<%BANNER%>

Multitrophic Interactions among Crapemyrtles, Lagertroemia spp., Crapemyrtle Aphids, Sarucallis kahawaluokalani and Aphi...

Permanent Link: http://ufdc.ufl.edu/UFE0024268/00001

Material Information

Title: Multitrophic Interactions among Crapemyrtles, Lagertroemia spp., Crapemyrtle Aphids, Sarucallis kahawaluokalani and Aphid Predators
Physical Description: 1 online resource (158 p.)
Language: english
Creator: Herbert, John
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: aphid, chrysopidae, coccinellidae, crapemyrtle, tritrophic
Entomology and Nematology -- Dissertations, Academic -- UF
Genre: Entomology and Nematology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Crapemyrtles, Lagerstroemia spp., are one of the most widely planted ornamentals in the southeastern United States, and are popular for their continuous floral beauty throughout the summer months. Since its introduction into the U.S., plant breeders have produced numerous cultivars that vary in plant parentage, disease resistance and mature plant height. One of the most conspicuous and widespread pests of crapemyrtle is the crapemyrtle aphid, Sarucallis kahawaluokalani (Kirkaldy 1907), and S. kahawaluokalani can be found worldwide in association with crapemyrtle. Although crapemyrtle aphids are a pest from an ornamental perspective, they are host specific and in the U.S., crapemyrtle aphids do not attack plants outside of the genus Lagerstroemia. In Florida, crapemyrtle is a potential augmentation crop for pecans because crapemyrtle aphids reach peak populations 1-3 wk before the peak populations of pecan aphids. Crapemyrtle aphids can serve as alternative prey to attract or sustain aphid predators near pecan orchards. Several experiments were conducted to understand how various attributes of crapemyrtle cultivars affect crapemyrtle aphids and aphid natural enemies at the third trophic level. Crapemyrtle host suitability and aphid host preference experiments were conducted using no-choice and choice experiments in the laboratory. Host suitability or aphid host preference studies did not explain why suitable and preferred hosts in the laboratory were showing resistance in the field, which may be caused by interactions with natural enemies. To test if natural enemies were affected by crapemyrtle cultivar, experiments were conducted on the green lacewing, Chrysoperla rufilabris (Burmeister), and the lady beetle, Harmonia axyridis (Pallas). Larvae of C. rufilabris were fed crapemyrtle aphids that were reared on different crapemyrtle cultivars in ad libitum or calorie-restricted diets. A study to assess the abundance and response of H. axyridis to crapemyrtle aphid populations was conducted in a large plot of crapemyrtle. Plants were monitored for crapemyrtle aphids and H. axyridis populations. Sticky traps were also used to monitor the movement and distribution of H. axyridis. Results from laboratory experiments in this study indicate that crapemyrtle aphid lifetime fecundity and host preference varied according to crapemyrtle cultivar, plant parentage, and mature plant height. Chrysoperla rufilabris larvae fed aphids ad libitum had differences in survivorship that were associated with crapemyrtle cultivar that the aphid prey were reared upon. Adult dry mass and larval development from ad libitum and the restricted diet experiments differed according to cultivar, plant parentage and mature plant height. Field experiments indicated that H. axyridis had a numerical response that was density dependent to the presence of crapemyrtle aphids, but the scale of this response was not on a plant by plant basis. However, when aphid numbers were high and the distribution of aphids was clustered, the distribution of H. axyridis was clustered and associated with the distribution of crapemyrtle aphids more frequently. Host suitability and host preference results are useful for plant breeders wishing to create aphid-resistant cultivars, and results from the experiments conducted using insect predators are useful to biological control programs interested in using crapemyrtle as an augmentative crop.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by John Herbert.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Mizell, Russell F.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024268:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024268/00001

Material Information

Title: Multitrophic Interactions among Crapemyrtles, Lagertroemia spp., Crapemyrtle Aphids, Sarucallis kahawaluokalani and Aphid Predators
Physical Description: 1 online resource (158 p.)
Language: english
Creator: Herbert, John
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: aphid, chrysopidae, coccinellidae, crapemyrtle, tritrophic
Entomology and Nematology -- Dissertations, Academic -- UF
Genre: Entomology and Nematology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Crapemyrtles, Lagerstroemia spp., are one of the most widely planted ornamentals in the southeastern United States, and are popular for their continuous floral beauty throughout the summer months. Since its introduction into the U.S., plant breeders have produced numerous cultivars that vary in plant parentage, disease resistance and mature plant height. One of the most conspicuous and widespread pests of crapemyrtle is the crapemyrtle aphid, Sarucallis kahawaluokalani (Kirkaldy 1907), and S. kahawaluokalani can be found worldwide in association with crapemyrtle. Although crapemyrtle aphids are a pest from an ornamental perspective, they are host specific and in the U.S., crapemyrtle aphids do not attack plants outside of the genus Lagerstroemia. In Florida, crapemyrtle is a potential augmentation crop for pecans because crapemyrtle aphids reach peak populations 1-3 wk before the peak populations of pecan aphids. Crapemyrtle aphids can serve as alternative prey to attract or sustain aphid predators near pecan orchards. Several experiments were conducted to understand how various attributes of crapemyrtle cultivars affect crapemyrtle aphids and aphid natural enemies at the third trophic level. Crapemyrtle host suitability and aphid host preference experiments were conducted using no-choice and choice experiments in the laboratory. Host suitability or aphid host preference studies did not explain why suitable and preferred hosts in the laboratory were showing resistance in the field, which may be caused by interactions with natural enemies. To test if natural enemies were affected by crapemyrtle cultivar, experiments were conducted on the green lacewing, Chrysoperla rufilabris (Burmeister), and the lady beetle, Harmonia axyridis (Pallas). Larvae of C. rufilabris were fed crapemyrtle aphids that were reared on different crapemyrtle cultivars in ad libitum or calorie-restricted diets. A study to assess the abundance and response of H. axyridis to crapemyrtle aphid populations was conducted in a large plot of crapemyrtle. Plants were monitored for crapemyrtle aphids and H. axyridis populations. Sticky traps were also used to monitor the movement and distribution of H. axyridis. Results from laboratory experiments in this study indicate that crapemyrtle aphid lifetime fecundity and host preference varied according to crapemyrtle cultivar, plant parentage, and mature plant height. Chrysoperla rufilabris larvae fed aphids ad libitum had differences in survivorship that were associated with crapemyrtle cultivar that the aphid prey were reared upon. Adult dry mass and larval development from ad libitum and the restricted diet experiments differed according to cultivar, plant parentage and mature plant height. Field experiments indicated that H. axyridis had a numerical response that was density dependent to the presence of crapemyrtle aphids, but the scale of this response was not on a plant by plant basis. However, when aphid numbers were high and the distribution of aphids was clustered, the distribution of H. axyridis was clustered and associated with the distribution of crapemyrtle aphids more frequently. Host suitability and host preference results are useful for plant breeders wishing to create aphid-resistant cultivars, and results from the experiments conducted using insect predators are useful to biological control programs interested in using crapemyrtle as an augmentative crop.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by John Herbert.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Mizell, Russell F.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024268:00001


This item has the following downloads:


Full Text

PAGE 1

1 MULTITROPHIC INTERACTIONS AMONG CRAPEMYRTLES, Lagerstroemia spp., CRAPEMYRTLE APHIDS, Sarucallis kahawaluokalani AND APHID PREDATORS By JOHN JOSEPH HERBERT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 John Joseph Herbert

