.182b 11 August 1992 P=0.026 34.700a 13.091 ab 12.364ab 7.000b 19 August 1992 P=0.098 8.333 8.500 11.333 3.667 4 Sept. 1992 P=0.815 0.556 1.000 1.091 0.700 14 Sept. 1992 P=0.532 0.417 0.500 1.000 0.200 Table 2.1. Results of analysis of variance tests and means comparisons for female PLH data. Data were initially screened using Levene’s test. Those dates with data to which this test could not be applied, or which produced significant values (a=0.05) when Levene's test was applied, were not tested further. The unit of data used as the dependent variable was a log^ (X +1) transformation of the mean number of females collected per D-Vac bag, averaged over the experimental plot (3 bags per plot, 3 plots per cropping system in 1991, 4 plots per cropping system in 1992). The single main effect tested was cropping system (4 levels). A separate analysis was made for each collection date. The P value indicates the probability that an equal or greater F (ONEWAY Procedure, Minitab 11) would occur by chance. Each mean shown is the overall average for all plots of the cropping system for the date indicated. Within the same date, m eans followed by the sam e letter were not found significantly different using Tukey’s w procedure. Upper-case letters indicate means comparison results when a=0.01, lower-case letters indicate results when a=0.05. 24 Table 2.2 Results of Analysis of Variance Tests and Means Comparisons for Male PLH Collection Date Result of Cropping System Analysis of Variance Test 0:3 3:3 7:3 14:3 4 June 1991 P=0.596 0.111 0.000 0.000 0.111 14 June 1991 P<0.001 2.333A O.OOOB 0.111B 0.111B 20 June 1991 P=0.021 4.556a 0.000b 0.111b 0.000b 27 June 1991 P=0.001 8.222A O.OOOB O.OOOB 0.111B 17 July 1991 P=0.012 7.333a 2.889ab 2.000b 1.111b 26 July 1991 P=0.002 9.000a 1.556b 2.667b 2.000b 10 August 1991 P=0.301 6.889 6.000 8.556 2.556 28 August 1991 P=0.626 1.111 0.444 2.000 1.222 15 June 1992 P=0.588 0.000 0.083 0.083 0.000 25 June 1992 P=0.011 0.917a 0.000b 0.083b 0.000b 2 July 1992 P=0.004 0.917a 0.000b 0.083b 0.083b 9 July 1992 P<0.001 7.333A 0.167B 0.583B 0.182B 16 July 1992 P=0.065 10.556 0.222 0.333 0.333 23 July 1992 P<0.001 52.091A 3.091 B 3.667B 2.333B 1 August 1992 P=0.005 50.111a 12.000ab 6.143b 2.455b 11 August 1992 P=0.063 19.900 8.818 12.182 3.889 19 August 1992 P=0.162 5.667 5.250 9.222 3.000 4 Sept. 1992 P=0.280 0.444 1.125 1.273 0.700 14 Sept. 1992 P=0.141 0.417 0.667 0.917 0.100 Table 2.2. Results of analysis of variance tests and means comparisons for male PLH data. Data were Initially screened using Levene's test. Those dates with data to which this test could not be applied, or which produced significant values (a=0.05) when Levene’s test was applied, were not tested further. The unit of data used as the dependent variable was a log^o (X +1 ) transformation of the mean number of males collected per D-Vac bag, averaged over the experimental plot (3 bags per plot, 3 plots per cropping system in 1991,4 plots per cropping system in 1992). The single main effect tested was cropping system (4 levels). A separate analysis was made for each collection date. The P value indicates the probability that an equal or greater F (ONEWAY Procedure, Minitab 11) would occur by chance. Each mean shown is the overall average for all plots of the cropping system for the date indicated. Within the same date, m eans followed by the sam e letter were not found significantly different using Tukey’s w procedure. Upper-case letters indicate means comparison results when a=0.01, lower-case letters indicate results when a=0.05. 25 ABUNDANCE OF FEMALE PLH BY CROPPING SYSTEM, 1991 y a. I -A— 14:3 Q ÛC m CL Ui UJ ai LL I 9 9 9 9 9 9 9 9 9 9 c c c c O) O) a t O) & S' 3 3 3 3 3 3 3 3 0> s 7 7 7 7 N J) (6 (4 < < < < 9 00 CO lA (6 d> ch CNJ (M CM CM COLLECTION DATE Fig. 2.2. Mean adult female PLH captured per D-Vac sample by cropping system, 1991. Each mean represents 3 samples per plot, 3 plots per cropping system (see text). Arrows indicate the beginning and end of the wheat harvest. ABUNDANCE OF FEMALE PLH BY CROPPING SYSTEM, 1992 0:3 "O ' 3:3 a. o- 7:3 S -Û - 14:3 w i a: lij CL K) u.i z S5 CM CN (N CM CM CM CM CM CM CM CM CM 9 9 9 9 9 9 9 9 9 CJ) O ) O ) C c C C 0 0 O) O) O)&&& 3 3 3 3 3 3 3 3 3 0) 0) 0) ■7 < < < < < f (6 (6 9 9 9 CO [A CM d ) c4 d> CM CM CM CO CM COLLECTION DATE Fig. 2.3. Mean adult female PLH captured per D-Vac sample by cropping system, 1992. Each mean represents 3 samples per plot, 4 plots per cropping system (see text). Arrows indicate the beginning and end of the wheat harvest. ABUNDANCE OF MALE PLH BY CROPPING SYSTEM, 1991 11 10 -O- 0:3 9 ■O' 3:3 y 8 I 7 -A-- 14:3 6 1 5 o: 4 tu QL 3 •O. _ — — A' ÎS 2 1 00 0 1 cn o> O) o> O) O) O) O) O) 9 O) %%% O) CM o> CM COLLECTION DATE Fig. 2.4. Mean adult male PLH captured per D-Vac sample by cropping system, 1991. Each mean represents 3 samples per plot, 3 plots per cropping system (see text). Arrows Indicate the beginning and end of the wheat harvest. ABUNDANCE OF MALE PLH BY CROPPING SYSTEM, 1992 60 55 50 -O - 0:3 ai D 3:3 _j 45 Q. o 7:3 40 —A" 14:3 I 35 ; 30 25 g 20 a. 15 ÎS 10 5 I 0 5 C4 CD % Î o> Î CO 0 0 O) CM CM cO CO COLLECTION DATE Fig. 2.5. Mean adult male PLH captured per D-Vac sample by cropping system, 1992. Each mean represents 3 samples per plot, 4 plots per cropping system (see text). Arrows indicate the beginning and end of the wheat harvest. DISCUSSION This study found substantial and significant levels of suppression of PLH population density in soybean-wheat relay intercropping systems, compared with the soybean monoculture. This was particularly in evidence during the early portion of the sampling period of each year. Prior to the wheat harvest, overall densities for cropping systems ranged from 1.4-14.8 females per sample in 1991 and 1.4-26.8 females per sample in 1992 in the soybean monoculture, with the range among all intercropping systems only 0-1.1 females over both years. This time period has a critical influence upon seasonal population dynamics of PLH because it coincides with the spring influx of migrant PLH that colonize crop habitats in the upper Midwest. The intercropping systems that had negligible densities prior to the wheat harvest persisted in having considerably lower densities than the soybean monoculture long after the wheat was removed; within collection dates, none of the mean densities of the intercropping systems exceeded those of the soybean monoculture until the second (1991 ) or third (1992) week of August. At that point of the season, abundance of PLH in soybean had entered into a state of decline. This trend may be related to decreases in quality of leaf substrate and to higher temperatures, which inhibit oviposition and survivorship (DeLong 1938, Kieckheffer & Medler 1964, Kouskolekas & Decker 1966, Decker et al. 1971, Taylor 1989). Reduced density in the intercropping systems translates into considerably diminished (or at least displaced) population growth, relative to the monoculture. While PLH usually does not cause economic damage to varieties of soybean that are commonly grown, those pests that develop on soybean may disperse to damage- susceptible crops (e.g. alfalfe, potato, snap and dry beans, omamentals). Except for the exceptions noted previously, the temporal dynamics and relationships of male density among the cropping systems studied were similar to those of the females. Because of the similarity, the results with males tend to buttress the conclusions reached from the female data. 30 During the two years of the experiment, there were never any significant differences detected among females in soybean-centered traps of the three intercropping systems. This is surprising in light of the wide range in the amount of wheat present, in the similarity in the vegetative characteristics (Appendix D) of soybean in the 3:3 cropping system to those in the soybean monoculture, and the relative proximity among experimental plots (i.e. considerably higher densities were nearby). Not only were there no significant differences found among the intercropping systems, there was not a trend toward higher density as the amount of wheat decreased. Over 19 sample collection dates, the 7:3 cropping system had the greatest mean among the intercropping systems 12 times, and the 3:3 system only 4 times (for males it was 11 times to 2 times). The purpose of this study was to quantify PLH abundance across a gradient of crop diversity. Because of the large differences between soybean monoculture and the intercropping systems, the null hypothesis, that there was no difference among cropping systems in PLH abundance, was rejected. However, the proposed alternative hypothesis, which predicted that PLH would be distributed among cropping systems in numbers inversely proportionate to the amount of wheat present, cannot be accepted, because of the absence of significant differences among the intercropping systems. Therefore, the results of this research lend only qualified support to the resource concentration hypothesis. When comparing monoculture to the more vegetationally diverse intercropping systems in as a group, the hypothesis is strongly supported by the magnitude and consistency of the differences recorded. However, within the range of intercropping systems there was no discernable response of herbivore abundance to concentration of host. The failure to find differences in PLH density among the intercropping systems suggests that host plant quality differences are a less important factor than the presence of wheat in influencing PLH abundance. For additive intercropping designs (Vandermeer 1989), such as 31 was used in this study, superimposing a pattern of one plant type over a fixed density and dispersion of another type generally increases the intensity of interspecific competition as overall plant density is enhanced. Increasing the number of rows of wheat by a factor of 4.7 (3:3 to 14:3) may have a detrimental impact on soybean leaf area, biomass accumulation, phenology (Appendix D), and stress level (e.g. as indicated by wilting), and is likely to influence habitat selection by herbivores (Kareiva 1983, Andow 1983b, Lamp 1991). In 1991, a drought year, such signs of reduced plant quality were observed by this author on soybean in the 14:3 cropping system, and to a lesser extent in the 7:3 cropping system. Hammond & Jeffers (1990) found that for pubescent soybean, intercropping with wheat produced significant decreases in leaf area index, and in som e cases significant seed yield reductions. PLH has been found to respond to differences in host plant quality. For example, Hammond & Jeffers (1990) and the present author have observed that PLH may abandon severely damaged soybean plants, allowing for some recovery from injury. Roltsch & Gage (1990b) concluded that although bean plants in their experiment had reduced quality and reduced resource availability to PLH as competition with non-hosts increased, PLH response to reduced host quality/quantity was unable to account for the suppression of abundance observed in the presence of non-hosts. The results of this study corroborate the conclusions of Hammond & Jeffers (1990), that PLH abundance is suppressed in the presence of wheat occurring in soybean-wheat relay intercropping systems, relative to soybean monoculture. Those conclusions are extended to include densities of wheat that are considerably less than that used in the previous study. Aside from what has already been discussed regarding the relationship of PLH density to host plant quality, the methods of this study were not well-designed to explore the possible mechanisms responsible for the differences in abundance. A companion study (Chapter 3) conducted concurrently in the same experimental plots explored differential colonization as a possible explanation for these differences. The patterns of PLH density observed in the present 32 work appear to directly reflect the pattern of differential colonization reported. They strengthen the argument that the primary influence of the presence of wheat on PLH is through influencing its movement into habitats. Smith (1987), Smith etal. (1992), and Lamp & Zhao (1993) provide additional experimental evidence specifically addressing the effect of grasses on PLH movement. The significance of this work is that it extends our understanding of how PLH interacts with mixed-crop habitats, and its evidence is supportive of other work. Of particular interest was the substantial level of PLH population suppression produced in the cropping system with the lowest amount of wheat. If comparable or lower densities of a non-host can reliably interfere with PLH, intercropping with low densities of pest-suppressing plants might find applicability in producing crops to which PLH is a significant threat 33 CHAPTER 3 COLONIZATION OF FOUR CROPPING SYSTEMS BY THE POTATO LEAFHOPPER, AS STUDIED USING WATER PAN TRAPS INTRODUCTION Mixed cropping, the practice of growing different species of crops in close temporal or spatial proximity, has been used in myriad forms in early and traditional agriculture, and is the dominant mode of producing some crops in tropical regions today. Recently there has been an increase in interest in mixed cropping in North America as growers and agricultural technologists explore alternative and improved ways of growing crops. In general, the advantage sought is that the efficiency of land use, usually expressed as relative yield total or land equivalence ratio\ should be maximized (Vandermeer 1989), or that profitability for the mixed cropping system should be greater than for other competing cropping systems (Hildebrand 1976, Hook & Gascho 1988). However, in some cases other advantages have been discovered for particular systems, such as greater price stabilization or improved pest management, that may recommend their use even when the relative yield total is not as great as other alternatives (Gliessman 1985, Vandermeer 1989). Relay intercropping describes a subclass of mixed cropping systems being examined as an innovative alternative for efficient use of land and favorable growing conditions. In the northern Midwest, relay intercropping of soybean and winter wheat has been shown to have ’ land equivalence ratio or relative yield total = the ratio of yield per unit area in polyculture to yield per unit area in monoculture for each crop, summed for all crops in the polyculture 34 economic and agronomic advantages over monocropping (i.e. solitary monoculture) and conventional double cropping (i.e. sequential monocultures) (Jeffers etal. 1977; Jeffers and Triplett 1979; Chan etal. 1980; Jeffers 1987,1990). Advantages include higher profitability, better soil and water conservation, and potentially, reductions in herbicide use. A fortuitous discovery is that soybean-wheat relay intercropping systems exhibit substantially reduced density of the potato leafhopper, Empoasca fabae (Harris), relative to soybean monocultures (Hammond & Jeffers 1990). Similar suppression of abundance of this insect has been found when alfalfa has been grown in the presence of grassy weeds (Lamp et ai. 1984a), oat (Lamp 1991), or forage grasses (Coggins 1991) compared with alfalfa monocultures. While this leafhopper infrequently causes economic damage to commercial varieties of soybean, the crop grown in monoculture is a good host for this leafhopper during spring to mid summer, facilitating reproduction of massive numbers of the insects. Many of these may disperse to infest and damage other, more susceptible crops. The potato leafhopper (PLH) is a serious pest of a wide variety of crops over much of the eastern United States, including farm and home garden situations (see Chapter 1). PLH is believed to be unable to withstand severe cold weather, such as prevails in the northern Midwest (DeLong & Caldwell 1935; Decker & Maddox 1967; Decker & Cunningham 1967, 1968; Specker et al. 1990; Shields & Sher 1992) and its winter range in the United States appears to be mainly restricted to southern states bordering the Gulf of Mexico and Atlantic Ocean (NC-193 Technical Committee 1993). Each year, as temperatures increase in the spring, PLH expands its range to include most of the lower-elevation areas of the eastern U. S. (DeLong 1938). This expansion in range is assisted by flights that take advantage of rapidly moving, northward-directed air m asses typically associated with storm fronts (Medler 1957, 1962; Pienkowski & Medler 1964; Decker & Cunningham 1967; Taylor & Reling 1986; NC-193 Technical Committee 1990). 