PAGE 3

3 To my wonderful loving parent s David and Margaret Herbert

PAGE 4

4 ACKNOWLEDGMENTS I thank my advisor Dr. Russell F. Mizell III for al l the support he gave me for the past four years. Dr. Mizell has provided advice, guidance, and contributed significan tly to my growth not only as a student, but as a person as well. Without his academic, emotional and monetary support, this research could not have been po ssible. I thank the members of my advisory committee Drs. Gary Knox, Heather McAuslane, a nd J. Howard Frank. My committee provided me with invaluable feedback and construc tive advice on how to im prove and achieve my academic and personal goals. I thank Tobin Nort hfield for helping me learn SADIE and helping with the plot design for the spatial experiment Charles Riddle provided help with managing plots, preparing leaf disks, and sticky trap data collection. Ci hangir Gokalp helped with aphid rearing and spent many nights in the laboratory helping me prepar e leaf disks. Meghan Brennan and George Papageorgio helped a rrange the most appropriate statis tical analyses and aided with writing SAS code. I thank Dr. Karla Addesso for providing editorial advice. I thank Dawn Atchison for all the love support, friendship, companionship, and compassion that she gave me, and my life has been enriched because of her. Leslie Rios is a dear friend who was there to support me whenever it was needed. Frank Wessels and Erin Vrzal taught me how to balance work and play. I th ank my best friends fr om Ohio, Andy and Cindy Simmons, who have been there to support me for the last ten years of my academic career. I thank all of my other friends for contributi ng to my academic and personal growth. I thank my family, especially my parents, for giving me the values, morals, and work ethic to succeed in life. My sister Susan was instru mental in helping me through tough times. I thank my brother Mike, sisters, Ann, a nd Maria, and their families for all their love and support.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .......10 ABSTRACT....................................................................................................................... ............13 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW..............................................................15 Crapemyrtle Aphids............................................................................................................. ...15 Tritrophic Interactions........................................................................................................ ....17 Crapemyrtle as an Augmentation Crop..................................................................................19 Objectives..................................................................................................................... ..........21 2 HOST PREFERENCE OF TH E CRAPEMYRTLE APHID Sarucallis kahawaluokalani (KIRKALDY) (HEMIPTERA: APHIDIDAE) AND HOST SUITABILITY OF CRAPEMYRTLE Lagerstroemia spp. CULTIVARS...........................................................22 Introduction................................................................................................................... ..........22 Materials and Methods.......................................................................................................... .24 No-Choice Experiment....................................................................................................24 Choice Experiment..........................................................................................................26 Statistics..................................................................................................................... ......27 Results........................................................................................................................ .............28 No-Choice Experiment....................................................................................................28 Choice Experiment..........................................................................................................29 Discussion..................................................................................................................... ..........30 No-Choice Experiment....................................................................................................30 Choice Experiment..........................................................................................................31 Aphid Study.................................................................................................................... .32 3 TRITROPHIC INTERACTIONS AM ONG CRAPEMYRTLE, CRAPEMYRTLE APHIDS, AND THE GREEN LACEWING Chrysoperla rufilabris (BURMEISTER)........39 Introduction................................................................................................................... ..........39 Materials and Methods.......................................................................................................... .41 Ad Libitum Experiment....................................................................................................41 Suboptimal Diet Experiment...........................................................................................43 Results........................................................................................................................ .............46 Ad Libitum Experiment....................................................................................................46 Suboptimal Diet Experiment...........................................................................................47

PAGE 6

6 Discussion..................................................................................................................... ..........48 Ad Libitum Experiment....................................................................................................48 Suboptimal Diet Experiment...........................................................................................50 Lacewing Study...............................................................................................................50 4 LANDSCAPE ECOLOGY OF THE CRA PEMYRTLE APHID AND THE APHID PREDATOR Harmonia axyridis (PALLAS).........................................................................64 Introduction................................................................................................................... ..........64 Materials and Methods.......................................................................................................... .66 Experimental Setup.........................................................................................................66 Sampling....................................................................................................................... ...67 Statistics..................................................................................................................... ......68 Results........................................................................................................................ .............72 2007 Data...................................................................................................................... ...72 2008 Data...................................................................................................................... ...76 Discussion..................................................................................................................... ..........79 Habitat Level Effects.......................................................................................................79 Cultivar Level Effects......................................................................................................81 Spatial Relationships.......................................................................................................81 Field Study.................................................................................................................... ...83 5 GENERAL CONCLUSIONS...............................................................................................110 Host Suitability and Host Preference....................................................................................110 Tritrophic Interactions........................................................................................................ ..111 Spatial and Temporal Interactions........................................................................................114 Multitrophic Interactions...................................................................................................... 115 APPENDIX A GIS MAPS....................................................................................................................... .....118 B ACTIVITY OF ADULT VIRGIN OPAROUS CRAPEMYRTLE APHIDS.......................146 Materials and Methods.........................................................................................................1 46 Results and Discussion......................................................................................................... 146 C PERSONAL OBSERVATIONS..........................................................................................148 Field Observations............................................................................................................. ...148 Laboratory Observations of Crapemyrtle Suitability...........................................................148 LIST OF REFERENCES............................................................................................................. 150 BIOGRAPHICAL SKETCH.......................................................................................................158

PAGE 7

7 LIST OF TABLES Table page 2-1 Lagerstroemia cultivars and their attributes used for the construction of a priori hypotheses and contrasts....................................................................................................33 3-1 Analysis of variance results for the ad libitum diet experiment with contrast statements..................................................................................................................... ......52 3-2 Analysis of variance resu lts with contrasts for the dr y mass of crapemyrtle aphids reared on different cultivars of crapemyrtle.......................................................................52 3-3 Regression analyses results for the comparison of aphid dry mass to aphid fresh mass for different crapemyrtle cultivar s using samples of 25 aphids.........................................53 3-4 Analysis of variance results for C. rufilabris in the suboptimal diet experiment..............53 4-1 Crapemyrtle cultivars planted in the bl ock of crapemyrtle used for aphid and lady beetle sampling................................................................................................................ ..84 4-2 Analysis of variance results for the 2007 repeated measur es analysis for aphids per plant.......................................................................................................................... ..........85 4-3 LS means and mean separation for the e ffect of cultivar in the 2007 analysis of variance for aphids per plant..............................................................................................85 4-4 Analysis of variance results for th e 2007 repeated measures analysis for H. axyridis per plant...................................................................................................................... .......85 4-5 Analysis of variance results for th e 2007 repeated measures analysis for H. axyridis trapped in a 24-h period.....................................................................................................86 4-6 Analysis of variance results for th e 2007 repeated measures analysis for H. axyridis in traps for all sampling dates regardless of trapping duration..........................................86 4-7 LS means and mean separation for eff ect of cultivar in th e 2007 analysis of H. axyridis trapped for all sampling dates re gardless of trapping duration............................86 4-8 Probability of spatial clustering for the response variables samp led once per week in 2007........................................................................................................................... .........87 4-9 Probability of spatial clustering for the 2007 data created by summing the number of aphids and H. axyridis sampled on plants adjacent to traps using trap coordinates..........87 4-10 Probability of spatial clustering for H. axyridis trapped in sticky traps over a 3-d period in 2007................................................................................................................. ...88

PAGE 8

8 4-11 Spatial associations for the distributions of H. axyridis per plant and aphids per plant in 2007........................................................................................................................ .......89 4-12 Spatial association of H. axyridis sampled on plants adjacent to sticky traps and H. axyridis trapped in a 24-h period in 2007..........................................................................89 4-13 Spatial association of H. axyridis per plant and aphids per plant using sampling dates separated by 1 wk in 2007.................................................................................................90 4-14 Spatial association of H. axyridis trapped in a 24-h period and aphids using sampling dates separated by 1 wk in 2007........................................................................................90 4-15 Spatial association of H. axyridis trapped in a 24-h period and H. axyridis on plants using sampling dates separated by 1 wk in 2007...............................................................91 4-16 Spatial association of H. axyridis on plants and H. axyridis trapped over a 3-d period before plant sampling in 2007............................................................................................91 4-17 Spatial association of H. axyridis on plants and H. axyridis trapped over a 3-d period after plant sampling in 2007..............................................................................................92 4-18 Analysis of variance results for the 2008 repeated measur es analysis for aphids per plant.......................................................................................................................... ..........92 4-19 Analysis of variance results for th e 2008 repeated measures analysis for H. axyridis per plant...................................................................................................................... .......92 4-20 Analysis of variance results for th e 2008 repeated measures analysis for H. axyridis trapped in a 24-h period.....................................................................................................93 4-21 Analysis of variance results for th e 2008 repeated measures analysis for H. axyridis trapped on all sampling dates, re gardless of trapping duration.........................................93 4-22 LS means and mean separation for the e ffect of cultivar in the 2008 analysis of H. axyridis trapped for all sampling dates, re gardless of trapping duration...........................93 4-23 Probability of spatial clustering for the response variables samp led once per week in 2008........................................................................................................................... .........94 4-24 Probability of spatial clustering for the 2008 data created by summing the number of aphids and H. axyridis sampled on plants adjacent to traps using trap coordinates..........94 4-25 Probability of spatial clustering for H. axyridis sampled in traps on dates other than weekly sampling in 2008...................................................................................................95 4-26 Spatial associations for the distributions of H. axyridis per plant and aphids per plant in 2008........................................................................................................................ .......96

PAGE 9

9 4-27 Spatial associations for the distribution of aphids sa mpled on plants adjacent to sticky traps and H. axyridis trapped in a 24-h period in 2008...........................................96 4-28 Spatial associations of H. axyridis sampled on plants adjacent to sticky traps and H. axyridis trapped in a 24-h period in 2008..........................................................................97 4-29 Spatial associations of H. axyridis per plant and aphids per plant using sampling dates separated by 1 wk in 2008........................................................................................97 4-30 Spatial associations of H. axyridis trapped in a 24-h pe riod and aphids using sampling dates separated by 1 wk in 2008.........................................................................98 4-31 Spatial associations of H. axyridis trapped in a 24-h period and H. axyridis on plants using sampling dates separated by 1 wk in 2008...............................................................98 5-1 Adjusted efficiency ratings and reproductive poten tial of lacewings..............................117

PAGE 10

10 LIST OF FIGURES Figure page 2-1 Mean total fecundity ( SE) for crapemyrtle aphids in the no-choice experiment for 2005........................................................................................................................... .........34 2-2 Mean total fecundity ( SE) for crapemyrtle aphids in the no-choice experiment for 2005 and 2006.................................................................................................................. ..35 2-3 Relationship of aphid total fecundity to longevity for all samples from both years for the no-choice host suitability experiment..........................................................................36 2-4 Mean number of adult crapemyrtle aphids per disk ( SE) and orthogonal contrasts for the choice experiment...................................................................................................37 2-4 Continued.................................................................................................................. .........38 3-1 Sleeve cage attached to a bran ch of crapemyrtle (Natchez)............................................54 3-2 Vials and Petri-plates cont aining lacewing larvae in the s uboptimal diet experiment......54 3-3 Percentage of lacewing larvae that succe ssfully developed from first instar larva to adult in the ad libitum diet experiment with orthogonal contrasts.....................................55 3-4 Percentage of lacewing pupae emerging as adults in the ad libitum diet experiment with orthogonal contrasts...................................................................................................56 3-5 LS mean ( SE) for the square root of larval development for the ad libitum diet experiment with orthogonal contrasts................................................................................57 3-6 LS mean log dry mass ( SE) for adult lacewings in the ad libitum diet experiment with orthogonal contrasts...................................................................................................58 3-7 Mean dry mass ( SE) for aphid samples in the suboptimal diet experiment...................59 3-8 Relationship of dry mass to fresh mass for aphid samples................................................60 3-9 Mean developmental time of lacewings ( SE) for the suboptimal diet experiment.........61 3-10 Mean pupal mass of lacewings ( SE) in the suboptimal diet experiment........................62 3-11 LS mean log adult dry mass ( SE) for lacewings that successfully emerged without deformation in the suboptimal experiment........................................................................63 4-1 Maps of the crapemyrtle pl ot used for insect sampling.....................................................99 4-2 Trap locations of the 100 sticky tr aps used for sampling Coccinellidae.........................100

PAGE 11

11 4-3 Mean number of insects per sample ( SE) for crapemyrtle aphids and H. axyridis sampled on plants for all dates of sampling in 2007........................................................100 4-4 Mean number of insects per sample ( SE) on crapemyrtle plants for crapemyrtle aphids and H. axyridis in 2007........................................................................................101 4-4 Continued.................................................................................................................. .......102 4-5 Mean number of H. axyridis ( SE) per trap day trapped inside and outside the crapemyrtle plot in 2007..................................................................................................103 4-6 Relationship of the square root of H. axyridis per plant and the square root of aphids per plant in 2007.............................................................................................................. 103 4-7 Relationship of the square root of H. axyridis per trap and the square root of aphids per plant in 2007.............................................................................................................. 104 4-8 Relationship of the square root of H. axyridis per trap and H. axyridis per plant in 2007........................................................................................................................... .......104 4-9 Relationship of the proportion of H. axyridis trapped inside the plot and aphids per plant in 2007.................................................................................................................. ..105 4-10 Relationship of the log H. axyridis trapped outside the plot and the log H. axyridis trapped inside the plot in 2007.........................................................................................106 4-11 Mean number of insects per sample ( SE) for crapemyrtle aphids and H. axyridis sampled on plants for all dates of samplings in 2008......................................................106 4-12 Mean number of H. axyridis ( SE) per trap day trapped inside and outside the crapemyrtle plot in 2008..................................................................................................107 4-13 Relationship of H. axyridis per plant and aphids per plant in 2008.................................107 4-14 Relationship of H. axyridis per trap and H. axyridis per plant in 2008...........................108 4-15 Relationship of the square root of H. axyridis trapped outside the plot and the square root of H. axyridis trapped inside the plot in 2008..........................................................108 4-16 Mean number of crapemyrtle aphids per sample ( SE) in 2007 and 2008...................109 4-17 Mean number of H. axyridis trapped on all sticky traps ( SE) in 2007 and 2008.........109 A-1 Distribution of crapemyrtle aphids and H. axyridis on 06-August 2007.........................119 A-2 Distribution of crapemyrtle aphids and H. axyridis on 13-August 2007.........................120 A-3 Distribution of crapemyrtle aphids and H. axyridis on 20-August 2007.........................121

PAGE 12

12 A-4 Distribution of crapemyrtle aphids and H. axyridis on 27-August 2007.........................122 A-5 Distribution of crapemyrtle aphids and H. axyridis on 03-September 2007...................123 A-6 Distribution of crapemyrtle aphids and H. axyridis on 10-September 2007...................124 A-7 Distribution of crapemyrtle aphids and H. axyridis on 17-September 2007...................125 A-8 Distribution of crapemyrtle aphids and H. axyridis on 24-September 2007...................126 A-9 Distribution of crapemyrtle aphids and H. axyridis on 01-October 2007........................127 A-10 Distribution of crapemyrtle aphids and H. axyridis on 08-October 2007........................128 A-11 Distribution of crapemyrtle aphids and H. axyridis on 15-October 2007........................129 A-12 Distribution of crapemyrtle aphids and H. axyridis on 22-October 2007........................130 A-13 Distribution of crapemyrtle aphids and H. axyridis on 28-October 2007........................131 A-14 Distribution of crapemyrtle aphids and H. axyridis on 05-November 2007....................132 A-15 Distribution of crapemyrtle aphids and H. axyridis on 12-November 2007....................133 A-16 Distribution of crapemyrtle aphids and H. axyridis per plant on 02-July 2008...............134 A-17 Distribution of crapemyrtle aphids and H. axyridis per plant on 08-July 2008...............135 A-18 Distribution of crapemyrtle aphids and H. axyridis per plant on 15-July 2008...............136 A-19 Distribution of crapemyrtle aphids and H. axyridis per plant on 22-July 2008...............137 A-20 Distribution of crapemyrtle aphids and H. axyridis on 22-July 2008..............................138 A-21 Distribution of crapemyrtle aphids and H. axyridis on 04-August 2008.........................139 A-22 Distribution of crapemyrtle aphids and H. axyridis on 11-August 2008.........................140 A-23 Distribution of crapemyrtle aphids and H. axyridis per plant on 19-August 2008..........141 A-24 Distribution of crapemyrtle aphids and H. axyridis per plant on 26-August 2008..........142 A-25 Distribution of crapemyrtle aphids and H. axyridis per plant on 03-September 2008....143 A-26 Distribution of crapemyrtle aphids and H. axyridis on 10-September 2008...................144 A-27 Distribution of crapemyrtle aphids and H. axyridis per plant on 16-September 2008....145 B-1 LS mean number of crapemyrtle aphids ( SE) caught in all sticky traps used to measure activity patterns..................................................................................................147

PAGE 13

13 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MULTITROPHIC INTERACTIONS AMONG CRAPEMYRTLES Lagerstroemia spp., CRAPEMYRTLE APHIDS Sarucallis kahawaluokalani AND APHID PREDATORS By John Joseph Herbert May 2009 Chair: Name Russell F. Mizell III Major: Entomology Crapemyrtles, Lagerstroemia spp., are one of the most wide ly planted ornamentals in the southeastern United States, and are popular for their continuous floral beauty throughout the summer months. Since its introdu ction into the U.S., plant br eeders have produced numerous cultivars that vary in plant parentage, disease resistance and mature plant height. One of the most conspicuous and widespread pests of crapemyrtle is the crapemyrtle aphid, Sarucallis kahawaluokalani (Kirkaldy 1907), and S. kahawaluokalani can be found worldwide in association with crapem yrtle. Although crapemyrtle aphids are a pest from an ornamental perspective, they are host specific and in the U.S., crapemyrtle aphids do not attack plants outside of the genus Lagerstroemia In Florida, crapemyrtle is a potential augmentation crop for pecans because crapemyrtle aphids reach peak popul ations 1-3 wk before the peak populations of pecan aphids. Crapemyrtle aphids can serve as alternative prey to at tract or sustain aphid predators near pecan orchards. Several experiments were conducted to understand how various attributes of crapemyrtle cultiv ars affect crapemyrtle aphids a nd aphid natural enemies at the third trophic level. Crapemyrtle host suitabil ity and aphid host prefer ence experiments were conducted using no-choice and choice experiments in the laboratory. Host suitability or aphid host preference studies did not ex plain why suitable and preferre d hosts in the laboratory were