35 Because PLH must reestablish itself throughout much of the northern portion of its range in spring or early summer of each year, understanding the migration and crop colonization process would greatly enhance our ability to predict and possibly control PLH populations. Considerable advancement has been made recently in our knowledge of different aspects of the migration process, largely through the efforts of members of the NC-193 Technical Committee (NC-193 Technical Committee Annual Reports 1990-1994). Some of them have explored PLH- habitat interactions, including movement, in cropping systems involving alfelfa and grasses (Lamp 1991, Lamp etal. 1984ab, Lamp & Zhao 1993) and tomato and bean (Roltsh & Gage 1990ab). Hammond & Jeffers (1990) found that the presence of wheat in a soybean-wheat relay intercropping system was sufficient to protect from damage even glabrous soybean which is highly susceptible to injury by PLH. However, their research was not able to determine the specific mechanism responsible for reduced PLH densities when wheat was present. They employed a commercially viable version of an intercropping system (“full relay intercropping") that contained a relatively large amount of wheat The present series of field studies included a very similar “full relay intercropping" system (different cultivars and a somewhat different seeding rate for wheat) and a soybean monoculture, and added two cropping systems that had amounts of wheat intermediate between the extremes. Together, the four cropping systems represent a gradient of vegetational diversity (Roltsh & Gage 1990b). The study of PLH adult abundance presented in Chapter 2, and Hammond & Jeffers (1990) found that PLH is detected in only relatively low numbers, if at all, in intercropping systems during the early part of the season while wheat is present. However, the sampling methods used to determine this (D-Vac® suction device, Dietrick 1961), indeed most commonly used leafhopper sampling methods (e.g. use of D-Vac, sweep net, or direct observation; Helm et ai. 1980), have the disadvantage of measuring only very short intervals of time. They are almost 36 exclusively used during daylight hours when weather conditions are relatively good. Insects that might be present in plots only briefly, or available to be sampled during times of the day beyond the range of the usual sampling devices, will not be detected. This study used water pan traps for continuous sampling, in order to capture migrants that might have entered the habitats of the different cropping systems, but been removed (e.g. by emigration or predators) before they might be detected by other methods. This experiment used water pan traps located within the soybean canopy to measure habitat colonization. For the purposes of this study, habitat colonization is defined as making physical contact with host plants (soybean) or arriving within the host plant canopy (including traps placed therein). This definition is similar to Thorsteinson’s (1960) use of host finding: arriving near or on a resource. While a more widely accepted use of the term colonization might include, in addition to finding, examining and consuming hosts (Miller & Strickler 1964), it is assum ed that migrating PLH have a high probability of remaining on host plants, at least for a short period of time, once contact is made (Flinn et al. 1990b). Therefore recovery from traps located in the soybean canopy is considered to be an indication of colonization. Even if contact is only momentary, detection of PLH within the host canopy of a cropping system provides evidence that the leafhoppers have the ability to find and the opportunity to exploit (consume) the resources there. This study was used provided a test of the resource concentration hypothesis (Tahvanainen & Root 1972, Root 1973). This states that diet-specialized herbivores are more likely to find and remain in habitats where their host plants are concentrated. Evaluation of resource concentration is made by herbivores based upon an interaction among internal states and environmental stimuli (Dethier 1982, Miller & Strickler 1984). For herbivores, stimuli from vegetation is thought to be especially important. Reduction in the density of host plants, unfavorable alteration of their spatial arrangement, or increase in the frequency of non-host 37 plants is believed to inhibit host location or exploitation. Herbivores may demonstrate a greater propensity to leave a habitat with less concentrated host resources, or exhibit reduced rates of population increase. The objective of this study was to determine whether PLH colonized host plants (soybean) in the habitats of the different cropping systems differentially, in response to varying levels of vegetational diversity. The null hypothesis predicted that there was no difference in habitat colonization. An altemative hypothesis, as predicted by the resource concentration hypothesis, was that there would be a graded response in diminished habitat colonization as the amount of wheat (non-host) increased. 38 MATERIALS AND METHODS experimental design, agronomic procedures.The experiment was conducted on a Wooster silt loam at the Ohio Agricultural Research and Development Center in Wooster, Ohio. Adjacent fields were used during the two years of the study. The experimental design employed four treatments replicated three times in a randomized complete block design. Three treatments were relay intercropping systems that combined different amounts of soft red winter wheat {Triticum aestivum cv. Becker) with soybean {Glycine max cv. Flyer); one treatment was a soybean monoculture. The seeding rate for wheat within rows was the same for all cropping systems. The amount of wheat in cropping systems was determined by controlling the number and spacing of wheat rows planted. Wheat was drilled (Moore Unidrill, Moore Uni-Drill Ltd., Bally money, N. Ireland, U. K.) on 8 October 1990 and 15 October 1991 in patterns that produced 14:3, 7:3, 3:3, and 0:3 ratios of wheat to soybean rows. The numerator of each ratio is the sum of wheat rows planted within the distance defined by a single planter pass through the field (8 feet wide, ca. 2.4 m); similarly, the denominator of these ratios is the sum of the soybean rows planted in that same space. Because soybean was planted at the same density and row spacing in all cropping systems, the denominator is always the same. These ratio-designations will be used to refer to the cropping systems throughout this dissertation. The cropping systems that have wheat and soybean co-present for part of the year are also called intercropping systems. I have already referred to the cropping system without wheat for all of the year as a soybean monoculture, and will continue this practice. The seeding rates for wheat were ca. 106, 53, and 23 kg/ha for the 14:3, 7:3, and 3:3 cropping systems, respectively. Soybean (inoculated with Bradyrhizobium Japonlcum) was planted into gaps in the wheat planting on 9 May 1991 and 14 May 1992, using a no-till planter (Model 71 Flexi-planter, John Deere Corporation, Moline, IL, 39 u. s. A.). The estimated plant populations for soybean at germination were 305,100 plants/ha in 1991 and 319,800 plants/ha in 1992. Each plot (unique combination of treatment and block, N=12) measured 30.5 m by 19.4 m, and contained 24 soybean rows. Rows were oriented north-south. Blocks were oriented east-west in 1990-91 and north-south in 1991-92. Within the study field, mowed grass driveways, 4.9 m wide and oriented east-west, were superimposed between blocks in 1990-91 and cut across blocks (separating cropping systems within blocks) in 1991-92. The driveways were planted to wheat at the time of wheat planting, but during the course of the experiment were colonized by a variety of plant species common in turf. The fields in which the experimental plots were located were surrounded by a driveway/buffer area 5.5-15.2 m wide, that was inhabited by perennial grasses, predominantly fescues {Festuca spp.) and ryegrass {Lolium spp.). Driveways and buffers within and adjacent to the experimental plots were kept mowed to <10 cm in height throughout the experiment. Adjacent fields surrounding the experiment were occupied by alfalfa, field com, soybean, wheat, or were left fallow. Research plots and driveways were fertilized in mid-April of each year with 448 kg/ha 0- 26-26; 224 kg/ha 34-0-0 (NH4NO3) was applied to driveways and research plots containing wheat. Plots containing wheat were treated with 2,4-D ester at 560.4 g a.i./ha on 22 April 1991 and 6 May 1992; during 1992, thifensulfuron also was applied to these plots at 17.5 g. a.i./ha. Soybean monoculture plots were treated with 840.5 g a.i. linuron/ha and 1,681.0 g a.i. alachlor/ha within 2 d after planting. Weed escapes were spot-treated with a solution of glyphosate (1% in 1991, 3% in 1992) and 0.5% non-ionic surfactant (George II, Central Petroleum Company, Cleveland, OH, U. S. A.) during May of each year. 40 description of water pan traps.Each trap consisted of a clear polystyrene pan (Model No. 395C, Tri-State Plastics, Dixon, KY, U. S. A.) with exterior dimensions 32.5 cm (length), 26.0 cm (width), and 10.3 cm (height); the interior trapping surfece was 30.0 cm by 23.5 cm. Each pan was supported by a frame constructed of 1.3 cm interior diameter PVC irrigation pipe and wire. By adjusting the length of the frame’s legs, the pan was always positioned at the top of the crop canopy. Pans were maintained at this height in order to preferentially intercept PLH flying into the plant canopy from above, and to avoid collection of adults moving laterally among plants. An aqueous solution of 1.5% NaCI, 1.0% glycerine, and 0.5% surfactant (George II) was placed in each pan, sufficient to cover the bottom to a depth of 1.3-2.5 cm. This solution was replaced at weekly intervals, or supplemented or replaced more frequently in response to dilution by rain, desiccation, or a large influx of arthropods, dirt, or other contaminants. Preliminary trials during 1990 indicated that these traps passively intercepted PLH moving into plots from above but did not attract PLH. The clear bottom of the pan permitted the plants beneath it to be seen through the fluid. The total trapping area of the three traps with the sam e placement located within each experimental unit was 2115 cm^. placement of traps in the field.In each plot (including those with wheat), 3 traps were centered in soybean rows, and located near a line running east-west through the center of the plot. Traps were placed in rows 6 or 7, 12 or 13, and 17 or 18, such that traps were located in the eastern third, central, and westem third of each plot. In plots containing wheat, additional traps were centered in wheat sections adjacent to soybean-centered traps, ca 1.5-2 m north or south of soybean-centered traps. Thus, each intercropping plot contained 3 soybean-centered traps and 3 wheat-centered traps. As mentioned earlier, adjustments were made as needed to keep each trap at the same height as the top of the plant canopy in the row(s) where it was centered. In the intercropping systems, soybean-centered traps were lower than wheat- 41 centered traps because soybean were shorter than wheat until the wheat harvest. Traps were placed in the field prior to the onset of seasonal mass migration of the potato leafhopper into the area. The decision to begin trapping was based on the coincidence of weather conditions favorable to migration, and local and regional reports of initial recovery of this species. Sampling w as initiated on 19 May 1991 and 22 May 1992. Immediately prior to the wheat harvest (5-11 July 1991 and 15 July 1992), traps were removed from research plots. In 1991 the soybean- centered traps were returned to the field (12 July) and sampling continued until soybean senescence (1 October). retrieval of trap contents.Arthropods were collected from each trap by pouring the trap fluid through a patch of fine-mesh organdy (pore size ca. 0.24 mm), which filtered out all except the smallest solid matter. The cloth patches with arthropods were then placed individually in labelled portion cups (Model LUS-3 or LUS-5,, Sweetheart Plastics, Inc., Chicago, IL, U. S. A.), that were stored at ca. -23®C until they could be examined. The interval during which traps could intercept arthropods, preceding retrieval (“sampling interval”), varied between 2 and 10 d; however, of 48 sampling intervals, only 4 in 1990-91 and 2 in 1991-92 were > 5 d. Arthropods were usually collected from traps twice per week (on Monday and Thursday). The contents of all of the traps within the field were retrieved within a 1-2.5 h period, beginning in early afternoon. laboratory procedures.Samples were examined using a Wild 5A stereomicroscope. Adult PLH were counted and classifed as to sex. In addition to PLH, other species of leafhoppers, and species of Mymaridae were also collected and classified. Some mymarids have been found to be potentially important natural enemies of PLH and other leafhoppers (Fenton & Hartzell 1923; Steiner 1938; Doutt & Nakata 1965, 1973; DeLong 1971; Waloff 1980; Frey tag 1985; 42 Huber 1986; Letoumeau 1990; TJM unpublished; D. Liewehr personal communication). Other species of leafhoppers may provide resources for development and maintenance of populations of natural enemies of PLH. A list of species derived from sample examination is provided in Appendix B. The total number of traps examined was 2452. analysis. In this study, the experimental unit is a single research plot The basic unit of data is the sum of PLH recovered from the three traps with the sam e placement located within the plot. Males and females were considered separately. Raw data were manipulated in three ways in order to provide different views of the pattem of PLH behavior. 1 ) presence in sampling intervals: This variable considers whether any PLH were recovered from traps of the group of interest (e.g. cropping system) during the sampling interval. Its utility is in providing an elementary tool to view patterns of colonization where the magnitude of numbers of insects recovered might tend to obscure the pattem. 2) cumulative numbers: Raw plot counts for sampling intervals are summed over the period of interest. My primary interest was in the period between the initiation of sampling and the wheat harvest. 3) PLH per plot per day of the sampling interval: This variable is most useful when comparing data from different sampling intervals (which may vary in length). It was used to create graphs and to produce transformed data for statistical analysis. The means of males and females recovered from all plots of the same cropping system per day were graphed for all sampling intervals so that differences could be compared visually (3D Sequential Ribbon Graphs, Statistica for Windows Version 5, StatSoft Inc., Tulsa, OK, U. S. A.). In order to determine whether differences observed among cropping systems were statistically significant, analysis of variance tests were performed comparing PLH recovered from traps with the sam e placement within sampling intervals (ONEWAY Procedure, Minitab 11, Minitab Inc., State College, PA, U. S. A.). One series of one-way analysis of variance tests was conducted for the soybean-centered traps (soybean monoculture and three 43 intercropping systems) and one series for the wheat-centered traps (three Intercropping systems only). Means comparisons were performed when analysis of variance tests showed significant results (Tukey’s w Procedure; Zar 1974, Steel & Tom e 1980). The test used is a relatively conservative test applicable to both planned and unplanned comparisons. It yields an experimentwise error rate, rather than a comparisonwise error rate (Steel & Tome 1980, Sokal & Rohlf 1981). In addition to the one-way analyses of variance, a series of two-way analysis of variance tests were conducted (ANOVA Procedure, Minitab 11) for the intercropping systems, which contained both soybean-centered traps and wheat-centered traps. The purpose of these tests was to determine whether there were significant differences among the relatively similar cropping system s, and to leam whether traps with different placement yielded significantly different results. For many sampling intervals, numbers of PLH recovered were very low, or there were no PLH present, for some groups for which comparison was desirable. For the purpose of performing statistical analysis, data were transformed by adding 0.5 to the mean of PLH recovered per day from each plot, then taking the square root. This transformation succeeded in reducing the heterogeneity of variances among groups compared and rendering their residuals more normal. In most cases, application of Levene’s Test (Levene 1960; %VARTEST Macro, Minitab 11) to the transformed data indicated that variances were sufficiently homogeneous to permit valid analyses of variance to be made. When this test indicated that the heterogeneity of variances for the groups of interest exceeded the test criterion (p<0.05), analysis of variance was not done for that sampling interval. Residuals from analysis of variance tests were examined graphically in order to detect excessive deviation from a normal distribution (ONEWAY Procedure, GNORMAL Option, Minitab 11). Only when the above criteria were met were analysis of variance test results and means comparisons reported here. 44 RESULTS As described so far, each study year spanned an autumn to autumn period, and they have been referred to as 1990-91 and 1991-92. However, for the sake of brevity and in order to reduce confusion, in the results and discussion sections each study year will be referred to by the portion during which samples and observations were made, 1991 and 1992. Table 3.1 shows the numbers of sampling intervals that had traps with PLH present, separated according to cropping system, placement of trap, sex of recovered specimens, and time period of interest. Table 3.2 provides cumulative counts for PLH recovered from traps in these categories. Figures 3.1 through 3.8 show the comparative temporal dynamics of PLH migration, for different cropping systems, sexes, and placement of traps. Each figure compares the mean of PLH captured per day of each sampling interval to that for traps with the same placement (soybean-centered traps or wheat-centered traps) in the same sampling interval in other cropping systems. Female data (Fig. 3.1 - Fig. 3.4) are presented separately from male data (Fig. 3.5 - Fig. 3.8). Fem ales soybean-centered traps. For the period between the installation of traps in experimental plots and the wheat harvest, there were great differences in the frequency with which PLH were detected in soybean-centered traps among the cropping systems studied. During 1991, when PLH migration into the area started earlier and continued over a longer period than in 1992, females were recovered during every sampling interval in soybean monoculture traps prior to the 45 wheat harvest, but rarely recovered from traps in any of the three intercropping systems (Table 3.1). There were three waves of substantial PLH immigration into the experimental plots in 1991 (Fig. 3.1, 3.2, 3.5, and 3.6); the first occurred within the period 23-31 May, the second 10-17 June, and the third 27 June -1 July. A total of 202 females was collected prior to the wheat harvest in soybean-centered traps in the monoculture, as compared with 1, 3, and 1 females collected in 3:3, 7:3, and 14:3 intercropping systems, respectively. These low numbers represented only 0.5-1.5 % of the quantity recovered from the soybean monoculture. Based on the criterion of female presence in sampling units (Table 3.1), the comparison for the same groups was less disparate for 1992. This was because the