PAGE 14

14 showing resistance in the field, which may be cause d by interactions with natural enemies. To test if natural enemies were affected by crapem yrtle cultivar, experiment s were conducted on the green lacewing, Chrysoperla rufilabris (Burmeister), and the lady beetle, Harmonia axyridis (Pallas). Larvae of C. rufilabris were fed crapemyrtle aphids that were reared on different crapemyrtle cultivars in ad libitum or calorie-restricted diets. A study to assess the abundance and response of H. axyridis to crapemyrtle aphid populations was conducted in a large plot of crapemyrtle. Plants were monitored for crapemyrtle aphids and H. axyridis populations. Sticky traps were also used to monitor the movement and distribution of H. axyridis Results from laboratory experiments in this st udy indicate that crap emyrtle aphid lifetime fecundity and host preference varied according to crapemyrtle cultivar, plant parentage, and mature plant height. Chrysoperla rufilabris larvae fed aphids ad libitum had differences in survivorship that were associated with crapemyrtle cultivar that the aphid prey were reared upon. Adult dry mass and larval development from ad libitum and the restricted diet e xperiments differed according to cultivar, plant parentage and mature plant height. Field experi ments indicated that H. axyridis had a numerical response that was density dependen t to the presence of crapemyrtle aphids, but the scale of this response was not on a plant by pl ant basis. However, when aphid numbers were high and the distribution of aphids was clustered, the distribution of H. axyridis was clustered and associated with the distributi on of crapemyrtle aphids more fr equently. Host suitability and host preference results are useful for plant breeders wishing to create aphid-resistant cultivars, and results from the experiments conducted using inse ct predators are useful to biological control programs interested in using crapemyr tle as an augmentative crop.

PAGE 15

15 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW Crapemyrtle Aphids Nomenclature. The scientific name of the crapemyrtle aphid, Sarucallis kahawaluokalani has followed a convoluted path in th at the genus has been changed several times since its original descrip tion in 1907. The crapemyrtle aphi ds original scientific name Myzocallis kahawaluokalani was assigned by Kirkaldy in 1907, but a specimen was later used by Shinji (1922) as a type specimen for the species Sarucallis lythrae to represent the genus Sarucallis (Quednau 2003). Because the species was previously described, but the genus was incorrect, nomenclature was changed to reflect the new genus and the scientific name became Sarucallis kahawaluokalani (Kirkaldy, 1907). The genus was changed to Tinocallis in 1997 by Remaudiere and Remaudiere, but Quednau (2003 ) concluded that there were sufficient differences between the crapemyrtle a phid and other members of the genus Tinocallis and revalidated the genus Sarucallis The current and most co rrect scientific name is Sarucallis kahawaluokalani (Kirkaldy, 1907). Life cycle. Crapemyrtle aphids have a holocyclic lifecycle that uses sexual and asexual reproduction. Starting from the egg stage, the lifec ycle begins in the spring. Eggs hatch in spring and early summer giving rise to first instar nymphs that immediately begin feeding on the abaxial portion of a crapemyrtle leaf. A nymph emerging from an egg is female in gender and referred to as a fundatrix (Dixon 1973). Crapem yrtle aphid fundatrices develop through four nymphal instars before molting into an adu lt (Alverson and Allen 1991, 1992a). The adult fundatrix reproduces through asexua l reproduction giving live birt h to only female offspring (parthenogenesis). Offspring of the fundatrix and subsequent gene rations of crapemyrtle aphids reproduced through parthenogenesis during the summer months, and referred to as virginoparae

PAGE 16

16 (Dixon 1973, Alverson and Allen 1991, 1992a, Dixon 1998). Virginoparae persist until environmental cues such as photoperiod and temper ature condition the virginoparae to give birth to another form of female offspring know n as a sexupara (Dixon 1973, Alverson and Allen 1991). Sexuparae produce male and female offspr ing that are morphologi cally different than virginoparae or sexuparae. Female offspring of sexuparae are referred to as oviparae, and oviparae will mate with male aphids before layi ng eggs. Oviparae deposit eggs singly or in small clusters around bud scars or in crevices on the bark of crapemyrtle, where the eggs remain dormant throughout the winter months (Alverso n and Allen 1991, 1992a). The following spring, the cycle repeats itself when temperature a nd photoperiod cues signa l the eggs to hatch. Identification. Nymphs of the crapemyrtle aphid are yellow in color and the shade of yellow may vary from an extremely pale whitis h yellow to a dark yellow or light orange. Nymphs of the fundatrix, virginoparae, and sexup arae have large setae that project from the dorsum and marginal areas of the thorax and abdomen. Adult crapemyrtle aphids are yellow mottled with black and bear two large black tuberc les that arise from their dorsum. Unlike other aphid species that produce winged adults in response to environm ental or nutritional cues, all adult crapemyrtle aphids, except oviparae, are fully winged and capable of flight (Dixon 1971, 1973, Alverson and Allen 1991, 1992a, Dixon 1998, Quednau 2003). Oviparae nymphs and adults are morphologically different from all other stages in that they ar e yellow to light orange with black spots and do not bear wing pads as nymphs or wi ngs as adults (Quednau 2003). Males are smaller than females and easily identi fied under a stereomicroscope by the differences with respect to their size, shap e of abdomen, and genitalia. Feeding habit and damage. Crapemyrtle aphids damage crapemyrtle ( Lagerstroemia spp.) cosmetically and are not known to vector pl ant diseases. Crapemyrtle aphid damage is an

PAGE 17

17 indirect result of feeding and doe s not appear to result in perman ent damage or long term effects on plant vigor. Aphids feed on phloem sap of th e plant, which carries the majority of plant sugars and amino acids, and although the nutrient to sugar ratio within the phloem is extremely low, aphids have evolved mechanisms that allo w them to utilize the tr ace amounts of nutrients within the phloem (Dixon 1973, Weibull 1988, D ouglas 1993, Wilkinson and Douglas 2003). Amino acids and other essentia l nutrients are separated with a special filter chamber located in the gut, and excess s ugar and water are excreted from the body in sugary droplets called honeydew (Dixon 1973). Crapemyrtle aphids eject honeydew away from their feeding location preventing them from becoming entangl ed within the sticky secretion. Honeydew accumulates on objects below aphid populations and is commonly seen as a shiny coating on the tops of leaves and stems. Molds and other microorganisms can grow on these surfaces using the rich sugary honeydew as a food source. Black sooty molds in the genus Capnodium grow on honeydew produced by crapemyrtle aphids and can turn an entire crapemyrtle plan t an unsightly black co lor, detracting from the visual aesthetics of cr apemyrtle (Dozier 1926). When aphi d infestations are severe, thick coatings of black sooty mold may interfere with photosynthesis, and it is common for leaves that are covered in sooty mold to drop from the plant. However, defoliation is unlikely to affect long term plant health or vigor, and from persona l observations, plants typically rebound and bloom beautifully in the following years. Tritrophic Interactions Plants are known to affect insect natural enemies through nutritiona l (Giles et al. 2000, Francis et al. 2001, Giles et al. 2002a, Giles et al. 2002b), morphological (Clark and Messina 1998, Legrand and Barbosa 2003), and chemical at tributes (Malcolm 1990, Fuentes-Contreras and Niemeyer 1998). These attributes affect ins ect natural enemies eith er directly or through

PAGE 18

18 multitrophic interactions (Bottrell et al. 1998). Plant attributes th at directly affect predators influence the ability of predators to locate or utilize prey (Van Emden 1995, Bottrell et al. 1998, Brewer and Elliott 2004). Pea plan t leaf morphology directly influe nces the foraging ability of the lady beetles Coccinella septempunctata L. and Hippodamia convergens Gurin-Mneville. Plants with tendrils instead of normal leafle ts increased the searching efficiency of C. septempunctata and H. convergens which resulted in higher pred ation rates of aphids on pea plants with tendrils (Kareiva and Sahakian 1990) Morphological complexity of a plant also influences foraging success of lady beetle pr edators, and morphologically complex plants decrease efficacy of C. septempunctata due to a larger search area (Legrand and Barbosa 2003). In addition to directly affecting predator fora ging success and fitness, plants affect aphid predators indirectly through multitro phic interactions via the aphid prey. Aphids are exposed to nutritional and allelochemical c onstituents of phloem sap during feeding and may sequester these compounds and store them within their tissues (Rothschild et al. 1970, Duffey 1980,). Specific compounds found within an aphids tissues or h oneydew are dependent on the host plant species or cultivar that the aphid is feeding upon (Duffey 1980, Malcolm 1990). Aphis nerii Boyer de Fonscolombe feeds on both oleander and milkweed acquiring different toxins (cardenolides) from each plant (Rothschild et al. 1970). Nutri tional or toxic constituents of phloem sap can alter an aphids suitability to serve as prey and a ffect the fitness or life hi story characteristics of aphid natural enemies (Rothschild et al. 1970). A phid prey that differ in suitability can affect longevity, survivorship, size, weig ht (larva, pupa or adul t), or fecundity of aphid natural enemies (Okamoto 1966, Hodek 1993, Van Emden 1995, Hauge et al. 1998, Giles et al. 2000, Francis et al. 2001, Giles et al. 2001, Giles et al. 2002a, Gi les et al. 2002b, Brewer and Elliott 2004).

PAGE 19

19 Predator fitness is affected through either qua ntitative or qualitative differences in prey suitability. Quantitative differences in prey su itability are dependent on the number of prey consumed and are cumulative, whereas qualitative differences are fundame ntal differences and are independent of the number of prey eaten. Qualitative or quantitativ e differences in prey suitability are resolved by using diets that diffe r in the amount of prey (Giles et al. 2000). Quantitative differences exist when a predators f ecundity or survivorship is influenced by the number of prey consumed (Giles et al. 2000, Giles et al. 2001, Giles et al 2002a). Predators are capable of overcoming quantitative differences by varying the number of prey eaten. Qualitative differences in prey suitability exist when na tural enemies show a difference in fitness or survivorship among all levels of daily prey (Giles et al. 2002b). Differences in prey suitability are further complicated because a single prey speci es can differ qualitatively for natural enemy X and quantitatively for natural enem y Y. Giles et al. (2000, 2001) fed pea aphids reared on faba bean or alfalfa to C. rufilabris H. convergens and Coleomegilla maculata (DeGeer) and found that differences in prey suitability of pea aphids for C. rufilabris were qualitative, but differences in prey suitability of pea aphids for H. convergens and C. maculata were quantitative. Aphid natural enemies (e.g., C. septempunctata ) are not always capable of distinguishing between suitable and unsuitable prey, and conservation bi ological control programs often use cover or augmentation crops to provide alternative pr ey to aphid predator s (Hodek 1956, Blackman 1967, Hauge et al. 1998). When provi ding cover or augmentation crops in agroecosystems for conservation biological control programs, it is important to understand the multitrophic interactions among plants, herbivores, and predators. Crapemyrtle as an Augmentation Crop Yellow pecan aphid, Monelliopsis pecanis Bissell, black pecan aphid, Melanocallis carvaefoliae (Davis), and blackmargined aphid, Monellia caryella (Fitch), damage pecan leaves

PAGE 20

20 directly through feeding activity causing economic injury (T edders et al. 1982, Wood et al. 1982). Use of insecticidal sprays in pecans produces resurgence of pecan aphid populations resulting from a combination of insecticidal re sistance and aphid natural enemy susceptibility to insecticides (Dutcher 1985, Dutc her and Htay 1985). Developing a biological control program for pecans will reduce the dependence on chemical insecticides, decrease the incidence of pecan aphid resurgence, and promote natural contro l of pecan aphids (Tedders 1983, Mizell and Schiffhauer 1987b, 1989). Pecan aphid natural enemies overwinter within pecan orchards and begin to emerge or become active in early spring (Mizell and Sc hiffhauer 1987a). Tedders (1983) suggested intercropping clover and vetch that harbor the pea aphids, Acyrthosiphon pisum (Harris), and Aphis craccivora Kotch, to provide alternative aphid pr ey for pecan aphid natural enemies. However, populations of yellow pecan aphid an d blackmargined aphid crash during mid summer and clover and vetch do not subsist through th is period (Tedders et al. 1982, Dutcher 1985, Dutcher and Htay 1985, Mizell and Schiffhauer 1987b) Additional biological control measures or techniques are needed during th is period to retain natural enem ies and promote natural control of pecan aphids. Mizell and Schiffhauer (1987b) recommend the use of crapemyrtle as an augmentation crop for pecans because most cultivars of crapem yrtle are attacked by the crapemyrtle aphid, and S. kahawaluokalani is host specific in the US at tacking plants within the genus Lagerstroemia (Alverson and Allen 1991, Alverson and Allen 1992a). Populations of S. kahawaluokalani in Florida peak twice per year once in late July-August and ag ain in late September-November and occur during the seasonal dip in pecan aphid pop ulations (Mizell and Schiffhauer 1987b). Mizell

PAGE 21

21 and Schiffhauer (1987b) collected pecan aphi d predators from crapemyrtle foliage and documented that these predators feed on crapemyrtle aphids. Objectives The objectives of this research were to unde rstand the interactions among crapemyrtles, crapemyrtle aphids, and aphid natural enemies. The first objective seeks to benefit the ornamental industry by identifying cultivars that ar e resistant to aphid a ttack. In addition to analyzing individual cultivars, information rega rding other plant attributes such as plant parentage and mature plant height may be useful to plant breeders that are interested in developing aphid resistant cultiv ars. To complete the first objective, experiments were conducted to assess the host suitability of crapem yrtle cultivars in a no-choice experiment and the host preference of the crapemyrtl e aphid in a choice experiment. The second objective of this research was to e xplore the potential of using crapemyrtle as an augmentation crop for pecans. Testing cultivar s for their effect on aphid predators, either directly or indirectly, may help with the selection of the most a ppropriate cultivar for use as an augmentation crop. Plant attributes such as plant parentage, and mature pl ant height could also play a role in tritrophic interactions, allo wing a broader range of cultivars to be used. Experiments assessing the potential of crapem yrtle to serve as an augmentation crop were conducted on the aphid predators C. rufilabris and H. axyridis Experiments using C. rufilabris investigated the effect of consuming aphids that were reared on different crapemyrtle cultivars on the life history charac teristics and fitness of C. rufilabris larvae. A field experiment was conducted in a large plot of crap emyrtle at the North Florida Res earch and Education Center to assess the spatial and temporal response of H. axyridis to crapemyrtle aphid populations.

PAGE 22

22 CHAPTER 2 HOST PREFERENCE OF TH E CRAPEMYRTLE APHID Sarucallis kahawaluokalani (KIRKALDY) (HEMIPTERA: APHIDIDAE) AND HOST SUITABILITY OF CRAPEMYRTLE Lagerstroemia spp. CULTIVARS Introduction Crapemyrtles are one of the most popular orna mental plants throughout the southeastern U.S. and were introduced for their floral beauty and horticultural properti es. Unfortunately, the cultivars of crapemyrtle that were originally brought to the U. S. suffered from powdery mildew, Erisyphe lagerstroemiae E. West and cercospora leaf spot, Cercospora lythracearum Heald & Wolf (Dix 1999). In the U.S., resistance to powdery mildew was established in crapemyrtle cultivars through a breeding program at th e National Arboretum in Washington, DC. Hybridizing L. fauriei Koehne with L. indica L. produced several pow dery mildew-resistant cultivars, but insect resistance was not cons idered during the selec tion process (Egolf 1981a, 1981b, 1986a, 1986b, 1987a, 1987b, 1990b) Studies have indicated that L. fauriei parentage can affect the susceptibility of cr apemyrtle to insect herbivores (Mizell and Knox 1993, Pettis et al. 2004). Lagerstroemia fauriei parentage has at leas t two and possibly three separate sources of germplasm, which may also influence in sect resistance. The first source of L. fauriei germplasm came from seeds collected in a mountain range near Kurio, Yakushima Japan, and this source was used to produce more than 19 powdery mild ew-resistant cultivars that are among the most popular and widely grown in the U.S. (Egolf 1981a, 1981b, 1986a, 1986b, 1987a, 1987b, 1990a, 1990b, Pooler and Dix 1999, Pooler 2003). Another source of L. fauriei germplasm is from a cultivar known as Bashams Party Pink that was discovered in Texas and was not produced through a plant breeding program. Bashams Party Pink is believe d to be a natural hybridization of L. indica and L. fauriei but its germplasm may be re lated to the seedling source