major migration of PLH into the area, which usually begins in late May, in 1992 did not begin until the third week of June (Fig. 3.3, 3.4, 3.7, and 3.8). Only 5 female and 1 male PLH were recovered from the combined traps (all cropping systems) between 22 May and 16 June (25 d). Between 16 June and the wheat harvest on 14 July (28 d), the sums of females collected in soybean-centered traps were 322 for the 0:3 cropping system (soybean monoculture), and 4, 5, and 5 for the 3:3, 7:3, and 14:3 systems. For the entire period preceding the 1992 wheat harvest, the numbers recovered from the intercropping systems ranged from 1.5% to 1.9% of that of the soybean monoculture (Table 3.2). Analysis of variance testing of differences for soybean-centered traps among cropping systems prior to the wheat harvest was problematic. Because of the lack of females in intercropping systems in most sampling intervals, tests for homogeneity of variance were often inapplicable (Levene’s Test, Minitab 11), rendering analysis of variance results questionable. Only cases in which homogeneity of variance tests were applicable and non-significant (i.e. variances of groups tested were relatively homogeneous) were subjected to further testing. Table 3.3 provides results of analysis of variance tests and means comparisons for female data when soybean-centered traps were compared. Three sampling intervals in early 1991 and four 46 in 1992 had differences among cropping systems that were significant (p<0.05). In five of ttiese cases the soybean monoculture was significantly different from all of the intercropping systems; in six of seven sampling intervals intercropping systems were not significantly different from one another. In 1991, immediately following the wheat harvest, there was an increase in numbers of female PLH recovered from soybean-centered traps in the intercropping systems (Table 3.2 and Fig. 3.1). The numbers were approximately equal in all three intercropping systems, approximately half that of the soybean monoculture. Numbers remained at about the same level for the soybean monoculture as in the week immediately preceding the wheat harvest, and were still somewhat greater than for the intercropping systems. Following the removal of wheat, females were usually found in all cropping systems (Table 3.1), and were widely distributed among individual plots, in every sampling interval until late August. They were found only occasionally thereafter. Numbers recovered from traps declined in all cropping systems beginning in late July through the end of sampling in October. Fifteen of twenty-one sampling intervals following the wheat harvest in 1991 had levels of female PLH sufficient for testing. Only in the sampling interval immediately after the wheat harvest was there a significant difference found among any of the cropping systems by a one way analysis of variance test, and in this case the means comparison test did not reveal any significant differences among means. wheat-centered traps. In contrast to the situation with soybean-centered traps in intercropping systems, females were much more likely to be present in wheat-centered traps in 1991, and slightly more likely in 1992 (Table 3.1, Fig. 3.1 vs. Fig. 3.2, and Fig. 3.3 vs. Fig. 3.4). In both years, the number of females recovered across all sampling intervals was intermediate between the extreme lows of the soybean-centered traps in the intercropping systems and the highs for 47 soybean-centered traps in the soybean monoculture (Table 3.2). In 1991, the cumulative numbers of females recovered from wheat-centered traps were 51.0, 14.7, and 34.0 times greater than from soybean-centered traps in the sam e period for the 3:3, 7:3, and 14:3 cropping systems, respectively. The cumulative numbers recovered from wheat-centered traps in the intercropping systems were 25.2% (3:3), 21.8% (7:3), and 16.8% (14:3) of that of the soybean- centered traps in the soybean monoculture. In 1992, the numbers of females recovered from the wheat-centered traps were 4.0, 1.7, and 1.5 times greater than for the soybean-centered traps in the 3:3, 7:3, and 14:3 cropping systems, respectively. Compared with the soybean- centered traps in the soybean monoculture, wheat-centered traps recovered only 6.2% (3:3), 3.1% (7:3), and 2.8% (14:3) as many females in the period preceding the wheat harvest. There was a trend toward greater numbers of females collected in wheat-centered traps as the amount of wheat in cropping systems decreased. This trend was not evident in the soybean-centered traps in the intercropping systems. There were ten of fourteen sampling intervals in 1991, and five of thirteen in 1992, that had sufficient female PLH distributions among intercropping systems to permit one-way analysis of variance testing of wheat-centered trap data. Nine sampling intervals in 1991 and three sampling intervals in 1992 had female PLH present in all three intercropping systems. One additional date in 1991 and one in 1992 had females co-present in two intercropping systems. One-way analyses of variance and means comparison tests applied to the data from these sampling intervals found no significant differences among cropping systems and no statistically significant separation among means. In another series of analyses, the data from the soybean monoculture were ignored and data from traps with both types of placement in the 3:3, 7:3, and 14:3 intercropping systems were submitted to two-way analysis of variance tests (ANOVA Procedure, Minitab 11). Once again, sampling intervals were tested separately. Once testing for homogeneity of variances 48 and normality of residuals was completed, for female PLH there were ten sampling intervals in 1991 and seven in 1992 available for further testing. Results from analysis of variance tests is presented in Table 3.4. The model tested used the (transformed) number of females per plot per day as the dependent variable, \with cropping system (three levels - 3:3, 7:3, and 14:3) and trap placement (two levels - soybean-centered traps and wheat-centered traps) as main effects (both fixed effects), and tested for a significant interaction effect of cropping system and placement of trap. In no case was there found to be a significant difference due to cropping system or interaction effect. All of the sampling intervals that showed significant differences due to placement of trap (seven sampling intervals in 1991, and two in 1992) had significantly greater numbers of females in wheat-centered traps than in soybean-centered traps (p<0.05). Of these, only four in 1991 and one in 1992 had females present in traps with both placements. Overall, there were four sampling intervals in 1991 and five in 1992 that had females copresent in traps with both placements. Males The patterns for male presence in sampling intervals (Table 3.1), cumulative numbers recovered in traps (Table 3.2), and trapping dynamics (Fig. 3.5 - 3.8) were similar to those of the females. soybean-centered traps. During the period preceding the wheat harvest, males were recovered in greater numbers, and over a much greater range of sampling intervals, in soybean- centered traps in the soybean monoculture than in the intercropping systems (Fig. 3.5 and 3.7). In 1991, males were present in 85.7% of all possible sampling intervals for the soybean 49 monoculture, compared with 33.3% for the 3:3 cropping system, 50.0% for the 7:3 system, and 16.7% for the 14:3 system; in 1992, they were present in 53.8% of sampling intervals for the monoculture, 7.7% for the 3:3 system, 15.4% for the 7:3 system, and 23.1% for the 14:3 system (Table 3.1). Cumulative numbers of males recovered in the intercropping systems were 1.4- 6.2% of the total for soybean monoculture in 1991 and 0.6-3.6% in 1992 (Table 3.2). As with the females, in the soybean-centered trap data, there did not appear to be a consistent correlation between males recovered or male presence in sampling intervals and the amount of wheat present Following the wheat harvest, as with the females, relative differences between soybean monoculture and intercropping systems in measurements of male abundance decreased, and these measurements remained similar among the intercropping systems (Fig. 3.5). The cumulative numbers of males recovered from the individual intercropping systems were 55.6- 60.6% of that from the soybean monoculture (Table 3.2). Males were present in 61.9%, 71.4%, 76.2%, and 80.6% of the possible sampling intervals for the 0:3, 3:3, 7:3, and 14:3 cropping systems, respectively (Table 3.1). Table 3.5 provides a summary of the results of one-way analysis of variance tests of male data for soybean-centered traps during the period preceding the wheat harvest. In 1991, there were seven of fourteen sampling intervals to which a one-way analysis of variance could be applied. Of these, five sampling intervals showed significant differences. Means comparison tests showed that the soybean monoculture was significantly different from each of the intercropping systems, but that the intercropping systems were not significantly different from one another. During 1992, there were five of thirteen sampling intervals eligible for analysis of variance; of these 4 had significant differences. In the period after the wheat harvest in 1991, sixteen of twenty-one sampling intervals were eligible for analysis of variance testing; only one had significant differences (not shown in Table 3.5). 50 wheat-centered traps.In the period preceding the wheat harvest, as with females, males were more likely to be present in wheat-centered traps than soybean-centered traps within the same intercropping systems, and there was little difference among the intercropping systems in this regard (Table 3.1). Males were present in 57.1%, 78.6%, and 64.3% of all possible sampling intervals for 1991, and 38.5%, 30.7%, and 38.5% of all intervals in 1992, for the 3:3, 7:3, and 14:3 cropping systems, respectively. The cumulative numbers of males recovered from the wheat-centered traps were 19.7% (3:3), 37.4% (7:3), and 35.6% (14:3) of the total for soybean monoculture during the same period in 1991, and 14.9% (3:3), 21.4% (7:3), and 17.9% (14:3) in 1992 (Table 3.2). W heat-centered traps within the sam e intercropping system s recovered 14.3 (3:3), 18.0 (7:3), and 5.7 (14:3) times the quantity of corresponding soybean-centered traps in 1991, and 6.3 (3:3), 36.0 (7:3), and 5.0 (14:3) times the quantity in 1992. A summary of analysis of variance and means comparison tests applied to the male data from the wheat-centered traps is provided in Table 3.6. Nine of the fourteen sampling intervals had sufficient males recovered from the wheat-centered traps to permit one-way analysis of variance tests to be applied. Two of the intervals showed significant differences, with the cropping system having the least amount of wheat, 3:3, found to be significantly different from the 7:3 and 14:3 intercropping systems. In 1992, there were six sampling intervals qualified for analysis of variance testing; none was determined to have significant differences among intercropping systems. In comparison, for females there were no sampling intervals with significant differences in recovery from wheat-centered traps. The results of the two-way analysis of variance, comparing placement of trap and cropping system (intercropping systems only), are shown in Table 3.7. In 1991, for eight sampling intervals for which a two-way analysis of variance was applicable, there was found to be a significantly more males recovered from wheat-centered traps than soybean-centered traps 51 in seven sampling intervals (p<0.01). Two of the seven sampling intervals also showed significant differences due to cropping system (p<0.01 ), and significant interaction effects of cropping system and trap placement (p<0.05). In these cases there was a significant inverse relationship between the number of males recovered from wheat-centered traps and the amount of wheat present, and there were few males recovered from soybean-centered traps. In 1992, analyses of variance were applicable to five sampling intervals; three of these showed significant variability due to trap placement. There were no significant differences found that year due to cropping system or interaction effect Figures 3.5-3.S also show a similar pattern in the comparative dynamics of male recovery in greater detail than the tables. The notable differences between females and males are that less males than females are recovered in early sampling inten/als, and the numbers of males recovered near the wheat harvest are greater than females. 52 Table 3.1 TOTAL NUMBER OF SAMPLING INTERVALS AND NUMBER OF SAMPLING INTERVALS WITH PLH PRESENT MEASUREMENT CROPPING FEMALES MALES SYSTEM 1991 1992 1991 1992 SOYBEAN-CENTERED TRAPS number of sampling intervals preceding wheat harvest 14 13 14 13 number of sampling intervals preceding wheat harvest in which PLH were recovered from plots 0:3 14 8 12 7 3:3 1 4 4 1 7:3 2 2 6 2 14:3 1 3 2 3 number of sampling intervals following wheat harvest 21 0 21 0 number of sampling intervals following wheat harvest during which PLH were recovered 0:3 14 NA 13 NA 3:3 14 NA 15 NA 7:3 12 NA 16 NA 14:3 14 NA 17 NA WHEAT-CENTERED TRAPS number of sampling intervals 14 13 14 13 number of sampling inten/als in which PLH were recovered from plots 3:3 12 5 8 5 7:3 11 4 11 4 14:3 10 6 9 5 Table 3.1. Measurements of the relative frequency of PLH recovery within soybean-centered and wheat-centered traps. The number of sampling intervals which had PLH present in traps can be compared among cropping systems and to the total possible number of intervals for which recovery might have occurred. Female and male data are presented separately. This table is calculated on a presence/absence basis for each cropping system; i.e. recovery of any PLH in a single trap (3 traps/plot, 3 plots for each cropping system) gives a positive score for the sampling interval. 53 Table 3.2 CUMULATIVE NUMBER OF PLH RECOVERED BEFORE AND AFTER THE WHEAT HARVEST 1991 1991 1992 1992 SEX OF TRAPPED CROPPING SOYBEAN- WHEAT- SOYBEAN- WHEAT- LEAFHOPPER, SYSTEM CENTERED CENTERED CENTERED CENTERED TIME PERIOD TRAPS TRAPS TRAPS TRAPS Females, prior to wheat harvest 0:3 202 NA 324 NA 3:3 1 51 5‘ 20 7:3 3 44 6 10 14:3 1 34 6 9 Females, after wheat harvest 0:3 111 NA NA NA 3:3 51 NA NA NA 7:3 55 NA NA NA 14:3 56 NA NA NA Males, prior to wheat harvest 0:3 289 NA 168 NA 3:3 4 57 4 25 7:3 6 108 1 36 14:3 18 103 6 30 Males, after wheat harvest 0:3 216 NA NA NA 3:3 123 NA NA NA 7:3 120 NA NA NA 14:3 131 NA NA NA Table 3.2. The cumulative number of PLH recovered over the series of sampling intervals before and after the wheat harvest. Separate sums are listed for sex. year, trap placement, and cropping system. NA- not applicable as cropping system was not sampled in the time period or with the trap placement referenced. 54 Table 3.3 ANALYSIS OF VARIANCE AND MEANS COMPARISON TEST RESULTS FOR FEMALE PLH IN SOYBEAN-CENTERED TRAPS, PRIOR TO WHEAT HARVEST SAMPLING INTERVAL ANALYSIS MEANS COMPARISON CROPPING OF TEST RESULT SYSTEMS WITH VARIANCE PLH PRESENT RESULT 27-31 May 1991 NS NS 0:3, 3:3 10-13 June 1991 P<0.001 0:3A, 7:3b, 3:3b, 14:3b 0:3, 7:3 27 June -1 July 1991 P<0.001 0:3A, 14:3b, 3:3b, 7:3b 0:3, 14:3 1-4 July 1991 P=0.009 0:3a, 7:3ab, 3:3b, 14:3b 0:3. 7:3 29 May - 2 June 1992 NS NS (7:3,14:3) 16-23 June 1992 P<0.001 0:3A, 14:38, 3:3c, 7:3c 0:3, 14:3 23-30 June 1992 NS NS 0:3, 3:3 4-7 July 1992 P=0.G16 0:3a, 7:3ab, 3:3b, 14:3b 0:3, 7:3, 3:3 7-10 July 1992 P<0.001 0:3A, 3:3b, 7:3b, 14:3b 0:3, 3:3 10-14 July 1992 P<0.001 0:3A, 14:3b, 3:3b, 7:3b 0:3, 14:3, 3:3 Table 3.3. Summary of results of one-way analysis of variance tests and means comparison tests for female data from soybean-centered traps preceding wheat harvest. MODEL; fixed- effect model with number of female PLH recovered per plot per day (transformed, see text) as the dependent variable, with cropping system as the single independent variable (4 levels). Sampling intervals were analyzed separately. Only sampling intervals which had PLH in more than one cropping system and produced non-significant values for Levene’s Test were submitted to analyses of variance and means comparisons. The P value indicates the probability that an equal or greater value of F (from one-way test) would occur by chance. For means comparison, means followed by the same letter were not significantly different from one another at least at the a=0.05 level. Upper-case letters indicate that the differences between the marked mean and all other means were significant at the a=0.01 level. NS = no significant differences. Cropping systems listed are ranked in order of descending value of means (left to right); parentheses indicate groups of equal means. 55 Table 3.4 RESULTS OF TWO-WAY ANALYSIS OF VARIANCE TESTS ON FEMALE PLH, 1991 CROPPING SYSTEMS WITH SAMPLING INTERVAL EFFECT TESTED ANOVA RESULT. FEMALE PLH PRESENT (by trap placement) 19-21 May 1991 cropping system NS 3:3w = 7:3w = 14:3w trap placement NS interaction NS 23-27 May 1991 cropping system NS 3:3w>7:3w> 14:3w trap placement P=0.014 interaction NS 27-31 May 1991 cropping system NS 3:3w > 14:3w > 7:3w >3:3s trap placement P<0.001 interaction NS 6-10 June 1991 cropping system NS 3:3w = 7:3w = 14:3w trap placement NS interaction NS 10-13 June 1991 cropping system NS 7:3w > 3;3w = 14;3w >7:3s trap placement P=0.004 interaction NS 13-17 June 1991 cropping system NS 14:3w > 3:3w = 7:3w trap placement P=0.002 interaction NS 17-20 June 1991 cropping system NS 3:3w > 7:3w > 14:3w trap placement P=0.024 interaction NS 20-24 June 1991 cropping system NS 14:3w > 7:3w > 3:3w trap placement NS interaction NS 27 June -1 July 1991 cropping system NS 3:3w > 7:3w > 14:3w > 14:3s trap placement P=0.013 interaction NS 1-4 July 1991 cropping system NS 7:3w > 3:3w = 14:3w > 7:3s trap placement P=0.026 interaction NS Table is continued on the following page. Table 3.4. Summary of results of two-way analyses of variance tests for female PLH data within specified sampling intervals. MODEL; fixed-effect model with number of female PLH recovered per day as the dependent variable (transformed, see text), with cropping system and trap placement as main effects, and with the interaction of cropping system and trap placement as the interaction effect. The P value indicates the probability that an equal or greater value of F (from two-way test) would occur by chance. NS = no significant differences were found for the effect indicated (a=0.05). The column at the far right lists the combinations of cropping system and trap placement that had any female PLH present during the sampling interval specified. Designations for the cropping systems (see text) are followed by s for soybean-centered traps or w for wheat- centered traps. The numerical relationships among means for the cropping systems listed are provided by greater than (>) or equal to (=) signs between pairs of means. Rankings do not imply significant differences among means. 