PAGE 23

23 of germplasm because it was discovered after th e seedlings were brought to the U.S. (Pooler 2003). The final source of L. fauriei germplasm present in L. indica x L. fauriei cultivars is from L. fauriei cuttings taken along a river in Yakushima (Japan) (Pooler and Dix 1999, Pooler 2003). This lineage is separate from the seedlings or Bashams Party Pink, and Apalachee is currently the only widely available cultivar from this source of germ plasm (Pooler and Dix 1999, Pooler 2003). Crapemyrtle aphid damage is most noticeable as a black sooty mold, Capnodium sp., that grows on plant surfaces that have been coated with aphid honeyde w. Thick coatings of sooty mold detract from the visual aesthetics of crap emyrtle by turning the plant black and can cause early leaf drop or complete defoliation (Doz ier 1926). Studies on the susceptibility of crapemyrtles to aphid attack indi cated that there are differences among cultivars, plant parentage (pure L. indica vs. L. indica L. fauriei ), and mature plant height (dwarf, medium, or tall) (Alverson and Allen 1992b, Mizell and Knox 1993). Mizell and Knox (1993) demonstrated that pure L. indica cultivars were more resistant to crapemyrtle aphid attack than L. indica L. fauriei hybrids and that dwarf plants were more resi stant to aphids than medium or tall. Differences in susceptibility may be attributed to several factors, including, but not limited to, host plant suitability, aphid host preference or aphid host findi ng. Finding and utilizing a host may be a challenge for aphids because they te nd to be host specific and have poor control over their flight path (Dixon 1998, Powell et al. 2006). Even if crapemyrtle aphids locate a suitable species or cultivar, the plant may have antibiotic or antixenotic mechanisms that help it resist aphid attack (Painter 1968, Kogan and Ortman 1978, Panda and Khush 1995, Hill et al. 2004). These mechanisms are defined as plant attributes but insect attributes like host preference or host searching behavior can also in fluence the susceptibility of a pl ant to insect attack. In this

PAGE 24

24 study, the word preference is used as an aphid attribute under choice conditions, suitability is used as a synonym of antibiosis; a plant attribute that affects an insects health or reproduction, and susceptibility as the population of aphids that develop on a plant under field conditions (Singer 2000, Powell et al. 2006). The purpose of this study was to investigate the host suitability of different crapemyrtle cultivars and the host preference of the crapemyrtle aphid. The study also evaluated attributes that are shared by more than one cultivar of crapemyrtle like plant parentage, source of L. fauriei germplasm, and mature plant height. Materials and Methods Crapemyrtle. Seven cultivars of crapemyrtle were selected based on their parentage, source of germplasm, mature plant height, and availability, as follows: Carolina Beauty, Byers Wonderful White, Lipan, Tuscarora Apalachee, Natchez, and Sioux. The selected cultivars represent the two major parent ages and a medium (4-6 m) and tall (> 7 m) cultivar from three of the four sources of ge rmplasm (Table 2-1.). Apalachee is the only cultivar commercially produced from the L. indica L. fauriei (cuttings) line of germplasm, and thus, is the only representative from this source of germplasm. All seven crapemyrtle cultivars were available in a 5-yr-old planting at the North Florida Research and Education Center, Quincy, FL. No-Choice Experiment Agar Petri-plates were prepared following the procedures outlined in Reilly and Tedders (1990). We combined 6 g of granular agar (F isher Scientific Atlant a, GA) and 500 ml of distilled water in a 1-liter autoclave bottle, auto claved the mixture for 20 min, and poured 5 ml of agar into each 10 40 mm Petri-plate. After the agar mixture solidified, the plates were stacked and stored agar side up.

PAGE 25

25 Leaf disks were prepared by cutting one 40-mm disk from each leaf. Disks from the same plant were washed in cold running tap water for 10 min. After washing, the disks were disinfected by placing them in a 1: 3 bleach:water solution for 2 min and rinsed four times with sterilized deionized wa ter (Reilly and Tedders 1990). One disk was placed on each 10 40 mm plate with the adaxial side of the leaf in contact with the agar. The plates were inverted allowing the aphids to feed on the underside of the leaf in their normal inverted orientation. Four aphid colonies were used in the no-c hoice experiment, and each colony was reared on a separate crapemyrtle cultivar. The procedure for starting the Apalachee colony was as follows: 1) a single adult virginoparous aphid (foundress) was collected from an Apalachee plant in the field, 2) the aphid was brought to the laboratory and placed on a 40-mm Apalachee leaf disk, 3) the foundress, her progeny, and s ubsequent generations were reared on 40-mm Apalachee leaf disks on agar plates. To pr oduce four unique colonies, the procedure above was repeated for the cultivars Natchez, Byers Wond erful White, and Tuscarora. Colonies were started with the procedures desc ribed here to test if the foundr ess selected the most suitable cultivar for her offspring and their descendants. To start the no-choice experime nt, leaf material was collected from crapemyrtle plants in the field. Four plants from C arolina Beauty, Byers Wonderful White, Lipan, Tuscarora, Apalachee, Natchez, and Sioux were used in the experiment. Four leaves from each of the 28 plants were removed from the plant, placed in a plastic bag (1 bag per plant), and stored in a cooler that contained ice p acks until reaching the laborator y. Leaves were chosen by haphazardly selecting a branch, identifying the thir d pair of fully expanded leaves from the branch terminal, and removing a single leaf from the pair. Four leaves per plant were selected from each plant used in the experiment. In the laboratory, one 40-mm leaf disk was removed

PAGE 26

26 from each leaf and prepared as outlined in the pr ocedures above. Leaf disks were labeled with the cultivar and plant of origi n. The four leaf disks that came from the same plant were randomly assigned to one of the four aphid colo nies. One adult virginoparous aphid that was from the appropriate aphid colony and less than 24 h post adult emergence was placed on each disk. Petri-plates containing l eaf disks were kept at 21 2 C on a laboratory bench under fluorescent lighting that maintained a 14:10 light:d ark photoperiod. Light fi xtures contained two 40 W tubes that were 0.75 m from the Petri dishes. Fecundity was recorded daily and defined as the number of new nymphs produced within 24 h. New nymphs were gently removed using a small vacuum pump fitted with plastic tubing and a pipette. Th is method of removal was chosen because it caused the least amount of disturbance to the adult virginoparous aphid. In 2006, the no-choice experiment was repeate d, but because of results obtained in 2005, aphid colony was not used as a factor. The 2006 procedures differed in two ways: 1) the disks were placed in a growth chamber at 22 C with a 14:10 light: dark photoperiod, 2) the adult virginoparous aphid was moved to a fresh leaf disk every 7 d. Choice Experiment In addition to the seven cultivars used in the no-choice experiment the aphid preference choice experiment included an eighth cultivar of crapemyrtle, Lagerstroemia speciosa L. Lagerstroemia speciosa acted as a negative control because it is one of the few species of Lagerstroemia that is not a suitable for the growth and development of the crapemyrtle aphid. All aphids used in the choice e xperiment were descendants of a single aphid that was reared on a cultivar not tested (Tonto). The choice arena was constructed from a 20 4 cm agar plate. Leaf disks were prepared by rins ing with water to remove forei gn particles, but mold was not a concern during this short duration and the leaf disks were not disi nfected before their use in the

PAGE 27

27 experiment. Disks were placed on th e plates as in the previous experiment with the adaxial side of the leaf in contact with the agar. Each aren a had one leaf disk from each cultivar, and the disks were randomly arranged around the outer edge of the Petri-plate. The experiment used eight arenas and each arena received 24 adult virgi noparous aphids. Four ar enas were placed in a growth chamber at 22 C with a 14:10 light:dark photoperiod and the other four arenas were placed in the growth chamber in complete darkness. Complete darkness was tested to investigate if crapemyrtle aphid movement or host discriminati on might differ when visual cues were absent. Arenas were checked once per day for 3 d and the number of adults and nymphs on each disk was recorded. Statistics The experimental design for the 2005 leaf disk experiment was a split-plot design with crapemyrtle cultivar as the main plot factor and aphid clone as the subplot factor. Thus, cultivar is replicated by plant(cultivar) whil e aphid clone is replicated by disk(plant cultivar). Daily fecundity data were analyzed using repeated measures analysis PROC MIXED SAS 9.1.3 (SAS Institute 2000-2004). Total fec undity data from 2005 and 2006 were combined and analyzed using PROC GLM SAS 9.1.3 (SAS institute 20002004). Because both years used the same plants, the analysis used plant(cultivar) as the e rror term for cultivar and plant(cultivar year) as the error term for year and year cultivar. Means of significant factors were separated using Tukeys HSD (SAS Institute 2000-2004). A regr ession analysis on aphid total fecundity and aphid longevity was performed in SAS usi ng PROC REG SAS 9.1.3 (SAS Institute 2000-2004). Choice data were analyzed using PROC GL IMMIX SAS 9.1.3 (SAS Institute 2000-2004). Data were square root transformed and anal yzed using a repeated measures analysis.