56 Table 3.4 (continued) RESULTS OF TWO-WAY ANALYSIS OF VARIANCE TESTS, FEMALES, 1992 CROPPING SYSTEMS WITH SAMPLING INTERVAL EFFECT TESTED ANOVA RESULT, FEMALE PLH PRESENT (by trap placement) 29 May-2 June 1992 cropping system NS 7:3s = 14:3s trap placement NS interaction NS 16-23 June 1992 cropping system NS 14:3s > 3:3w = 7:3w = 14:3w trap placement NS interaction NS 23-30 June 1992 cropping system NS 3:3s > 14:3w trap placement NS interaction NS 30 June-4 July 1992 cropping system NS 3:3w> 7:3w = 14:3w trap placement P<0.001 interaction NS 4-7 July 1992 cropping system NS 7:3s = 3:3w > 3:3s = 14:3w trap placement NS interaction NS 7-10 July 1992 cropping system NS 3:3w > 7:3w > 14:3w > 3:3s trap placement P=0.050 interaction NS 10-14 July 1992 cropping system NS 7:3w > 14:3s > 3:3s = 14:3w trap placement NS interaction NS 57 Table 3.5 ANALYSIS OF VARIANCE AND MEANS COMPARISON TEST RESULTS FOR MEANS COMPARISON TEST RESULTS FOR MALE PLH IN SOYBEAN- CENTERED TRAPS PRIOR TO WHEAT HARVEST ANALYSIS CROPPING SAMPLING INTERVAL OF MEANS COMPARISON SYSTEMS WITH VARIANCE TEST RESULT PLH PRESENT RESULT 10-13 June 1991 P<0.001 0:3A, 3:3b, 7:3b, 14:3b 0:3, 3:3 13-17 June 1991 P<0.001 G:3A, 3:3b, 7:3b, 14:3b 0:3, (3:3, 7:3) 17-20 June 1991 P<0.001 Q:3A, 7:3b, 3:3b, 14:3b 0:3, 7:3 20-24 June 1991 NS NS 0:3, (7:3, 14:3) 24-27 June 1991 P<0.001 0:3A, 7:3b. 3:3b, 14:3b 0:3, 7:3 27 June - 1 July 1991 P=0.011 0:3a, 7:3b, 3:3b, 14:3b 0:3, 7:3, 3:3 1-4 July 1991 NS NS 0:3, 7:3, 14:3, 3:3 23-30 June 1992 NS NS 0:3, 14:3 30 June - 4 July 1992 P<0.001 0:3A, 14:3b, 7:3b, 3:3b 0:3, 14:3 4-7 July 1992 P=0.003 0:3A, 7:3b, 3:3b, 14:3b 0:3, 7:3 7-10 July 1992 P<0.001 0:3A, 7:3b, 3:3b, 14:3b 0:3, 7:3 10-14 July 1992 P<0.001 0:3A, 3:3b, 14:3b, 7:3b 0:3, (3:3, 14:3) Table 3.5. Summary of results of one-way analysis of variance tests and means comparison tests for male data from soybean-centered traps preceding the wheat harvest. MODEL: fixed- effect model with number of male PLH recovered per plot per day (transformed, see text) as the dependent variable, with cropping system as the single independent variable (4 levels). Sampling intervals were analyzed separately. Only sampling intervals which had PLH in more than one cropping system and produced non-significant values for Levene’s Test were submitted to analyses of variance and means comparisons. The P value indicates the probability that an equal or greater value of F (from one-way test) would occur by chance. For means comparison, means followed by the sam e letter were not significantly different from one another at least at the a=0.05 level. Upper-case letters indicate that the differences between the upper-case marked mean and all other means were significant at the a=0.01 level. NS = no significant differences. Cropping systems listed are ranked in order of descending value of means (left to right); parentheses indicate groups of equal means. 58 Table 3.6 ANALYSIS OF VARIANCE AND MEANS COMPARISON TEST RESULTS FOR MALE PLH IN WHEAT-CENTERED TRAPS ANALYSIS SAMPLING INTERVAL OF MEANS COMPARISON CROPPING SYSTEMS VARIANCE TEST RESULT WITH PLH PRESENT RESULT 31 May - 3 June 1991 NS NS 7:3. 3:3 6-10 June 1991 NS NS (7:3.14:3) 10-13 June 1991 P<0.015 3:3a. 7:3b. 14:3b 3:3. 7:3.14:3 13-17 June 1991 NS NS 3:3. 7:3,14:3 17-20 June 1991 NS NS 3:3. 7:3,14:3 20-24 June 1991 NS NS 3:3, 7:3,14:3 24-27 June 1991 P=0.022 3:3a. 7:3b, 14:3b 3:3, 7:3,14:3 27 June -1 July 1991 NS NS 3:3, 7:3,14:3 1-4 July 1991 NS NS 7:3, 14:3, 3:3 30 June - 4 July 1992 NS NS 14:3, 3:3, 7:3 4-7 July 1992 P=0.003 NS 3:3, (7:3,14:3) 7-10 July 1992 P<0.001 NS 7:3, 3:3,14:3 10-14 July 1992 P<0.001 NS 7:3. 3:3,14:3 Table 3.6. Summary of results of analysis of variance tests and means comparison tests for male data from wheat-centered traps. MODEL; fixed-effect model with number of male PLH recovered per plot per day (transformed, see text) as the dependent variable, with cropping system as the single independent variable (4 levels). Sampling intervals were analyzed separately. Only sampling intervals which had PLH in more than one cropping system and produced non significant values for Levene’s Test were submitted to analyses of variance and means comparisons. For m eans comparison, means followed by the sam e letter were not significantly different from one another at least at the a=0.05 level. Upper-case letters indicate that the differences between the upper-case marked mean and all other means were significant at the a=0.01 level. NS = no significant differences. Cropping systems listed are ranked in order of descending value of m eans (left to right); parentheses indicate groups of equal means. 59 Table 3.7 RESULTS OF TWO-WAY ANALYSIS OF VARIANCE TESTS FOR MALE PLH ANOVA CROPPING SYSTEMS WITH SAMPLING INTERVAL EFFECT TESTED RESULT FEMALE PLH PRESENT (by trap placement) 31 May - 3 June 1991 cropping system NS 7:3w > 3:3w trap placement NS interaction NS 10-13 June 1991 cropping system P=0.003 3:3w > 7:3w > 14:3w = 3:3s trap placement P=0.001 interaction P=0.014 13-17 June 1991 cropping system NS 3:3w > 7:3w > 14:3w > 3:3s = 7:3s trap placement P=0.009 interaction NS 17-20 June 1991 cropping system NS 3:3w > 7:3w > 14:3w > 7:3s trap placement P=0.009 Interaction NS 20-24 June 1991 cropping system NS 3:3w > 7:3w > 14:3w > 7:3s = 14:3s trap placement P=0.001 interaction NS 24-27 June 1991 cropping system P=0.009 3:3w > 7:3w > 14:3w > 7:3s trap placement P<0.001 interaction P=0.011 27 June -1 July 1991 cropping system NS 3:3w > 7:3w > 14:3w > 7:3s > 3:3s trap placement P<0.001 interaction NS 1-4 July 1991 cropping system NS 7:3w > 14:3w > 3:3w > 7:3s > 14:3s trap placement P=0.002 >3:3s interaction NS Table is continued on the following page. Table 3.7. Summary of results of two-way analyses of variance tests for male PLH data within specified sampling intervals. MODEL; fixed-effect model with number of male PLH recovered per plot per day (transformed, see text) as the dependent variable, with cropping system and trap placement as main effects, and with the interaction of cropping system and trap placement as the interaction effect. The P value indicates the probability that an equal or greater value of F (from two-way test) would occur by chance. NS = no significant differences were found for the effect indicated (a=0.05). The column at the far right lists the combinations of cropping system and trap placement that had any female PLH present during the sampling interval specified. Designations for the cropping systems (see text) are followed by s for soybean-centered traps or w for wheat- centered traps. The numerical relationships among means for the cropping systems listed are provided by greater than (>) or equal to (=) signs between pairs of means. Rankings do not imply significant differences among means. 60 Table 3.7 (continued) RESULTS OF TWO-WAY ANALYSIS OF VARIANCE TESTS FOR MALES ANOVA CROPPING SYSTEMS WITH SAMPLING INTERVAL EFFECT TESTED RESULT FEMALE PLH PRESENT (by trap placement) 23-30 June 1992 cropping system NS 3:3w> 14:3s trap placement NS interaction NS 30 June - 4 July 1992 cropping system NS 14:3w > 3:3w > 7:3w >14:3s trap placement P=0.001 interaction NS 4-7 July 1992 cropping system NS 3:3w > 7:3w = 14:3w > 7:3s trap placement P=0.007 interaction NS 7-10 July 1992 cropping system NS 7:3w > 3:3w > 14:3w = 7:3s trap placement P=0.010 interaction NS 10-14 July 1992 cropping system NS 7:3w > 3:3w = 14:3w = 3:3s = 14:3s trap placement NS interaction NS 61 FEMALE PLH IN SOYBEAN-CENTERED TRAPS, 1991 Fig. 3.1. Comparison of female PLH recovered from soybean-centered traps In 4 cropping systems during 1991. The height of the ribbon at each articulation point indicates the mean of females recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot). The break in ribbons and date-axis labelling shows the period when sampling was interrupted by the wheat harvest. FEMALE PLH IN WHEAT-CENTERED TRAPS, 1991 Fig. 3.2. Comparison of female PLH recovered from wheat-centered traps In 3 cropping systems during 1991. The height of the ribbon at each articulation point Indicates the mean of females recovered per plot per day of the sampling Interval (3 plots per cropping system, 3 traps per plot). FEMALE PLH IN SOYBEAN-CENTERED TRAPS, 1992 2 'VVT'. Fig. 3.3. Comparison of female PLH recovered from soybean-centered traps in 4 cropping systems during 1992. The height of the ribbon at each articulation point indicates the mean of females recovered per plot per day of the sampling inten/al (3 plots per cropping system, 3 traps per plot). FEMALE PLH IN WHEAT-CENTERED TRAPS, 1992 .5 .0 a Fig. 3.4. Comparison of female PLH recovered from wheat-centered traps In 3 cropping systems during 1992. The height of the ribbon at each articulation point indicates the mean of females recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot). MALE PLH IN SOYBEAN-CENTERED TRAPS, 1991 g Fig. 3.5. Comparison of male PLH recovered from soybean-centered traps in 4 cropping systems during 1991. The height of the ribbon at each articulation point indicates the mean of males recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot). The break in ribbons and date-axis labelling shows the period when sampling was interrupted by the wheat harvest. MALES IN WHEAT-CENTERED TRAPS, 1991 5.5 ■...... i...... 5.0 ■....r.... i..... i 1 4.5 1 4.0 .....i.....!' S 3.5 ■ ...... : ; E 3.0 ; ...... m - (fl 2.5 r 2.0 ..... 3 1.5 ...... i..... r 2 10 .....:.....i....t ^ 0.5 ....k - ' O) N o « ^ ' Fig. 3.6. Comparison of male PLH recovered from wheat-centered traps in 3 cropping systems during 1991. The height of the ribbon at each articulation point indicates the mean of males recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot). MALE PLH IN SOYBEAN-CENTERED TRAPS, 1992 J.. r ; . r ' : ... ' -i : . r:..! T.. - i iV : i J 4 ■■■■ " i -i ...i 4- ! i.. ;■ ■■■■ g Fig. 3.7. Comparison of male PLH recovered from soybean-centered traps in 4 cropping systems during 1992. The height of the ribbon at each articulation point indicates the mean of males recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot). MALE PLH IN WHEAT-CENTERED TRAPS, 1992 11 : , ; ! I j - ' i- ■ - :k ik . i" : ■ i ■ j . 1. : l: I' 11 L i rf ■ t 1.1 ,.i. ! ■ki' k Hi: ■' ' ;. i i ■■■■ i'.:. k- ik ik 1 1 1 ' ! I'i 1 ■■ ■ ; r iJIIkk g Fig. 3.8. Comparison of male PLH recovered from wheat-centered traps in 3 cropping systems during 1992. The height of the ribbon at each articulation point indicates the mean of males recovered per plot per day of the sampling interval (3 plots per cropping system, 3 traps per plot). DISCUSSION This study found evidence that the presence of wheat in the intercropping systems was highly effective in inhibiting soybean colonization by PLH. Large differences were found between the soybean monoculture and all of the intercropping systems in the cumulative numbers of PLH recovered from soybean-centered traps in the period preceding the wheat harvest. For sampling intervals with statistically testable datasets (either female or male data), the majority had significantly greater numbers of PLH in the soybean monoculture than in intercropping systems. Over the two years of the study, cumulative numbers in soybean- centered traps in the intercropping systems were only 0.5-1.9% (females) and G.6-6.2% (males) of levels found in the soybean monoculture. Once substantial migration was initiated each year, PLH were found in almost every sampling interval for the soybean monoculture, but few sampling intervals had any PLH detected in traps located at the soybean canopy level (soybean-centered traps) in the intercropping systems. Among the intercropping systems, the cumulative numbers of PLH recovered prior to the wheat harvest, and the number of sampling intervals which had any PLH, were not proportionate to the amount of wheat present. Both these variables were approximately the sam e for all three intercropping systems. This is a surprising conclusion, in light of the relatively great differences in the amount and spacing of wheat among the intercropping systems. Considering data for both males and females, over two years there was only one sampling interval in the period preceding the wheat harvest with a significant difference among the intercropping systems. PLH did not fail to make contact with the intercropping systems entirely. Considerable numbers of both male and female PLH were recovered from wheat-centered traps, especially in 1991. In 1992 the relatively late onset of migration in mid-June meant that soybean plants in all 70 cropping systems were taller and had more leaf area than was the case when PLH migration started in 1991. Therefore, the soybean plants in the intercropping systems may have been somewhat more apparent to searching leafhoppers than in the previous year. The discrepancies between the traps with different placement in cumulative numbers and presence in sampling intervals was not as great as 1991. In both years for both sexes, the cumulative numbers recovered from the wheat-centered traps were less than those of the soybean- centered traps in the soybean monoculture, but always surpassed those of nearby soybean- centered traps within the same cropping system. Within intercropping systems, more sampling intervals had PLH present in wheat- centered traps than in soybean-centered traps. In 1991 within any sampling interval, when there were any females at all present in soybean-centered traps, the mean for soybean-centered traps in any intercropping system was never as great as the smallest mean for wheat-centered traps. In that year, female recovery was significantly greater in wheat-centered traps than soybean- centered traps in all seven of the sampling intervals which had significant differences. In 1992, this dominance was less complete. There were sometimes more females recovered from soybean-centered traps than wheat-centered traps in intercropping systems, but in those sampling intervals the differences were not significant. In those cases where a statistical difference due to trap placement was found, means for wheat-centered traps exceeded their soybean-centered trap counterparts in every cropping system. For males, seven of eight intervals tested showed significantly more PLH in wheat-centered traps than soybean-centered traps in 1991, and three of five in 1992; other intervals had no significant differences. During both years, there was a trend toward greater cumulative numbers of females recovered from wheat-centered traps as the amount of wheat decreased. This relationship was not found for males, nor for PLH of either sex recovered from the soybean-centered traps in the intercropping systems. For individual sampling intervals, there were usually more females and 71 males found in wheat-centered traps in the 3:3 cropping system than in the other intercropping systems. However, in one-way analysis of variance tests on individual sampling intervals, there were no significant differences found among cropping systems for females recovered from the wheat-centered traps. Males were significantly more abundant in the 3:3 cropping system than in other intercropping systems in two of nine sampling intervals for which analysis of variance tests were made. There is some evidence that PLH may prefer a more open plant canopy to one that is closely planted. Mayse (1978) reported that PLH attained higher densities (per unit ground area, per plant, and per unit leaf area) on soybean grown with wide row spacing than with narrow row spacing. Smith (1987) found that when the number of alfalfa stem cuttings within test cages was equal between groups compared, PLH preferred to oviposit in stems that were more evenly dispersed over those that were clumped. In a study of com leafhopper movement and plant dispersion. Power (1992), found that among several stands of equal host plants density, Dalbulus maidis DeLong & Wolcott preferred com in spatial arrangements that had the greatest interplant distance over other arrangements with greater clumping of hosts. In my study, female PLH may have initially shown a greater attraction to the openness of the 3:3 cropping system, but this feiled to translate into effective colonization, as judged by the low level of soybean-centered trap catches. Immediately following the removal of wheat in 1991, there were substantial increases in both male and female recovery from the soybean-centered traps in all of the intercropping systems. The numbers trapped in the soybean monoculture remained higher for several weeks, after which the number captured became approximately equal among all cropping systems. The fact that the differences among the cropping systems, which were so great immediately prior to the wheat harvest, diminished so rapidly is indicative of the magnitude of the shielding effect provided to soybean in the intercropping systems by the wheat. 72 As the season progresses there is a shift in the source of PLH recovered from traps, from long-distance migrants to PLH engaged in shorter-distance movement. This occurs as the conditions that ^cilitate long-distance migration (source populations and weather conditions) are reduced (usually by the end of June), and local abundance of PLH increases. Following the wheat harvest recovery of PLH from traps is believed to represent mostly short-distance movement and local redistribution of populations. I had relatively little interest in these phenomena, so while recovery from soybean-centered traps was monitored following the wheat harvest in 1991, it was not measured during 1992. Because of the substantial and significant differences between the soybean monoculture and intercropping systems in soybean canopy colonization, the null hypothesis was rejected. However, as was the case with the abundance study, the altemative hypothesis that was based upon the resource concentration hypothesis could not be accepted. This predicted that a gradient of host colonization behavior would exist, in opposition to the gradient of vegetational diversity. Across all cropping systems examined, the colonization response of PLH is poorly related to this gradient, and it is independent of the gradient for the intercropping plots during the critical early part of the season. Therefore the resource concentration hypothesis cannot be said to be well-supported by the evidence of this study. Considering habitats with hosts only (monoculture) versus any or all mixtures of hosts and non-hosts (intercropping systems), the herbivore colonized the less vegetationally diverse habitats more abundantly. Over the wide range of vegetational diversity in mixtures, there was no difference in colonization. As is the common approach with many intercropping experiments, the host plant seeding density and spatial arrangement was the same for all cropping systems. This study differed from most other research on PLH and vegetational diversity (e.g. Lamp ef a/. 1984a, 73 Hammond & Jeffers 1990) in that it used four levels of host plant concentration that were quantitative differences in only two components (soybean and wheat). Inevitably, there were differences in host quantity (e.g. reduced leaf area - Appendix D and Hammond & Jeffers 1990) and probably quality (e.g. reduced nitrogen - Roltsch & Gage 1990b, and water stress) resulting from competition between intercropped species and modification of the environment (e.g. reduced light intensity - Appendix C). While specific factors of host quality or quantity to which PLH might respond are not well known, some evidence exists that such factors can influence leafhopper abundance (Roltsch & Gage 1990ab). Differences in host quality and quantity in this study were believed to follow the gradient defined by vegetational diversity (i.e. more wheat decreases soybean quality and quantity). The effect of differential host quality and quantity on PLH was expected to act in the same direction as presence of the non-host along the gradient of vegetational diversity. Ideally, it would be desirable to separate any effect that differential host quality and quantity has on PLH from the effect of non-host interference, in order to better understand the specific factors contributing to PLH suppression. The cropping system with the lowest amount of wheat, 3:3, was believed to be similar in host quality and quantity to the monoculture, so that the difference in PLH colonization and density (Chapter 2) between these cropping system s was expected to be primarily due to the presence of the non-host. In 1991 across all replications, in early July soybean leaf area in the 3:3 cropping system was 82.9% of that in the monoculture, and the mean leaf areas for two of the three blocks were 87.6% and 96.9% of monoculture plots in those blocks (Appendix D, Table D.3). Colonization of soybean in the 3:3 cropping system was about the same as in the 14:3 cropping system, which had only 32.1%, 22.6%, and 57.8% (for three replications) as much leaf area as the 3:3 cropping system (Appendix D, Table D.3). Over the range of the intercropping systems, there was no indication that PLH responded to differential host quality and quantity, because colonization of these cropping system s was similar, and at such a low level prior to the wheat harvest. Following the 74 wheat harvest, when the presence of wheat was probably less important, differential host quality and quantity still did not appear to be an important factor influencing colonization, because there were few significant differences detected among the cropping systems. The process of host selection by herbivores can be subdivided as a series of complex interactions of environmental stimuli with the internal states and sensory capabilities of herbivores. Thorsteinson (1960) and Schoonhôven (1968) have described these as habitat- finding, host-finding, host recognition, host acceptance, and host suitability. This line of analysis predicts that monophagous herbivores of necessity develop host selection mechanisms specific to their narrow range of hosts, which may be spatially and temporally unpredictable; they have the sensory capability to respond to host location cues over long distances, or to overcome problems presented by interfering stimuli ("noise"). Monophagous herbivores tend to be host- finding specialists. Polyphagous herbivores have hosts that are, in aggregate, abundant and easily located. Highly mobile, polyphagous herbivores such as PLH are unlikely to have developed specialized sensory capabilities and behaviors required to find specific hosts at a distance. Herbivores in this group are predicted to place greater emphasis on the host acceptance step in the host selection process (Jermy 1971, Lance 1983, Lamp & Zhao 1993). Host acceptance requires stimuli received through direct contact with plants. Lamp & Zhao (1993) have suggested that herbivores that emphasize this step in the host selection process are amenable to interference through habitat manipulations (such as intercropping hosts with non-hosts). These can be successful in reducing colonization of crop habitats if they reduce the likelihood of direct contact with host plants. They exploit the polyphagous herbivore’s relatively poorly developed capability to distinguish hosts at a distance, without direct contact. 75 This study presented evidence that differential colonization was a general mechanism that led to differences of PLH density among cropping systems. Flanders & Radcliffe (1989), studying potato and snap bean, and Flinn etal. (1990b), studying alfalfe, likewise showed that the dynamics of movement of PLH into the crop habitat from external sources was of primary importance in understanding within-crop density and damage relationships of PLH, more important than population increase within crop fields. Research by Lamp & Zhao (1993) also supported that conclusion. They determined that colonization of allalfe intercropped with oat (four cropping systems with different seeding densities of oat) was inversely proportional to the amount of oat in the mixture. The specific fectors associated with differential colonization are not known. Other research on these same cropping systems (Appendix C) found that soybean in the intercropping systems were subjected to increased amounts of shading as the amount of wheat present increased (see also All 1990). The reduced light intensity effecting host plants in polycultures may make the habitats less suitable for herbivore colonization or may render host plants more difficult to find (Lamp 1991). Lamp (reported in NC-193 Technical Committee 1991) also reported that when light intensity was reduced, PLH fight activity increased (negative photokinesis). Smith (1987) and Smith etal. (1992) concluded that olfactory stimuli from a grass were at least partially responsible for reduced PLH density in alfal^ when grasses were present. In the preference tests he conducted within cages, extracts from crabgrass (Digitaria sanguinalis L.) applied directly to alfalfe or in close proximity to it (outside of cages, excluding the possibility of contact) deterred oviposition and reduced residency of PLH. relative to untreated alfalfe or alfelfe further away from crabgrass extracts. Altieri et al. (1977) believed that a volatile chemical from grasses was responsible for reducing densities of a related leafhopper, Empoasca kraemeri Ross and Moore, when grasses were growing with beans. Smith (1987) and Smith et al. (1992) showed that the presence of grass with alfelfe stimulated PLH flight 76 behavior, compared with alfalfa alone. In the field this response might be expected to inhibit colonization of habitats containing grass or promote relocation of PLH out of grassy habitats. PLH appears to have a greater rate of population increase on soybean during May and June because weather conditions are closer to optimal and the host plant is undergoing rapid expansion in leaf area. In mid-late summer, higher temperature, lower humidity, and reduced vegetative growth may negatively influence PLH survival, growth, and reproduction (Delong 1938, Kieckheffer & Medler 1964, Kouskolekas & Decker 1966, Decker ef a/. 1971, Taylor 1989). Therefore, delay of soybean colonization will probably greatly limit the ability of this insect to realize its reproductive potential by denying it the crop resource during the most favorable time of the year. 77 CHAPTER 4 SUMMARY AND DISCUSSION Review of Experimental Results Abundance Study In the first study presented (Chapter 2), the measurement of population density found that once substantial spring migration into the area was underway, PLH were significantly more abundant in the soybean monoculture than in any of the intercropping systems. Mean densities were extremely low in the intercropping systems prior to the wheat harvest. The disparities between the soybean monoculture and intercropping systems gradually decreased after the harvest, and fewer collection dates showed significant differences. However, PLH densities continued to be substantially greater in the soybean monoculture than in the other systems until early to mid-August of each year. Males were less abundant than females for most of the sample collection period, but otherwise showed similar population trends and differences among cropping systems. There were no significant differences in density detected among the intercropping systems. Colonization Study While the first study determined abundance of PLH, the second study provided a measurement of behavior involved with habitat colonization. The results showed that the presence of wheat in the intercropping systems strongly inhibited PLH from making contact with the soybean canopy level (i.e. soybean-centered traps) until after the wheat harvest. During the 78 early portion of the year, cumulative numbers of PLH intercepted within the soybean canopy in the intercropping systems were only 0.5-1.9% (females) and G.6-6.2% (males) as great as those in the soybean monoculture over two years of the study. Once substantial migration was initiated each year, PLH were found in almost every sampling interval for the soybean monoculture, but few had any PLH detected in the intercropping systems at the soybean canopy level. The great majority of sampling intervals tested had significantly more PLH detected in soybean-centered traps in the soybean monoculture than for equivalent traps in intercropping systems. While colonization of the soybean canopy overwhelmingly favored the soybean monoculture over the intercropping systems, there were essentially no differences among the intercropping systems in PLH colonization of the soybean canopy. Among both males and females, over the two years of the study, only a single sampling interval in the period preceding the wheat harvest showed a significant difference among the intercropping systems. Considerable numbers of both male and female PLH were recovered from traps located in wheat sections of the intercropping systems at canopy level, especially in 1991. Means from traps located within the wheat canopy were always greater (usually significantly greater) than for corresponding traps located within the soybean canopy of the same cropping system. They were always less than for traps located in the soybean monoculture. Immediately following the removal of wheat in 1991, there were substantial increases in both male and female recovery from the soybean-centered traps in all of the intercropping systems. The numbers trapped in the soybean monoculture remained higher for several weeks, after which the number captured became approximately equal among all cropping systems. The fact that the differences among the cropping systems, which were so great immediately prior to the wheat harvest, diminished so rapidly is indicative of the magnitude of the shielding effect provided to soybean in the intercropping systems by the wheat. 79 Analysis of Research Objectives and Hypotheses Tested The purpose of this research was to investigate the effects that vegetational diversity has on PLH, with emphasis on identification of mechanisms contributing to the pattern of population suppression that has been commonly reported. This was pursued through two experiments that examined whether there were differential responses of PLH to an experimentally-produced gradient of vegetational diversity. In the case of the first experiment (Chapter 2), the response variable measured was abundance; in the second experiment (Chapter 3) the response was colonization of host plants. The first study quantified density while the second measured a behavior important in contributing to density. The two studies tested similar null and alternative hypotheses. The null hypothesis in each case stated that there were no differences in the response variable among the cropping systems. The alternative hypothesis, which in both studies was based upon the resource concentration hypothesis, stated that the response variable, abundance or colonization, would be inversely proportionate to the amount of wheat present. The null hypothesis was rejected in both studies because of the large difference in the response variable that was found when the soybean monoculture was compared with the intercropping systems as a group. The alternative hypothesis was also rejected because the response variable was independent of the gradient of vegetational diversity among the intercropping systems. Had the soybean monoculture not been included with the intercropping systems in the experimental design, results of both studies would have supported the null hypothesis. If only a single version of a relay intercropping system (even that with the lowest amount of wheat) had been included in experiments to compare with the monoculture, then the alternative hypothesis would have been supported in both cases. As such the resource concentration hypothesis can neither be fully validated nor fully rejected based on this research. 80 it remains the most appropriate explanatory hypothesis for the current research because of its emphasis on movement behavior of herbivores and its interaction with abundance. Implications of Results Resource Concentration Hypothesis. The resource concentration hypothesis (Root 1973) states that the abundance and distribution of herbivores is in large part determined by their behavior in response to the combined stimuli received from plants in habitats. Three major effects interact to determine the herbivore's evaluation of the suitability or attractiveness of a habitat: 1) the quantity of host species present, and the relative preference for each species; 2) the absolute density and spatial arrangement of each host species; and 3) inteference from non-host species (Risch 1981). Over the range of vegetational diversity presented in the cropping systems of this experiment, PLH appeared to have a qualitative response, ratiier than a quantitative response, to the presence of the non-host. By this I mean that the leafhopper did not discriminate among cropping systems that had different relative amounts of soybean and wheat, but responded similarly to all cropping systems that had any wheat, at least over the range of non-host densities presented. PLH is a highly polyphagous herbivore, believed to be lacking in specific adaptations to host finding that typify true diet-specialized herbivores (Thorsteinson 1960, SchoonhOven 1968, Jermy 1971, Lance 1983, Lamp & Zhao 1993). However, within the context of the cropping systems used for the current research, PLH functions as a diet-specialized herbivore (Andow 1983a, 1991), that is, it can exploit only one of the plant species in the habitat as a host. In searching for new habitats the leafhopper may be unable to accurately assess host quantity, preference, density, and arrangement (effects 1 and 2 above) in the intercropping 81 systems because the interference effect of the non-host (effect 3 above) is dominant, in bicultures with crops arranged in rows, potential colonists may be most likely to encounter the plant canopy which has the greatest height In the case of the soybean-wheat relay intercropping systems, wheat is taller until after the wheat harvest If the behavioral response of host-searching PLH to landing on wheat is to take flight (Flinn et al. 1990b, Coggins 1991) and they poorly recognize nearby host plants within the same habitat, then we would expect to detect some contacts of PLH with the wheat canopy, but to detect far fewer contacts with the soybean canopy (lower height). This is the pattern observed in the colonization study. Significantly less colonization of soybeans in all the intercropping systems than in monoculture translates to significantly less density of PLH in those systems. That is the pattern observed in the abundance study. Immigration vs. Emigration The results of the current research also have implications concerning the relative importance of immigration and emigration as processes affecting abundance. Root (1973) considered both processes to make important contributions to determining density in relation to vegetational diversity. Some experimental evidence has indicated that emigration of PLH from habitats as a response to contact with non-hosts may be relatively important in cropping systems with alfalfe (Smith 1987, Coggins 1991). In the colonization experiment, if traps located within the soybean canopy detected that PLH had made contact there, but abundance sampling failed to find the insect subsequently, that would suggest that immigration had