PAGE 28

28 Lagerstroemia speciosa data were removed from the analys is because no adults were observed on the leaf disks at any time and only two nymphs were ever observed on a L. speciosa disk. Both experiments used a priori orthogonal contrasts to comp are the following crapemyrtle attributes: 1) plant parentage: pure L. indica cultivars vs. L. indica L. fauriei hybrids; 2) source of L. fauriei germplasm: cuttings vs. seedlings an d Bashams Party Pink 3) source of L. fauriei germplasm separate from Apalachee: U.S. B ashams Party Pink vs. seedlings; 4) mature plant height within the pure L. indica cultivars: medium (Carol ina Beauty) vs. tall (Byers Wonderful White); 5) mature plant height within the L. fauriei seedlings germplasm: medium (Sioux) vs. tall (Natchez); 6) mature plan t height within the Bashams Party Pink germplasm: medium (Lipan) vs. tall (Tuscarora). Results No-Choice Experiment Repeated measures analysis for the 2005 data did not indicate that daily fecundity was different among crapemyrtle cultivars or the four aphid colonies, but crapemyrtle aphid total fecundity was different among crapemyrtle cultiv ars (F = 6.24; df = 6, 21; P < 0.001; Figure 21). Total fecundity was not found to be different for the aphid colonies and there was no cultivar colony interaction. Repeated measures anal ysis from 2006 showed the same non-significance for daily fecundity as the test in 2005. To test for differences between ye ars and to test if the effect of cultivar was the same for both years, the total fecundity data from 2005 and 2006 were combined. Total fecundity data for the combined dataset from 2005 and 2006 showed differences for year (F = 50.71; df = 1, 21; P < 0.0001; Figure 2-2A) and cult ivar (F = 4.89; df = 6, 21; P = 0.0028, Figure 2-2B). Because there was no interaction between year and cultivar, the a priori contrasts were applied to the combined data set from both years. Crapemyrtle aphids reared on pure L. indica cultivars had a lower total fecundity than aphids reared on the cultivars

PAGE 29

29 that were L. indica L. fauriei hybrids (F = 17.3; df = 1, 21; P < 0.001, Figure 2-2B). Source of L. indica L. fauriei germplasm did not significantly affect aphid total fecundity, but there were differences in aphid total fecundity associated wi th mature plant height. Aphid total fecundity differed according to mature plant height within the pure L. indica source of germplasm and aphids reared on Byers Wonderf ul White had a higher total fec undity than aphids reared on Carolina Beauty (F = 7. 13; df = 1, 21; P = 0.014; Figure 2-2B). In the Bashams Party Pink source of germplasm aphids reared on Lipan had a greater total fecundity th an aphids reared on Tuscarora (F = 4.31; df = 1, 21; P = 0.05; Fi g 2-2B). Using all cultivars and both years, crapemyrtle aphid total fecundity was found to be positively and linearly correlated with adult longevity (Figure 2-3; R2 = 0.86). Choice Experiment The number of adult aphids that settled on a leaf disk was different among the crapemyrtle cultivars (F = 5.79; df = 6, 36; P = 0.0003), but th ere was a significant interaction between the main effects of cultivar and photoperiod (F = 2.54; df = 6, 36; P = 0.0375). An interaction between cultivar and photoperiod indicates that the effect of cultivar was not the same for arenas kept under a 14:10 light:dark photoper iod and the arenas kept in co mplete darkness. Therefore, the a priori contrast statements were evaluated sepa rately for arenas that received a 14:10 light:dark photoperiod and arenas kept in complete darkness. Data from the arenas kept at a 14:10 light:dark photoperiod showed diffe rences between the two parentages L. indica and L. indica L. fauriei (F = 4.64; df = 1, 18; P = 0.0451; Figure 2-4A) and between the L. fauriei Bashams Party Pink and L. fauriei seedlings sources of germpl asm (F = 10.65; df = 1,18; P = 0.0043; Figure 2-4A). The 14:10 light:dark photoperi od data also showed differences for mature plant height for the cultivars within the pure L. indica parentage (Byers Wonderful White and Carolina Beauty) (F = 8.01; df = 1, 18; P = 0.0111; Figure 2-4A) and the L. fauriei seedlings

PAGE 30

30 source of germplasm (Sioux and Natchez) (F = 5.87; df = 1, 18; P = 0.0262; Figure 2-4A). Data analysis for arenas kept in complete dar kness indicated that there was a difference between the L. fauriei cuttings source of germplasm and the grouping of the L. fauriei seedlings and Bashams Party Pink sources of germplasm (F = 4.81; df = 1, 18; P = 0.0417; Figure 2-4B). There was also a difference between medium and tall mature plant heights within the L. indica parentage and more aphids were found on Byers Wonderful White than Carolina Beauty (F= 11.25; df = 1, 18; P = 0.0035; Figure 2-4B). The number of nymphs on a disk was correlated with the number of adults on that leaf disk for arenas that were ke pt at a 14:10 light:dark photoperiod (y = 0.132 + 3.43x; R2 = 0.61) and arenas kept in complete darkness (y = 0.392 + 3.04x; R2 = 0.73), where y = (square root of nymphs) and x = (square root of adults) (data not shown). Discussion No-Choice Experiment Previous experiments investiga ting the host suitability of crapemyrtle cultivars focused on Carolina Beauty and Natchez as representatives of pure L. indica and L. indica L. fauriei hybrids, respectively (Alverson and Allen 1992a, 1992b). Data from this study agree with the results of Alverson and Allen (1992 b) in that total fecundity was different for aphids reared on the cultivars Natchez and Carol ina Beauty, but daily fecundity was not different for aphids reared on Natchez and C arolina Beauty. Results from the host suitability experiment expands the knowledge of how crapemyrtle cultivar affects crapemyrtle aphid fecundity for an additional pure L. indica cultivar and four additional L. indica L fauriei cultivars. Data from the host suitability experiment are also in agreement with host susceptibility data presented in Mizell and Knox (1993). Mizell and Knox (1993) found higher aphid populations on L. indica L. fauriei cultivars than cultivars of pure L. indica decent.

PAGE 31

31 Mature plant height is another plant attri bute that may be impo rtant for considering crapemyrtle aphid resistance. Mizell and Knox (1993) found that cultivars with a tall mature plant height were more susceptibl e than plants that had a medium mature plant height. The host suitability experiment detected differences in total fecundity when aphids were reared on cultivars with a different mature plant height, but the effect of mature pl ant height was different between pure L. indica cultivars and L. indica L. fauriei hybrids. Theref ore, mature plant height may be more important to suitability wh en it is considered in the context of plant parentage or source of germplasm. During the no-choice experiment, virginoparous aphids were rarely found dead with their stylets inserted into the plant tissue; and most aphids, found dead were on the agar or the bottom of the Petri-plate. Cultivar su itability measured in this study ma y differ due to rejection of the leaf disk and it is unknown whether this rejection is caused by antibiosis or antixenosis. It is known that many aphid specialists can use pl ant secondary metabolite s for host selection, feeding, or parturition stimulants, but overall f itness of an aphid is attributed to host plant nutritional quality (Fragoyiannis et al. 1998, Powell et al. 2006). Data from this study cannot determine if nutritional or chemical attributes significantly contributed to the rejection of a leaf disk, but because the rejection took place afte r several days of feeding and reproduction, nutritional or chemical constituents of the phloem are logical factors to consider for future work. Choice Experiment Crapemyrtle aphid host plant preference was as sociated with host plant suitability, and both experiments found fewer aphids on the pure L. indica cultivars in comparison to the hybridized L. indica L. fauriei cultivars. The only noticeable effect of complete darkness was the lower number of aphids seen on the cultiv ars Lipan and Tuscarora, which are derived from the same source of L. fauriei germplasm. It is unknown if complete darkness differentially

PAGE 32

32 affected plant chemistry or simply suppresse d aphid movement, but th e higher correlation of nymphs to adults for the arenas in complete darkness suggests th at aphid nymphs are less likely to move under dark conditions. Aphid Study Data from this study may be useful for cr apemyrtle plant breeding programs with the objective to produce aphid resistant cultivars. The most popular and widely available crapemyrtle cultivars are the hybridized L. indica L. fauriei cultivars, and their popularity is attributed to powdery mildew resistance. Unfo rtunately, these cultivars appear to be more suitable, more susceptible, and less resistant to a phid attack, and extensive use of these cultivars in urban landscapes could lead to more aphid problems in the future.