occurred, but that migrants had subsequently emigrated or otherwise had been removed (possibly by predation). The feilure to find more than negligible numbers of PLH in soybean- centered traps (during intervals when contact was made with the wheat canopy), together with 82 the very low densities recorded through abundance sampling, indicates that immigration was a more important process in the current studies. Specific Mechanisms Underlying Responses to Vegetational Diversity Differential habitat colonization was found to be a general mechanism that appeared to be capable of accounting for the abundance differences reported in this dissertation. This is in agreement with work by others who concluded that the dynamics of movement of PLH into the crop habitat from extemal sources was of primary importance in understanding within-crop density and damage relationships of PLH (Flanders & Radcliffe1989, Flinn etal. 1990b, Lamp & Zhao 1993). Specific mechanisms that have been suggested or investigated as having an influence upon colonization are differential shading in habitats (Smith 1987, Ali 1990, Lamp 1991, NC-193 Technical Committee 1991, Appendix 0). and chemical or contact stimuli from non-hosts (Smith 1987, Smith etal. 1992) (Altieri etal. 1977, with Empoasca kraemeri). Host Plant Quality and Quantity Factors involved in evaluation of host plant quality and quantity by PLH are not well understood, however there is some evidence that PLH adults respond to factors such as nitrogen content of hosts (Roltsch & Gage 1990b) and stress related to plant injury (Hammond & Jeffers 1990). R esponses reported include differential host exploitation (feeding & oviposition - Roltsch & Gage 1990b) and immigration/emigration (Hammond & Jeffers 1990). Across the gradient of vegetational diversity represented by the cropping systems of the current experiments, host plant phenology, height, leaf area, and biomass (Appendix D) were altered by plant competition and probably microhabitat modification related to plant density (e.g. reduced light intensity - Appendix C). In general, leaf area of soybean decreased with increased amounts of wheat in cropping systems. For example, in 1991 mean leaf areas in the 83 intercropping systems as a percentage of that in the soybean monoculture were 82.9% (3:3), 63.3% (7:3), and 28.9% (14:3). However, in some replications (blocks), mean leaf area of intercropping systems with low (3:3) or intermediate (7:3) amounts of wheat was approximately equal to or greater than in monoculture. Height of soybean tended to be greater in intercropping systems than in monoculture, probably related to reduced light intensity and protection from wind, factors that promote etiolation. Among the intercropping systems, the height rankings were not consistent, varying with replication and year. Vegetative stage development of soybean (Fehretal. 1971) was inhibited by the presence of wheat, and biomass accumulation reduced (Appendix D). However, the results of the two experiments did not suggest that colonization and abundance were closely related to the quality and quantity differences of host plants. There were no significant differences in PLH response variables among the intercropping systems while host characteristics, especially leaf area, varied considerably. If the interference effect due to wheat has an overriding influence upon colonization, then host quality and quantity may be relatively unimportant factors affecting abundance during the early portion of the year, as potential colonists may rarely make physical contact with hosts, nor even closely approach them. When PLH density increased in intercropping systems immediately following the wheat harvest, PLH still did not assume a pattem of abundance that was correlated with host quality-quantity effects, and soybean canopy colonization remained approximately the same among the intercropping systems. 84 Future Directions The soybean-wheat relay intercropping system is an interesting and useful platform for studying a variety of ecological phenomena such as crop interactions, insect ecology, and cropping system design. One version of the intercropping systems that was used for these studies, the “full relay intercropping” system, is a commercially viable altemative cropping system (if not widely accepted by growers). Because of the large area devoted to soybean in many parts of the east and Midwest, the crop serves as a nursery for PLH which may then go on to damage other, more susceptible crops. If a soybean-wheat relay intercropping system were substituted for monocultured soybean on a large scale, even extensively within a local area, it seems likely that PLH abundance and damage relationships could be greatly reduced. A question arises in connection with the matter of scale. In these experiments, plot size w as relatively small, relative to the leafhopperis ability to move. Migrant PLH that contacted intercropping systems and were diverted away could travel to nearby a soybean monoculture. If soybean-wheat intercropping systems or other systems (especially those with a low density of non-hosts) designed to inhibit PLH colonization were planted over a large area, would they remain effective? With a restricted ability to behaviorally express preference, females might show a greater inclination to colonize habitats with grass. Studies that examined development of PLH immatures in crop mixtures with grasses have indicated that nymphs were not greatly effected by vegetationally diverse habitats (Lamp etal. 1984a, Roltsch & Gage 1990a). The experiments in this dissertation determined that the intercropping system with the least amount of wheat (3:3), suppressed the PLH population at the same level as the intercropping systems with more wheat (7:3,14:3). It would be interesting and useful to know 85 whether cropping systems with even less than one wheat row per soybean row wheat than 3:3 be equally effective, or at a lower but acceptable level of effectiveness? If the gradient of resource concentration w as shifted in the direction of lower amounts of non-host, would graduated responses of density and colonization then become measurable? Pursuit of such a line of inquiry might yield especially valuable information if the intent w as to find optimal crop mixtures combining associational resistance with minimal yield loss due to plant competition. It should be worthwhile to pursue additional research oriented toward teaming more about the the specific behaviors of PLH that influence it to react to vegetational diversity. For example, it would be interesting to know whether the success of wheat in shielding soybean from colonization was due to some biologicai property or results from its effectiveness as a barrier. If a dead plant barrier or an artificial barrier might be effective in inhibiting PLH abundance, then the desirable properties of such might be worthy of investigation. Given the reports that a variety of different grasses and non-grasses have contributed to suppressing PLH abundance (e.g. Lamp etal. 1984a, Smith etal. 1988, Roltsch & Gage 1990ab, Lamp 1991), a biological property is unlikely to be specific to wheat. The responses of PLH to non-host stimuli would seem to warrant additional careful investigation. One approach that has been used with some success is to employ large cages, arranged in pairs within the greenhouse or field, and to create spatial arrangements of potted plants that test the importance of specific factors thought to influence PLH. In this way factors that are difficult to separate under field conditions, can be teased apart. For example, the problem of correlations between effects of vegetational diversity and effects of host plant quality and quantity can be reduced by using potted host plants, closely matched in terms of quality and quantity, and arranged in identical arrays on either side of a large rectangular cage. By superimposing an array or non-hosts among host plants on only one side of the cage pair. 86 habitats that differ by a single ^cto r can be created. By introducing PLH adults into the cages so that freedom of movement between the two habitats is possible, preference for vegetational diversity can be studied in isolation from other complicating factors (TJM unpublished). This research contributed knowledge about the ecology of a specific, important economic pest that might have application in other situations. It contributed incrementally to the development of the general knowledge bases about herbivore-habitat interactions and intercropping. 87 APPENDIX A SEX RATIOS OF POTATO LEAFHOPPERS FROM TWO FIELD STUDIES Table A.1 Sex Ratio of Potato Leafhoppers Captured During the Abundance Study DATE mean per sample, mean per sam ple, mean per sample, se x ratio m ales females total 04.06.1991 0.0556 0.5833 0.6389 0.9130 14.06.1991 0.6571 4.0286 4.6857 0.8599 20.06.1991 1.1667 2.8333 4.0000 0.7083 27.06.1991 2.0833 2.2778 4.3611 0.5223 17.07.1991 3.3333 8.1667 11.5000 0.7101 26.07.1991 3.8056 5.6111 9.4167 0.5959 10.08.1991 6.0000 7.0833 13.0833 0.5414 28.08.1991 1.1944 1.6667 2.8611 0.5825 DATE mean per sample, mean per sample, m ean per sam ple, se x ratio m ales females total 15.06.1992 0.0417 0.3542 0.3958 0.8947 25.06.1992 0.2500 6.8125 7.0625 0.9646 02.07.1992 0.2708 3.1875 3.4583 0.9217 09.07.1992 1.7500 5.1818 6.9318 0.7475 16.07.1992 3.0909 4.5758 7.6667 0.5968 23.07.1992 14.7609 17.4348 32.1957 0.5415 01.08.1992 17.1842 19.6579 36.8421 0.5336 11.08.1992 11.3415 16.8293 28.1707 0.5974 19.08.1992 6.1034 8.3448 14.4483 0.5776 04.09.1992 0.8947 0.8421 1.7368 0.4848 14.09.1992 0.5435 0.5435 1.0870 0.5000 Table A.I. Sex ratio of potato leafhoppers captured during the abundance study. The grand mean (all cropping system s) for each sample collection date is shown for males, females, and total potato leafhoppers (males + females). The sex ratio is expressed as the proportion of total potato leafhoppers that were female. A small number of potato leafhoppers were not sexed due to specimen damage, these were excluded from the calculations. 88 SEX RATIO OF POTATO LEAFHOPPERS RECOVERED FROM WATER PAN TRAPS, 1991 0 0 iS (O co T- 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 O) CJ) o i o> C c c 3 3 3 3 C3> C3) C3) CJ> 6 ) 6 . & 6 . & k 3 3 3 3 3 3 3 3 0) 0) 0) 0) f 7 2 2 6 ;z < : <Ç < : < f < f 9 9 9 9 2 2 CM CM CO o M" ■M- 00 i I 1 1 CM T— CM CM COLLECTION DATE Fig.A.1. Sex ratio of potato leafhoppers recovered from water pan traps, 1991. Sex ratio was expressed as the proportion of total potato leafhoppers recovered that were female. A small number of specimens were not sexed due to damage, and were not included. Labels near data points show the number of specimens upon which the sex ratio was based. SEX RATIO OF POTATO LEAFHOPPERS RECOVERED FROM WATER PAN TRAPS, 1992 1.10 1.00 0.90 0.80 0.70 0.60 196 O 0.50 118 ^ 0.40 X 0.30 (D UJ O w 0.20 0.10 0.00 - 0.10 CM CJ) CO COLLECTION DATE Fig. A.2. Sex ratio of potato leafhoppers recovered from water pan traps, 1992. Sex ratio was expressed as the proportion of total potato leafhoppers recovered that were female. A small number of specimens were not sexed due to damage, and were not included. Labels near data points show the number of specimens upon which the sex ratio was based. Gaps in the plot indicate sampling intervals during which no potato leafhoppers were recovered. APPENDIX B LEAFHOPPERS AND MYMARIDS IDENTIFIED FROM WATER PAN TRAP SAMPLES LEAFHOPPERS IDENTIFIED FROM WATER PAN TRAP SAMPLES’ AGALLIINAE Agalliini Aceratagallia sanguinolenta Provancher Agallia constricta Van Duzee APHRODINAE Aphrodini Aphrodes costatus (Panzer) Aptirodes fuscofasciata (Goeze) Aphrodes prob. flavostrigata CICADELLINAE Cicadellini Draeculacephala antica (Walker) Graphocephala coccinea (FOrster) Helochara communis Fitch COELIDIINAE CoelidiinI Coelidia olitona (Say) DELTOCEPHALINAE Athysanini Athysanus argentarius Metcalf Exitianus exitiosus (Uhler) Limotettix atricapillis (Boheman) Paraphlepslus irroratus (Say) Streptanus confinis (Reuter) ' specimens identified by Tim Miklasiewicz, based upon reference specimens in OARDC Entomology Museum and DeLong, Johnson, and Osborn Collections in the Ohio State University Museum of Biodiversity. Dr. H. D. Blocker, Department o f Entomology, Kansas State University, identified voucher specimens for virtually all species listed, except for Typhlocybinae. 91 Deltocephalinae (continued) Deltocephalini Amblysellus curtisi (Fitch) Colladonus clitellanus (Say) Endna inimica (Say) Exitianus exitiosus (Uhler) Graminella fitchi (Van Duzee) Graminella nigrifrons (Forbes) Latalus sayi (Fitch) Norvellina seminuda (Say) Planocephalus flavocostatus (Van Duzee) Polyamia apicata (Osborn) Scaphoideus sp. Hecalini Parabolocratus major Osbom Macrosteiini Macrosteles quadrilineatus (Forbes) Paralimnini Psammotettix lividellus (Zetterstedt) Scaphytopiini Scaphytopius acutus (Say) Scaphytopius frontalis (Van Duzee) GYPONINAE Gyponana expanda DeLong MACRGPSINAE Macropsini Macropsis insignis (Van Duzee) TYPHLOCYBINAE^ Alebrini Alebra sp. Dikraneurini Dikraneura sp. (probably D. fieberi LOw) Forcipata (Dicranoneura) /oca DeLong & Caldwell Erythroneurini Erythroneura sp. (indet #1 ) Erythroneura sp. (indet #2) Typhlocybini Empoasca erigeron DeLong Empoasca fabae (Harris) Empoasca sp. (indet #1) Empoasca sp. O’ndet. #2) Empoasca sp. Ôndet. #3) Typhlocyba sp. (indet #1 ) Typhlocyba sp. (indet #2) ■ Dr. A. Hamilton, CLBRR, Agriculture and Agri-Food Canada, provided additional assistance with identification of Typhlocybinae. 92 XESTOCEPHALINAE Xestocephalinini Xestocephalus desertonim (Berg) (= X. pulicarius) MYMARIDAE IDENTIFIED FROM WATER PAN TRAP SAMPLES SPECIES'’ PROBABLE HOST* Anagnis nigriventris Girault® Cicadellidae, reared on Empoasca fabae Anagrus sp. (indet. #1 )® Auchenorrtiyncha Anagnis sp. (indet. #2)® Auchenorrhyncha Anaphes iole Girauit Lygus spp. Anaphes {=Patasson) sp. "crassiscomis" group Curculionidae, Chrysomelidae Camptoptera sp. unknown Erythmelus rex Girauit Miridae Gonotocenis rivalis Girauit Cicadellidae Gonotocenis sp. “litoralis" group Cicadellidae Polynema sp. (indet. # 1 ) Miridae. Nabidae Polynema sp. (indet. # 2) Miridae, Nabidae Polynema sp. (indet. # 3) Miridae, Nabidae /Vfymarsp.® unknown 10 additional tentative species®, unidentified Except as noted, all identifications were made by Dr. John Huber, Canadian National Collection of Insects, Agriculture and Agri-Food Canada * host information provided by Dr. John Huber, Canadian National Collection o f Insects, Agriculture and Agri-Food Canada ’ determination by Dr. Serguei Triapitsyn, Department of Entomology, University of California Riverside ® tentative identification by Tim M iklasiewicz 93 APPENDIX C MEASUREMENTS OF LIGHT INTENSITY IN CROPPING SYSTEMS In order to evaluate differences in light intensity within habitats of the different cropping systems studied, m easurements were made using a Li-Cor Line Quantum Sensor and Datalogger (Models LI-191 SB and LI-1000, Li-Cor Inc., Lincoln, NE, U. S. A.). Potato leafhopper have been shown to exhibit positive phototrophism (Smith 1987). Recordings were made above the top plant canopy, at the top of the soybean canopy, and at ground level. The top plant canopy was the wheat canopy for the intercropping system; the soybean canopy was the top canopy for the soybean monoculture. Units of light intensity used in this appendix are pmoles*s'’*m'\ Measurements were made under cloudless conditions, as much as was possible. In order to illustrate how light intensity varies throughout the day. Fig. 0.1 shows fluctuations in light intensity at a height immediately above the top plant canopy (aggregate of all cropping systems). Measurements taken June 13,18, and 21,1990; July 16 and 17, 1990; June 21, 24, 25, and 26, 1991; and July 3, 4, 5, and 9,1991. Time of the day is shown as hours of the 24 hour clock (Eastem Daylight Saving Time). Figures 0.2, 0.3, 0.4, and 0.5 show mean light intensity (Y-axis) for 5 periods of the day (late June, 1991) for the 0:3, 3:3, 7:3, and 14:3 cropping systems, respectively. The periods were mid-moming (8:30-10:30 am), late morning (10:30 am - noon), early afternoon (noon - 2:30 pm), late aftemoon (2:30 - 5:00 pm), and early evening (6:30 - 8:30 pm). Means (heights of vertical bars) and standard errors (caps) for 5 light readings taken at each of 3 different positions within the intercropping system habitats are shown. The positions shown are above the wheat 94 canopy (i.e. unobstructed light intensity), immediately above the soybean canopy, and at ground level parallel to the soybean row (next to soybean stems). Only 2 positions are shown for the soybean monoculture (0:3, no measurement above wheat canopy). While the top of the soybean canopy of the soybean monoculture receives unobstructed light, the top of the soybean canopy in the intercropping systems is shaded by the wheat during part of the day. Fig. C.6 summarizes the discrepency between light intensity above the top canopy (unobstructed) and at the top of the soybean canopy for the 3 intercropping systems. Shading is most pronounced early and late in the day, when the sun is low on the horizon. Potato leafhoppers were observed to have their highest level of flight activity in late afternoon and early evening, if leaves are dry and temperatures are sufficiently warm. Shading of the top of the soybean canopy increases with the amount of wheat in cropping systems. Decreased light intensity illuminating soybean plants in intercropping systems may decrease immigration and/or increase emigration for the intercropping systems. 95 LIGHT ABOVE TOP CANOPY 2400 2200 A A 2000 1800 1600 v > 1400 z UJ 1200 1000 J 800 O 600 400 200 0 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 TIME OF DAY Fig. C.1. Variation in light intensity throughout the day, measured above the top canopy in the cropping systems studied. Units of light intensity are pmo!es*s '*m ’. Time of the day is expressed in hours of tfie 24 hour clock. LIGHT ABOVE SOYBEAN CANOPY AND AT GROUND LEVEL, 1991 2000 1800 1600 Z 1000 h- I 800 (D CO -I 600 “v j 400 200 MID-MORNING LATE MORNING EARLY AFT'NOON LATE AFTNOON EARLY EVENING TIME OF DAY ] 0:3 0:3 GROUND LEVEL SOYBEAN CANOPY Fig. C.2. Comparison of light intensity at the top of the soybean canopy and at ground level for the soybean monoculture (0,3). Heights of bars show means (N=5) with standard error caps. Units of light intensity are pmoles*s ’*m'. LIGHT ABOVE WHEAT AND SOYBEAN CANOPIES, AND AT GROUND LEVEL, 1991 2 0 0 0 1800 - 1600 - V) z 1 4 0 0 - UJ z 1 2 0 0 - t - X 1 0 0 0 - o - 8 0 0 - (O 0 0 600 - 4 0 0 - 2 0 0 — MID-MORNING LATE MORNING EARLY AFT'NOON LATE AFT'NOON EARLY EVENING TIME OF DAY 3:3 3:3 X////A 3:3 GROUND LEVEL SOYBEAN CANOPY WHEAT CANOPY Fig. C.3. Comparison of light intensity at the top of the wheat canopy, at the top of the soybean canopy, and at ground level for the 3:3 intercropping system. Heights of bars show means (N=5) with standard error caps. Units of light intensity are pmoles*s’*m'. LIGHT ABOVE WHEAT AND SOYBEAN CANOPIES, AND AT GROUND LEVEL, 1991 1600 — 1400 - 111 1200 - 1000 - 0 800 — (D MID-MORNING LATE MORNING EARLY AFT'NOON LATE AFT'NOON EARLY EVENING TIME OF DAY 7:3 7:3 XZZZZA 7:3 GROUND LEVEL SOYBEAN CANOPY WHEAT CANOPY Fig. C.4. Comparison of light intensity at the top of the wheat canopy, at the top of the soybean canopy, and at ground level for the 7:3 intercropping system. Heights of bars show means (N=5) with standard error caps. Units of light intensity are pmoles*s’*mV LIGHT ABOVE WHEAT AND SOYBEAN CANOPIES, AND AT GROUND LEVEL, 1991 1600 - (/) 1200 - k - 1 0 0 0 - MID-MORNING LATE MORNING EARLY AFT'NOON LATE AFT’NOON EARLY EVENING TIME OF DAY 14.3 t 2 Z ^ / l 14:3 GROUND SOYBEAN WHEAT Fig. C.5. Comparison of light intensity at the top of the wheat canopy, at the top of the soybean canopy, and at ground level for the 14:3 intercropping system. Heights of bars show means (N=5) with standard error caps. Units of light intensity are pmoles*s'*m’. LIGHT AT SOYBEAN CANOPY TOP AS PROPORTION OF LIGHT ABOVE WHEAT CANOPY 1.00 UJ 0.80 - MID-MORNING LATE MORNING EARLY AFTNOON LATE AFTNOON EARLY EVENING TIME OF DAY 14.3 7:3 XZZZZA 3 3 Fig. C.6. Light intensity at the top of the soybean canopy expressed as a proportion of the amount above the wheat canopy. Heights of the bars provide means (N=5), with standard error caps. Five periods of the day are shown for 3 intercropping systems. APPENDIX D MEASUREMENTS OF HOST PLANT QUANTITY AND QUALITY IN CROPPING SYSTEMS In order to quantify characteristics of cropping systems that may be related to potato leafhopper behavior and abundance, measurements of soybean plants from the cropping systems used for these studies were made in the field and in the laboratory. Measurements were made on 4 July 1991 and 24-27 June 1993. The 1993 samples were taken from research plots located within 200 m of the plots used for the experiments reported in Chapters 2 and 3; these plots used the same cropping systems and agronomic procedures that were almost identical to those reported in Chapters 2 and 3. The vegetative characteristics quantified were height, leaf area, epigeal dry biomass, and vegetative stage (Fehref a/. 1971). Locations for measurements were identified by selecting coordinates without bias within each plot using a random number generator; however samples were not taken within 1.5 m of plot borders. Coordinates were located within soybean rows. From the central point identified by these coordinates, a rectangular area was defined extending 0.5 m perpendicular to the soybean row to the east and w est (row orientation was north-south), and 0.125 m within the row (total area 0.25 m^). The maximum height of the soybean canopy above the soil surface was measured in the field at the sampling coordinates ("soybean canopy height". Table D.1). Then all plant material emerging from the soil within the rectangle was cut at ground level and removed to the laboratory for further measurement. An additional measurement of soybean height, the distance along the main stem between the cotyledonary node and the terminal node, was determined in 102 the lab (Table D.2). Canopy height was based on a single measurement per sample (at coordinates). The laboratory procedure for estimating soybean height involved measuring all the soybean plants that were han/ested from the 0.25 m^ sample rectangles. In the laboratory, the vegetative stage of each plant in a sample was rated (Fehr etal. 1971). Then the leaves were separated from other plant parts and the aggregate leaf area of the sample was determined (LI-3100 Area Meter, Li-Cor, Inc., Lincoln, NE, U. S. A.). Finally, leaves and other plant parts for each sample were air-dried and weighed (Metier PC4000, Mettler Instrument Corp., Hightstown, NJ, U. S. A.) to determine epigeal biomass (1993 only). Data is summarized for leaf area in Table D.3 and biomass in Table D.4. The summary data presented for vegetative stage (Table D.5) is the mean of all plants measured. Because the relationships among cropping systems were often substantially different for different replications, statistics for individual replications are provided in addition to overall statistics for cropping systems. 103 Table D.1 SOYBEAN CANOPY HEIGHT IN FOUR CROPPING SYSTEMS, MEASURED IN THE FIELD 1991 Cropping System 0:3 Replication 1 2 3 0:3 Total Number of Samples NA 5 5 10 Mean Height NA 42.6 32.7 37.7 Standard Deviation NA 6.6 6.6 8.1 Maximum NA 53.0 40.0 53.0 Minimum NA 35.0 22.0 22.0 Cropping System 3:3 Replication 1 2 3 3:3 Total Number of Samples NA 5 5 10 Mean Height NA 50.90 48.8 49.9 Standard Deviation NA 5.1 4.6 4.7 Maximum NA 58.0 53.0 58.0 Minimum NA 46.0 43.0 43.0 Cropping System 7:3 Replication 1 2 3 7:3 Total Number of Samples NA 5 5 10 Mean Height NA 45.5 45.3 45.4 Standard Deviation NA 8.1 8.8 8.0 Maximum NA 57.0 54.5 57.0 Minimum NA 35.0 31.0 31.0 Cropping System 14:3 Replication 1 2 3 14:3 Total Number of Samples NA 5 5 10 Mean Height NA 41.2 40.5 40.9 Standard Deviation NA 3.8 6.7 5.2 Maximum NA 47.0 48.5 48.5 Minimum NA 37.0 33.0 33.0 Cropping System ALL Replication ALL Number of Samples 40 Mean Height 43.3 Standard Deviation 8.0 Maximum 58.0 Minimum 22.0 Table is continued on the following page. Table D.1. Soybean canopy height in four cropping systems, measured in the field, 1991 and 1993. Summary statistics for each cropping system by replication and overall for each year. Units of height m easurem ent are cm. 104 Table D.1 (continued) SOYBEAN CANOPY HEIGHT IN FOUR CROPPING SYSTEMS. MEASURED IN THE FIELD 1993 Cropping System 0:3 Repiication 1 2 3 4 0:3 Total Number of Samples 5 5 5 5 20 Mean Height 20.1 16.5 19.5 20.4 19.1 Standard Deviation 2.2 3.6 4.8 1.1 3.4 Maximum 23.5 22.5 26.0 22.0 26.0 Minimum 18.5 14.0 14.0 19.0 14.0 Cropping System 3:3 Replication 1 2 3 4 3:3 Total Number of Samples 5 5 5 5 20 Mean Height 20.7 20.1 18.7 19.7 19.8 Standard Deviation 2.1 2.8 1.3 3.0 2.3 Maximum 23.5 23.0 20.0 22.0 23.5 Minimum 17.5 16.0 16.5 16.0 18.0 Cropping System 7:3 Replication 1 2 3 4 7:3 Total Number of Samples 5 5 5 5 20 Mean Height 25.8 22.2 18.7 21.9 22.2 Standard Deviation 5.2 5.0 2.1 2.9 4.5 Maximum 31.5 28.0 21.0 24.0 31.5 Minimum 20.0 17.0 16.0 17.0 16.0 Cropping System 14:3 Replication 1 2 3 4 14:3 Total Number of Samples 5 5 5 5 20 Mean Height 22.5 25.0 23.4 17.9 22.2 Standard Deviation 4.1 0.9 2.2 5.1 4.2 Maximum 27.5 26.0 26.0 23.0 27.5 Minimum 17.5 23.5 20.0 9.5 9.5 Cropping System ALL Replication ALL Number of Samples 80 Mean Height 20.8 Standard Deviation 3.9 Maximum 31.5 Minimum 9.5 105 Table D.2 HEIGHT OF SOYBEAN IN FOUR CROPPING SYSTEMS, DISTANCE BETWEEN COTYLEDONARY NODE AND TERMINAL NODE 1991 Cropping System 0:3 Replication 1 2 3 0:3 Total Number of Plants in Samples 21 12 10 43 Mean Height 27.50 29.71 23.70 27.23 Standard Deviation 5.37 6.58 6.58 6.25 Maximum 35.0 39.5 32.0 39.5 Minimum 18.0 17.0 13.5 13.5 Cropping System 3:3 Replication 1 2 3 3:3 Total Number of Plants in Samples 40 29 25 94 Mean Height 34.88 37.90 35.13 35.89 Standard Deviation 6.80 4.97 3.69 5.71 Maximum 45.5 47.5 40.0 47.5 Minimum 13.5 21.0 28.5 13.5 Cropping System 7:3 Replication 1 2 3 7:3 Total Number of Plants in Samples 42 34 35 111 Mean Height 35.31 31.02 34.97 33.90 Standard Deviation 8.07 8.58 9.20 8.70 Maximum 43.5 47.0 48.5 48.5 Minimum 0.0 14.0 14.0 0.0 Cropping System 14:3 Repiication 1 2 3 14:3 Total Number of Plants in Samples 29 30 40 99 Mean Height 29.86 29.85 30.09 29.95 Standard Deviation 7.45 7.08 5.98 6.70 Maximum 38.0 42.5 40.5 42.5 Minimum 1.5 12.5 17.0 1.5 Cropping System ALL Repiication ALL Number of Plants in Samples 347 Mean Height 32.44 Standard Deviation 7.69 Maximum 48.5 Minimum 0.0 Table is continued on the following page. Table D.2. Height of soybean in four cropping systems, distance between cotyledonary node and terminal node, 1991 and 1993. Summary statistics for each cropping system by replication and overall for each year. Units of height measurement are cm. 106 Table D.2 (continued) HEIGHT OF SOYBEAN IN FOUR CROPPING SYSTEMS, DISTANCE BETWEEN COTYLEDONARY NODE AND TERMINAL NODE 1993 Cropping System 0:3 Replication 1 2 3 4 0:3 Total Number of Plants in Samples 26 24 30 35 115 Mean 13.8 11.2 13.0 13.1 12.9 Standard Deviation 2.6 3.0 2.8 2.1 2.7 Maximum 17.0 15.5 17.0 17.0 17.0 Minimum 8.0 1.0 8.0 5.5 1.0 Cropping System 3:3 Replication 1 2 3 4 3:3 Total Number of Plants in Sam ples 37 30 26 29 122 Mean 12.0 12.1 10.5 12.9 11.9 Standard Deviation 2.3 3.0 1.9 2.9 2.7 Maximum 15.0 17.0 14.5 19.5 19.5 Minimum 6.0 4.0 7.0 8.0 4.0 Cropping System 7:3 Replication 1 2 3 4 7:3 Total Number of Plants in Samples 36 24 33 35 128 Mean 16.4 16.2 11.8 13.6 14.4 Standard Deviation 4.9 4.2 2.1 3.6 4.3 Maximum 23.0 22.5 16.0 19.0 23.0 Minimum 5.0 9.5 8.0 4.5 4.5 Cropping System 14:3 Replication 1 2 3 4 14:3 Total Number of Plants in Samples 30 24 24 33 111 Mean 14.7 179 15.4 11.7 14.6 Standard Deviation 3.6 3.0 3.0 2.8 3.8 Maximum 24.5 25.0 19.5 14.5 25.0 Minimum 8.0 10.0 8.0 3.5 3.5 Cropping System ALL Repiication ALL Number of Plants in Samples 476 Mean 13.4 Standard Deviation 3.6 Maximum 25.0 Minimum 1.0 107 Table D.3 LEAF AREA OF SOYBEAN IN FOUR CROPPING SYSTEMS 1991 Cropping System 0:3 Replication 1 2 3 0:3 Total Number of Samples 5 5 5 15 Mean 3281.83 2586.02 1738.76 2535.54 Standard Deviation 669.15 1308.83 1088.62 1175.81 Maximum 4444.21 4524.80 3289.62 4524.8 Minimum 2752.40 1451.00 409.57 409.57 Cropping System 3:3 Replication 1 2 3 3:3 Total Number of Samples 5 5 5 15 Mean 2274.43 2504.89 1523.72 2101.01 Standard Deviation 549.31 1493.19 463.74 986.26 Maximum 2734.85 4642.28 1951.48 4642.28 Minimum 1326.02 1187.89 736.23 736.23 Cropping System 7:3 Replication 1 2 3 7:3 Total Number of Samples 5 5 5 15 Mean 1970.73 1330.67 1515.92 1605.77 Standard Deviation 685.23 583.09 861.27 721.61 Maximum 2542.41 2153.87 2656.68 2656.68 Minimum 788.75 698.02 620.47 620.47 Cropping System 14:3 Replication 1 2 3 14:3 Total Number of Samples 5 5 5 15 Mean 730.47 566.74 880.24 725.81 Standard Deviation 545.18 315.43 306.11 397.09 Maximum 1605.71 946.60 1237.92 1605.71 Minimum 128.08 247.73 490.66 128.08 Cropping System ALL Replication ALL Number of Samples 60 Mean 1742.04 Standard Deviation 1086.26 Maximum 4642.28 Minimum 128.08 Table is continued on the following page. Table D.3. Leaf area for soybean in four cropping systems, 1991 and 1993. Statistics shown here are for aggregate leaf area of plants obtained from a sample corresponding to 0.25 m^ ground area. Units of area measurem ent are cm^. 108 Table D.3 (continued) LEAF AREA OF SOYBEAN IN FOUR CROPPING SYSTEMS, 1993 Cropping System 0:3 Replication 1 2 3 4 0:3 Total Number of Samples 5 5 5 5 20 Mean 738.99 534.64 902.603 1042.45 804.67 Standard Deviation 468.31 322.13 580.38 427.19 463.76 Maximum 1434.61 996.25 1816.71 1529.49 1816.71 Minimum 341.88 211.29 366.24 500.70 211.29 Cropping System 3:3 Replication 1 2 3 4 3:3 Total Number of Sam ples 5 5 5 5 20 Mean 660.87 553.51 399.57 438.61 513.14 Standard Deviation 211.95 281.11 165.22 212.46 228.85 Maximum 924.84 787.4 631.26 634.27 924.84 Minimum 421.82 246.63 252.81 121.17 121.17 Cropping System 7:3 Replication 1 2 3 4 7:3 Total Number of Samples 5 5 5 5 20 Mean 708.44 561.05 465.51 571.73 576.68 Standard Deviation 357.96 368.92 104.70 327.98 297.45 Maximum 1289.44 1103.81 602.24 1034.60 1289.44 Minimum 399.79 198.65 350.01 186.71 186.71 Cropping System 14:3 Replication 1 2 3 4 14:3 Total Number of Samples 5 5 5 5 20 Mean 373.76 341.84 315.43 267.29 323.67 Standard Deviation 307.71 199.95 152.11 157.96 200.16 Maximum 833.09 614.99 502.77 465.90 833.09 Minimum 65.41 134.18 158.54 36.91 36.91 Cropping System ALL Repiication ALL Number of Sam ples 80 Mean 557.46 Standard Deviation 354.30 Maximum 1816.71 Minimum 36.91 109 Table D.4 BIOMASS OF SOYBEAN IN FOUR CROPPING SYSTEMS, 1993 Cropping System 0:3 Replication 1 2 3 4 0:3 Total Number of Samples 5 5 5 5 20 Mean 5.7 3.7 6.6 6.7 5.7 Standard Deviation 3.4 2.2 3.8 2.5 3.0 Maximum 10.9 7.1 12.2 9.6 12.2 Minimum 3.0 1.5 2.8 3.5 1.5 Cropping System 3:3 Replication 1 2 3 4 3:3 Total Number of Samples 5 5 5 5 20 Mean 4.6 4.3 3.3 3.2 3.8 Standard Deviation 1.4 1.9 1.4 1.7 1.6 Maximum 6.3 6.7 5.0 5.0 6.7 Minimum 2.8 1.8 1.9 0.7 0.7 Cropping System 7:3 Replication 1 2 3 4 7:3 Total Number of Samples 5 5 5 5 20 Mean 3.9 3.1 3.4 3.7 3.5 Standard Deviation 2.4 2.2 1.0 2.1 1.9 Maximum 8.1 5.9 4.9 6.8 8.1 Minimum 2.4 0.5 2.3 1.3 0.5 Cropping System 14:3 Replication 1 2 3 4 14:3 Total Number of Samples 5 5 5 5 20 Mean 2.4 2.1 2.0 1.7 2.0 Standard Deviation 1.9 0.7 0.8 1.0 1.1 Maximum 5.2 3.2 2.9 2.8 5.2 Minimum 0.5 1.4 1.0 0.6 0.5 Cropping System ALL Replication ALL Number of Samples 80 Mean 3.8 Standard Deviation 2.4 Maximum 12.2 Minimum 0.5 Table D.4. Biomass for soybean in four cropping systems, 1993. Statistics shown here are for aggregate dry epigial biomass of plants obtained from a sample corresponding to 0.25 m^ ground area. Units of biomass measurement are grams. 110 Table D.5 VEGETATIVE STAGE OF SOYBEAN IN FOUR CROPPING SYSTEMS, 1991 Cropping System 0:3 Replication 1 2 3 0:3 Total Number of Plants in Samples 21 12 10 43 Mean 8.2 8.5 7.6 8.1 Standard Deviation 1.8 2.4 2.1 2.0 Maximum 11 12 11 12 Minimum 4 5 5 4 Cropping System 3:3 Replication 1 2 3 3:3 Total Number of Plants in Samples 40 29 25 94 Mean 5.6 6.3 6.6 6.1 Standard Deviation 1.2 1.3 0.8 1.2 Maximum 8 9 8 9 Minimum 2 4 5 2 Cropping System 7:3 Replication 1 2 3 7:3 Total Number of Plants in Samples 42 34 35 111 Mean 5.5 4.8 5.3 5.2 Standard Deviation 0.9 1.5 1.8 1.4 Maximum 7 8 8 8 Minimum 2 2 1 1 Cropping System 14:3 Replication 1 2 3 14:3 Total Number of Plants in Samples 29 30 40 99 Mean 4.4 3.9 3.8 4.0 Standard Deviation 1.0 1.1 0.9 1.0 Maximum 6 6 6 6 Minimum 1 2 2 1 Cropping System ALL Replication ALL Number of Plants in Samples 347 Mean 5.5 Standard Deviation 1.9 Maximum 12 Minimum 1 Table is continued on the following page. Table D.5. Vegetative stage for soybean in four cropping systems, 1991 and 1993. Statistics shown here are based upon stage ratings for individual plants. I ll Table D.5 (continued) VEGETATIVE STAGE OF SOYBEAN IN FOUR CROPPING SYSTEMS 1993 Cropping System 0:3 Replication 1 2 3 4 0:3 Total Number of Plants in Samples 26 24 30 35 115 Mean 4.5 4.0 4.4 3.7 4.1 Standard Deviation 0.6 0.7 0.7 0.6 0.7 Maximum 5 5 5 5 5 Minimum 3 2 2 2 2 Cropping System 3:3 Replication 1 2 3 4 3:3 Total Number of Plants in Samples 37 30 26 29 122 Mean 2.8 2.9 2.8 2.5 2.7 Standard Deviation 0.5 0.5 0.7 0.5 0.6 Maximum 4 4 4 3 4 Minimum 2 2 1 2 1 Cropping System 7:3 Replication 1 2 3 4 7:3 Total Number of Plants in Samples 36 24 33 35 128 Mean 2.8 3.2 2.6 2.7 2.8 Standard Deviation 0.7 0.6 0.7 0.5 0.6 Maximum 4 4 4 4 4 Minimum 1 2 2 2 1 Cropping System 14:3 Replication 1 2 3 4 14:3 Total Number of Plants in Samples 30 24 24 33 111 Mean 2.5 2.9 2.8 2.1 2.5 Standard Deviation 0.5 0.3 0.4 0.5 0.5 Maximum 3 3 3 3 3 Minimum 2 2 2 1 1 Cropping System ALL Replication ALL Number of Plants in Sam ples 476 Mean 3.0 Standard Deviation 0.9 Maximum 5 Minimum 1 112 REFERENCES Acker, D. 1987. Production Economies, pp. 101-107. In: J. E. Beuerlein, H. Lafever, and P. Lipps (éd.). The Soybean in Ohio. Extension Bull. 741. Ohio Cooperative Extension Service. The Ohio State University, Columbus, OH, U. S. A. 130 pp. Ackerman, A. J. 1931. The leaf hoppers attacking apple in the Ozarks. Technical Bulletin No. 263. U. S. Dept. Agric. Ali, A/I. 1990. Light interception and growth of intercropped soybean into winter wheat. Ph.D. Dissertation. The Ohio State Univesity, Columbus, OH, U. S. A. Altieri, M. A., A. van Schoonhoven, and J. Doll. 1977. The ecological role of weeds in integrated pest management systems: a review illustrated by bean {Phaseolus vulgaris) cropping systems. PANS 23:195-205. Altieri, M. A., C. A. Francis, A. V. Schoonhoven, & J. A. Doll. 1978. A review of pest prevalence in maize (Zea mays L.) and bean {Phaseolus vulgaris L.) polycultural systems. Field Crops Res. 1: 33-44. Andow, D. A. 1983a. Agricultural crop diversity and insect populations. In: D. Pimentel (ed.). Ecological Aspect of Pest Management. Dept. Entomology, Comell University, Ithaca, NY, U. S. A. 1983b. Plant diversity and insect populations in experimental agroecosystems: interactions among beans, weeds, and insects. Ph.D. Dissertation, Comell University, Ithaca, NY. University Microfilms International, Ann Arbor, Ml, U. S. A. 201 pp. 1988. Management of weeds for insect manipulation in agroecosystems, pp. 265-302. In: M. A. Altieri and M. Z. Liebman (ed.). W eed Management in Agroecosystems: Ecological Approaches. CRC, Boca Raton, FL, U. S. A. 1991. Vegetational Diversity and Arthropod Population Response. Annu. Rev. Entomol. 36: 561-586. 1992. Population density of Empoasca fabae (Homoptera: Cicadellidae) in weedy beans. J. Econ. Entomol. 85: 379-383. Batten, E. T., and F. W. Poos. 1938. Spraying and dusting to control the potato leafhopper on peanuts in Virginia. Bulletin 316. Bureau of Entomology, U. S. Dept. Agric. and Virginia Agric. Exp. Sta. 113 Carlson, J. D., M. E. Whaion, D. A. Landis, and S. H. Gage. 1991. Evidence for long-range transport of potato leafhopper Into Michigan. Proc. Amer. Meteorol. Soc. Conf. on Biometeorology. Sept, 1991. Chan, L. M., R. R. Johnson, and C. M. Brown. 1980. Relay Intercropping soybeans Into winter wheat and spring oats. Agron. J. 72: 35-39. Coggins, M. L. 1991. Potato leafhopper{Empoasca fabae) and alfalfa weevil {Hypera postica) density and damage In binary mixtures of alfelfe and forage grasses. M. S. Thesis, Michigan State University, East Lansing, Ml, U. S. A. Decker, G. C., and H. B. Cunningham. 1967. The mortality rate of the potato leafhopper and some related species when subjected to prolonged exposure at various temperatures. J. Econ. Entomol. 60; 373-379. 1968. Winter survival and overwintering area of the potato leafhopper. J. Econ. Entomol. 61(1): 154-161. Decker, C. G., and J. V. Maddox. 1967. Cold hardiness of Empoasca fabae and some related species. J. Econ. Entomol 60: 1641-1645. Decker, G. C., C. A. Kouskoulekas, and R. J. Dysart. 1971. Some observations on fecundity and sex ratios of the potato leafhopper. J. Econ. Entomol. 64:1127-1129. DIetrick, E. J. 1961. An Improved backpack motor fan for suction sampling of Insect populations. J. Econ. Entomol. 54: 394-395. DeLong, D. M. 1938. Biological studies on the leafhopper, Empoasca fabae, as a bean pest. Tech. Bull. 618, U. S. Dept. Agric. 60 pp. 1965. Ecological aspects of North American leafhoppers and their role In agriculture. Bull. Entomol. Soc. Amer. 11: 9-26. 1971. The bionomics of leafhoppers. Ann. Rev. Entomol. 16:179-210. DeLong, D. M., and J. S. Caldwell. 1935. Hibernation studies of the potato leafhopper (Empoasca fabae Harris) and related species of Empoasca occurring In Ohio. J. Econ. Entomol. 28: 442-444. DeLong, D. M., and Fenton, F. A., and A. Hartzell. 1923. Bionomics and control of the potato leafhopper, Empoasca mali Le Baron. Research Bulletin No. 78. Agricultural Experiment Station. Iowa State College of Agriculture and Mechanic Arts. Ames, lA, U. S. A. Dethler, V. G. 1982. Mechanisms of host-plant recognition. Entomol. exp. appl. 31: 49-56. Doutt, R. L., and J. Nakata. 1965. Overwintering refuge of Anagrus epos (Hymenoptera: Mymaridae). J. Econ. Entomol. 58: 586. 1973. The Rubus leafhopper and Its egg parasltold: An endemic biotic system useful In grape pest management. Environ. Entomol. 2: 381-386. 114 Fehr, W. R., C. E. Caviness, D. T. Burwood, and J. S. Pennington. 1971. Stage development description of soybeans {Glycine max (L.) Merr.). Crop Sel. 11: 929-931. Fleischer, S. J., W. A. Allen, J. M. Luna, and R. L. Pienkowski. 1982. Absolute-density estimation from sweep sampling, with a comparison of absolute-density sampling techniques for adult potato leafhoppers in alfelfa. J. Econ. Entomol. 75: 425-430. Flanders, K. L, and E. B. Radcliffe. 1989. Origins of potato leafhoppers (Homoptera: Cicadellidae) invading potato and snap bean in Minnesota. Environ. Entomol. 18:1015- 1024. Fiinn, P. W., R. A. J. Taylor, and A. A. Mower. 1986. Predictive model for the population dynamics of potato leafhopper, Empoasca fabae (Homoptera: Cicadellidae), on alfalia. Environ, Entomol. 15: 898-904. Flinn, P. W.. A. A. Mower, and D. P. Knievel. 1990a. Physiological response of alfalfa to injury by Empoasca fabae (Homoptera: Cicadellidae). Environ. Entomol. 19:176-181. Flinn, P. W., A. A. Mower, and R. A. J. Taylor. 1990b. Immigration, sex ratio, and local movement of the potato leafhopper (Homoptera: Cicadellidae) in a Pennsylvania alfalfa field. J. Econ. Entomol 83:1858-1862. Freytag, P. H. 1985. The insect parasites of leafhoppers, and related groups, pp. 423-468. In: L. R. Nault and J. G. Rodriguez (ed.). The Leafhoppers and Planthoppers. John Wiley & Sons, New York, NY, U. S. A. 500 pp. Gentsch, B. J. 1982. The influence of weeds on the incidence of the potato leafhopper in alfalfa. M. S. Thesis. Southem Illinois Univ., Carbondale, IL, U. S. A. Gyrisco, G. G. 1958. Forage insects and their control. Ann. Rev. Entomol. 3: 421-448. Hammond, R. B., and D. L. Jeffers. 1990. Potato leafhopper (Homoptera: Cicadellidae) populations on soybean intercropped into winter wheat. Environ. Entomol. 19:1810- 1819. Hammond, R. B., and B. R. Stinner. 1987. Soybean foliage insects in conservation tillage systems: effects of tillage, previous cropping history, and soil insecticide application. Environ. Entomol. 16: 524-531. Helm, C. G., M. Kogan, and B. G. Hill. 1980. Sampling leafhoppers on soybean, pp. 260-282. In: M. Kogan and D. C. Herzog (ed.). Sampling Methods in Soybean Entomology. Springer-Verlag, New York, Inc., New York, NY, U. S. A. 587 pp. Hildebrand, P. E. 1976. Multiple cropping systems are dollars and “sense" agronomy, pp. 347- 371. In: R. I. Papendick ef a/, (ed.). Multiple Cropping. Special Publ. 27. American Society of Agronomy, Madison, Wl, U. S. A. Hogg, D. B. 1985. Potato leafhopper (Homoptera: Cicadellidae) immature development, life tables, and population dynamics under fluctuating temperature regimes. Environ. Entomol. 14: 349-355. 115 Hogg, D. B., and G. D. Hoffman. 1989. Potato leafhopper population dynamics. In: E. J. Armbrust and W. O. Lamp (éd.). Proceedings of a Symposium: History and Perspectives of Potato Leafhopper (Homoptera: Cicadellidae) Research. Mise. Publications No. 72, Entomol. Soc. America Hook, J. E., and G. J. Gascho 1988. Multiple Cropping for Efficient Use of Water and Nitrogen, pp. 7-20. In: W. L. Hargrove (ed.). Cropping Strategies for Efficient Use of Water and Nitrogen. American Society of Agronomy, inc./Crop Science Society of America, lnc./Soii Science Society of America, Inc., Madison, Wl, U. S. A. Hower, A. A. 1989. Potato leafhopper as a plant stress factor on alfalfe. In: E. J. Armbrust and W. O. Lamp (ed.). Proceedings of a Symposium: History and Perspectives of Potato Leafhopper (Homoptera: Cicadellidae) Research. Misc. Publications No. 72, Entomol. Soc. America Huber, J. T. 1986. Systematics, biology, and hosts of the Mymaridae and Mymarommatidae (Insecta: Hymenoptera). Entomography 4:185-243. Huff, F. A. 1963. Relation between leafhopper influxes and synoptic weather conditions. J. Appl. Meteorol. 2: 39-43. Jeffers, D. L. 1987. Multiple cropping, pp. 90-95. In: J. E. Beuerlein, H. Lafever, and P. Lipps (ed ). The Soybean in Ohio. E^ension Bull. 741. Ohio Cooperative Extension Service. The Ohio State University, Columbus, OH, U. S. A. 130 pp. 1990. Multiple cropping, pp. 22-26. In: J. E. Beuerlein, H. Lafever, and P. Lipps (ed.). Profitable Wheat Management. Extension Bull. 811. Ohio Cooperative Extension Service. The Ohio State University, Columbus, OH, U. S. A. 88 pp. Jeffers, D. L., and G. B. Triplett, Jr. 1979. M anagement needed for relay cropping soybeans and wheat. Ohio Report 64: 67-70. Jeffers, D. L., G. B. Triplett, Jr., and H. N. Lafever. 1977. Relay intercropping soybeans and wheat. Ohio Rep. 64: 67-70. Jermy, T. 1971. Biological background and outlook of the antifeedant approach to insect control. Acta. Phytopathol. Acad. Sci. Hung. 6: 253-260. Kareiva, P. 1982. Experimental and mathematical analyses of herbivore movement: quantifying the influence of plant spacing and quality on foraging discrimination. Ecol. Monogr. 52:261-282. 1983. Influence of vegetation texture on herbivore populations: In: R. F. Denno and M. S. McClure (ed.). Variable Plants and Herbivores in Natural and Managed Ecosystems. Academic Press, New York, NY, U. S. A. pp. 259-289. Kennedy, J. S. 1985. Migration, behavioral and ecological. In: M. A. Rankin (ed.). Contributions in Marine Science 27. Marine Science Institute, University of Texas at Austin. Port Aransas, TX, U. S. A. 116 Kieckheffer, R. W., and J. T. Medier. 1964. Some environmental fectors influencing oviposition by the potato leafhopper, Empoasca fabae. J. Econ. Entomol 57: 482-484. Kingsley, P. C., J. M. Schriber. and R. G. Harvey. 1986. Relationship of weed density with leafhopper populations in alfalfa. Wisconsin Agric. Ecosyst. Environ. 17: 281-286. Kouskolekas, C., and G. C. Decker. 1966. The effect of temperature on the rate of development of the potato leafhopper, Empoasca fabae (Homoptera: Cicadellidae). Ann. Entomol. Soc. Am. 59: 292-298. 1968. A quantitative evaiuation of Actors affecting alfal^ yield reduction caused by potato leafhopper attack. J. Econ. Entomol. 61: 921-927. Ladd, T. L., and W. A. Rawlins. 1965. The effects of feeding of the potato leafhopper on photosynthesis and respiration in the potato plant. J. Econ. Entomol. 58: 623-628. Lamont, W. J., Jr. 1981. Effect of relay intercropping system design on the component crops, dry beans {Phaseolus vulgaris L.) with winter wheat {Triticum aestivum L.) or spring oats {Avena sativa L.) and on insect populations. Ph. D. Dissertation. Comell University, Ithaca, NY, U. S. A. Lamp, W. O. 1991. Reduced Empoasca fabae (Homoptera: Cicadellidae) density in oat-aifaife intercrop systems. Environ. Entomol. 20:118-126. Lamp, W. O., R. J. Barney, E. J. Armbrust, and G. Kapusta. 1984a. Selective weed control in spring-pianted alfelfa: effect on leafhoppers and planthoppers (Homoptera: Auchenorrhyncha), with emphasis on potato leafhopper. Environ. Entomol. 13(1): 207- 213. Lamp, W. O., M. J. Morris, and E. J. Armbrust. 1984b. Suitability of common weed species as host piants for the potato leafhopper, Empoasca fabae. Entomol. exp. appl. 36:125- 131. Lamp, W. O., and L. M. Smith. 1989. Sampling objectives and problems. In: E. J. Armbrust and W. 0. Lamp (ed.). Proceedings of a symposium: History and Perspectives of Potato Leafhopper (Homoptera: Cicadellidae) Research. Misc. Publications No. 72, Entomol. Soc. America. Lamp, W. O., and L. Zhao. 1993. Prediction and manipulation of movement by polyphagous, highly mobile pests. J. Agric. Entomol. 10: 267-281. Lance, D. R. 1983. Host-seeking behavior of the gypsy moth: the influence of polyphagy and highly apparent host plants, pp. 201-204. In: S. Ahmad (ed.). Herbivorous Insects. Academic Press, New York, NY, U. S. A. 257 pp. Letoumeau, D. K. 1990. Abundance patterns of leafhopper enemies in pure and mixed stands. Environ. Entomol. 19: 505-509. Levene, H. 1960. Robust tests for equality of variances. In: I. Olkin (ed.). Contributions to Probability and Statistics. Stanford University Press, Stanford, CA, U. S. A. pp. 278- 292 117 Mayse, M.. M. Kogan, and P. W. Price. 1978. Sampling abundance of soybean arthropods: Comparison of methods. J. Econ. Entomol. 71:135-141. Marston, N. L, C. E. Morgan, G. D. Thomas, and 0. M. Ignoffo. 1976. Evaluation of four techniques for sampling soybean insects. J. Kansas Entomol. Soc. 49: 389-400. Medier, J. T. 1957. Migration of the potato leafhopper-a report on a cooperative study. J. Econ. Entomol. 50: 493-497. Miller, J. R., and K. L. Strickler. 1984. Finding and accepting host plants, pp. 127-157. In: W. J. Bell and R. T. Cardé (éd.). Chemical Ecology of Insects. Chapman and Hall, New York, NY, U. S. A. NC-193 Technical Committee. 1990. Spatial Dynamics of Leafhopper Pests and their Management on Alfelfa. Annual Report for the NC-193 Technical Committee for 1989. pp. 18-26. 1991. Spatial Dynamics of Leafhopper Pests and their Management on Alfalfe. Annual Report for the NC-193 Technical Committee for 1992. p. 24. 1993. Spatial Dynamics of Leafhopper Pests and their Management on Alfalfa. Annual Report for the NC-193 Technical Committee for 1992. pp. 18-20. Nielsen, G. R., W. O. Lamp, and G. W. Stutte. 1990. Potato leafhopper (Homoptera: Cicadellidae) feeding disruption of phloem translocation in alfalfa. J. Econ. Entomol 83: 807-813. Ogunlana, M. O., and L. P. Pedigo. 1974. Economic injury level of the potato leafhopper on soybeans in Iowa. J. Econ. Entomol. 67: 29-32. Ohio Cooperative Extension Service. 1988. Ohio Agronomy Guide, 12th edition. The Ohio State University, Columbus, OH, U. S. A. Ohio State University Extension. 1997. Midwest Small Fruit Pest Management Handbook. Bulletin 861. R. C. Funt, M. A. Ellis, & C. Welty (ed.). The Ohio State University, Columbus, OH, U.S.A. 173 pp. (potato leafhopper on strawberry, brambles, and grape). Oloumi-Sadeghi, H., L. R. Zavaleta, W. O. Lamp, E. J. Armbrust, and G. Kapusta. 1987. Interactions of the potato leafhopper (Homoptera: Cicadellidae) with weeds in an alfalfa ecosystem. Environ. Entomol. 16:1175-1180. Oloumi-Sadeghi, H., L. R. Zavaleta, G. Kapusta, W. O. Lamp, & E. J. Armbrust. 1989. Effects of potato leafhopper (Homoptera: Cicadellidae) and weed control on a\fa\fa yield and quality. J. Econ. Entomol. 82: 923-931. Pienkowski, R. L., and J. T. Medier. 1964. Synoptic weather conditions associated with long range movements of the potato leafhopper, Empoasca fabae, into Wisconsin. Ann. Entomol. Soc. Amer. 57: 588-591. 118 Poos, F. W., and C. M. Haenseler. 1931. Injury to varieties of eggplant by the potato leafhopper, Empoasca fabae (Harris). J. Econ. Entomol. 24:890-892 (with 2 plates). Poos, F. W., and H. W. Johnson. 1936. Injury to alfalfa and red clover by the potato leafhopper. J. Econ. Entomol 29:325-331. Poos, F. W., and N. H. Wheeler. 1943. Studies of host plants of the leafhoppers of the genus Empoasca. Tech. Bull. 850, U. S. Dept. Agric. 51 pp. 1949. Some additional host plants of three species of leafhoppers of the genus Empoasca (Homoptera: Cicadellidae). Proc. Entomol. Soc. Wash. 51: 35-38. Power, A. G. 1987. Plant community diversity, herbivore movement, and an insect-transmitted disease of maize. Ecology 68:1658-1669. Ranne, E. L, and W. 0. Lamp. 1990. Measurement of the ovipositional potential of potato leafhopper, Empoasca fabae (Homoptera: Cicadellidae): a comparison of feral and culture populations. J. Kansas Entomol. Soc. 63:420-426. Risch, S. J. 1981. Insect herbivore abundance in tropical monocultures and polycultures: an experimental test of two hypotheses. Ecology 62:1325-1340. Risch, S. J., D. Andow, and M. A. Altieri. 1983. Agroecosystem diversity and pest control: data, tentative conclusions, and new directions. Environ. Entomol. 12: 625-629. Roltsch, W. J., and S. H. Gage. 1990a. Potato leafhopper (Homoptera: Cicadellidae) movement, oviposition, and feeding response patterns in relation to host and nonhost vegetation. Environ. Entomol. 19: 524-533. 1990b. Influence of bean-tomato intercropping on population dynamics of the potato leafhopper (Homoptera: Cicadellidae). Environ. Entomol. 19: 534-543. Root, R. B. 1973. Organization of a plant- arthropod association in simple and diverse habitats: the feuna of collards. Ecol. Monogr. 43: 95-124. Russell, E. P. 1989. Enemies hypothesis: a review of the effect of vegetational diversity on predatory insects and parasitoids. Environ. Entomol. 18: 590-599. Schoonhoven, L. M. 1968. Chemosensory bases of host-selection. Annu. Rev. Entomol. 13: 115-136. Schoonhoven, A. V., C. Cardona, J. Garcia, & F. Garzon. 1981. Effect of weed covers on Empoasca kraemeri Ross & Moore populations and dry bean yields. Environ. Entomol. 10: 901-907. Sheehan, W. 1986. Response of specialist and generalist natural enemies to agroecosystem diversification: a selective review. Environ. Entomol. 15:456-461. Shields, E. J., and R. B. Sher. 1992. Low-temperature sun/ival strategies of potato leafhopper (Homoptera: Cicadellidae). Environ. Entomol. 21: 301-306. 119 Simonet, D. E., R. L Pienkowski, D. G. Martinez, and R. D. Blakesiee. 1979. Evaluation of sampling techniques and development of a sampling program for potato leafhopper adults on alfalfa. Environ. Entomol. 8: 397-399. Smith, A. W., R. B. Hammond, and B. L. Stinner. 1988. Influence of rye-cover crop management on soybean foliage arthropods. Environ. Entomol. 17:109-114. Smith, F. F., and F. W. Poos. 1931. The feeding habit of some leafhoppers of the genus Empoasca. J. Agric. Res. 43: 267-285. Smith, L. M. 1987. Mechanisms by which grass reduces potato leafhopper, Empoasca fabae (Harris) (Homoptera: Cicadellidae), abundance in alfelfa. Ph.D. Dissertation, University of Illinois, Urbana-Champaign, IL, U. S. A. 200 pp. Smith, L. M., W. O. Lamp, and E. J. Armbrust. 1992. Potato leafhopper acceptance of alfalfe as a host; the role of non-host stimuli. J. Entomol. Sci. 27: 56-64. Specker, D. R., E. J. Shields, D. M. Umbach, and S. A. Allan. 1990. Mortality response of potato leafhopper (Homoptera: Cicadellidae) to low temperatures: Implications for predicting overwintering and early migrant mortality. J. Econ. Entomol 83: 1541-1548. Stanton, M. L. 1983. Spatial patterns in the plant community and their effects upon insect search, pp. 125-157. In: S. Ahmad (ed.). Herbivorous Insects. Academic Press, New York, NY, U.S.A. 257pp. Steiner, H. M. 1938. Effects of orchard practices on the natural enemies of the white apple leafhopper. J. Econ. Entomol. 31: 232-240. Taylor, R. A. J. 1989. Mystery of the retum migration. In: E. J. Armbrust and W. O. Lamp (ed.). Proceedings of a Symposium: History and Perspectives of Potato Leafhopper (Homoptera: Cicadellidae) Research. Misc. Publications No. 72, Entomol. Soc. America. Taylor, R. A. J., and D. Reling. 1986. Preferred wind direction of long-distance leafhopper {Empoasca fabae) migrants and its relevance to the retum migration of small insects. Anim. Ecol. 55:1103-1114. Tahvanainen, J. O., and R. B. Root. 1972. The influence of vegetational diversity on the population ecology of a specialized herbivore, Phyllotreta cruciferae (Coleoptera: Chrysomelidae). Oecologia (Berlin) 10: 321-346. Thorsteinson, A. J. 1960. Host selection in phytophagous insects. Ann. Rev. Entomol. 5: 193- 218. Vandermeer, J. H. 1989. The Ecology of Intercropping. Cambridge University Press, Cambridge, England, U. K. 237 pp. van Emden, H. F., & G. F. Williams. 1974. Insect stability and diversity in agro-ecosystems. Annu. Rev. Entomol. 19: 455-475. 120 Waloff, N. 1980. Studies of grassland leafhoppers (Auchenorrhyncha; Homoptera) and their natural enemies, pp. 82-215. In: A. MacFadyen (ed.). Advances in Ecological Research. Vol. II. Academic press, London, England, U. K. Washbum, F. L. 1908. The apple leaf hopper and other injurious insects of 1907 and 1908. Bulletin No. 112. Division of Entomology. Agricultural Experiment Station. University of Minnesota. St. Paul, MN, U. S. A. Womack, C. L. 1984. Reductions in photosynthesis and transpiration rates of alfalfe caused by the potato leafhopper (Homoptera: Cicadellidae) defoliations. J. Econ. Entomol 77: 508-513. Zar, J. H. 1974. Biostatistical Analysis. Prentice-hall, Inc., Englewood Cliffs, NJ, U. S. A. 620 pp. 121