PAGE 33

33 Table 2-1. Lagerstroemia cultivars and their attributes used for the construction of a priori hypotheses and contrasts Cultivar Mature height Parentage Germplasm source Carolina Beauty Medium L. indica Pure L. indica Byers Wonderful White Tall L. indica Pure L. indica Sioux Medium L. indica L. fauriei L. fauriei ; seedlings Natchez Tall L. indica L. fauriei L. fauriei ; seedlings Lipan Medium L. indica L. fauriei L. fauriei ; Bashams Party Pink Tuscarora Tall L. indica L. fauriei L. fauriei ; Bashams Party Pink Apalachee Medium L. indica L. fauriei L. fauriei ; Cuttings

PAGE 34

34 Cultivar CBBWApaSioNatLipTus Total Fecundity 0 20 40 60 80 100 120 140 a a a a a a b N =16 Figure 2-1. Mean total fecundity ( SE) for crapemyrtle aphids in the no-choice experiment for 2005 using the cultivars: Carolina Beauty (CB), Byers Wonderful White (BW), Apalachee (Apa), Sioux (Sio),Natchez (Nat), Lipan (Lip), and Tuscarora (Tus). Columns with differe nt letters are different at P < 0.05 using Tukeys HSD (SAS Institute 2000-2004).

PAGE 35

35 CBBWApaSioNatLipTus Total Fecundity 0 20 40 60 80 100 120 140 20052006 0 20 40 60 80 100 120 140 N=112 a b N=32 A B Med Tall Med TallMed Tall Seedlings BPP Cuttings Seedlings; BPP L. indica L. indica L. fauriei ** ** Figure 2-2. Mean total fecundity ( SE) for crapemyrtle aphids in the no-choice experiment for 2005 and 2006 comparing the effects of (A) year (2005 and 2006) and (B) cultivar (Carolina Beauty (CB), Byers Wonder ful White (BW), Apalachee (Apa), Sioux (Sio),Natchez (N at), Lipan (Lip), and Tuscarora (Tus)) with a priori orthogonal contrasts. *P < 0.05 **P < 0.01.

PAGE 36

36 Longevity in days 05101520 Total fecundity 0 20 40 60 80 100 120 140 160 y = -0.1246 + 6.42x R 2 = 0.86 Figure 2-3. Relationship of aphid total fecundity to longevity for all samples from both years for the no-choice host suitability experiment.

PAGE 37

37 A LspCBBWApaSioNatLipTus Aphids per disk 0 2 4 6 8 Med Tall Med TallMed Tall Seedlings BPP Cuttings Seedlings; BPP L. indica L. indica L. fauriei* ** *N=4 Figure 2-4. Mean number of adult crapemyrtle aphids per disk ( SE) and orthogonal contrasts for the choice experiment for the treatm ents of (A) 14:10 light:dark photoperiod and (B) complete darkness. The cultivars used were as follows: Lagerstroemia speciosa (Lsp), Carolina Beauty ( CB), Byers Wonderful White (BW), Apalachee (Apa), Sioux (Sio),Natchez (Nat), Lipan (Lip), and Tuscaror a (Tus). *P < 0.05 **P < 0.01.

PAGE 38

38 B LspCBBWApaSioNatLipTus Aphids per disk 0 2 4 6 8 Med Tall Med TallMed Tall Seedlings BPP Cuttings Seedlings; BPP L. indica L. indica L. fauriei** N=4 Figure 2-4. Continued

PAGE 39

39 CHAPTER 3 TRITROPHIC INTERACTIONS AMONG CR APEMYRTLE, CRAPEMYRTLE APHIDS, AND THE GREEN LACEWING Chrysoperla rufilabris (BURMEISTER) Introduction Many studies have indicated that plants can in fluence insect natural enemies at the third trophic level through nutritional, morphological, or chemical at tributes (Malcolm 1990, Clark and Messina 1998, Fuentes-Contreras and Niem eyer 1998, Francis et al. 2001, Messina and Sorenson 2001, Giles et al. 2002a, Giles et al. 2002b, Gonzales et al. 2002, Legrand and Barbosa 2003). Because tritrophic interac tions can have a considerable affect on biological control systems, it is imperative to consider these interactions when implementing or evaluating biological control agents (Price et al. 1980, Van Emden 1995, Bottrell et al. 1998, Vidal and Tscharntke 2001). One of the best known trit rophic pathways involves aphids and aphid predators, where plants indirec tly affect the fitness of aphid predators through the aphid prey (Van Emden 1995). Rothschild et al. (1970) demonstrated that the oleander aphid Aphis nerii Fonscolombe can sequester cardenolides from f eeding on oleander or milkweed. Further studies indicated that lacewing predators suffered higher mortality rates on A. nerii that contained cardenolides and the mortality rate was related to the concentration of cardenolides inside the plant (Rothschild et al. 1970, Malcolm 1990, Hodek 1993). Other herbivorous in sects protect their offspring from lacewing predation by provisioni ng their eggs with chemical protections that are derived from their diet. The green lacewing, Ceraeochysa cubana Hagen, avoids Utetheisa ornatrix (L.) eggs provisioned with alkaloids, bu t will devour eggs that are left unprotected (Eisner et al. 2000). In addition to allelochemi cals, plants affect pr edators through nutritional pathways. Pea aphids reared on al falfa and faba bean contain diffe rent levels of myristic acid, and Chrysoperla rufilabris (Burmeister) larvae fed pea aphids that were reared on different host plants had differences in survivorship and adult mass (Giles et al. 2000).

PAGE 40

40 The green lacewing C. rufilabris has been implicated as an important natural enemy in several biological control systems (New 1975, Liao et al. 1984, Liao et al. 1985, Wang and Nordlund 1994, Randolph et al. 2002, Stew art et al. 2002). In pecans, C. rufilabris is considered to be a key natural enemy and one of the most dom inant predators for contro lling three species of pecan aphids (Dinkins et al. 1994). Pecan aphi ds display signs of pesticide resistance, and pesticide applications reduce the number of na tural enemies in pecans increasing the need for future pesticide applications to control prim ary and secondary pests (Mizell and Schiffhauer 1989, Mizell and Sconyers 1992, Mizell 2003). The use of cover and augmentation crops in pecans to increase natural enemy recruitment an d retention was suggested by Tedders (1983) and Mizell and Schiffhauer (1987b). Tedders (1983) suggested using clover and vetch during the spring months when pecan aphid populations are low in order to retain lacewings and other natural enemies that overwinter on the trunks of pecans (Mizell and Schiffhauer 1987a). Mizell and Schiffhauer (1987b) suggested the use of crapemyrtle as an augmentation crop because crapemyrtle aphids have a peak population in the mid to late summer, which is when pecan aphid populations experience a seasonal collapse be fore increasing again in the fall. This purpose of this study was to investig ate the suitability of crapemyrtle aphids Sarucallis kahawaluokalani (Kirkaldy) as a food source for the green lacewing C. rufilabris Experiments tested the effects of crapemyrtle cultivar, mature plant height, and source of germplasm on the life history characteristics of C. rufilabris The first experiment involved feeding C. rufilabris larvae crapemyrtle aphids ad libitum from seven different cultivars of crapemyrtle. A second study used suboptimal di ets of aphids and le pidopteran eggs to investigate if aphids reared on different crapemyr tle cultivars affected the fitness of lacewings with respect to the amount of prey consumed.

PAGE 41

41 Materials and Methods Ad Libitum Experiment Aphid rearing. Crapemyrtle aphids were reared on seven cultivars of crapemyrtle Lagerstroemia spp. as follows: Apalachee, Natchez, Sioux, Tuscarora, Lipan, Carolina Beauty, and Byers Wonderful White. Aphids were reared on crapemyrtle plants by enclosing individual branches in sleeve cages that we re constructed from cloth mesh with 1-mm2 openings. Sleeve cages were 16 cm in diameter and 80 cm in length. Before securing the sleeve cage on a branch with twine that was sewn into the sleev e cage ends, the terminal portion containing the bloom was removed (Figure 3-1). To prevent invasion by fire ants, Solenopsis invicta Buren, all branches touching the sleeve cage were pruned, and the portion of the br anch 10 cm under the sleeve cage was covered with Tanglefoot (Grand Rapids, MI). Sleev e cages were inoculated with aphid infested leaf disk s taken from the laboratory colony being reared on Tonto and inoculated two weeks before starting the ad libitum feeding experiment. Four plants per cultivar were used in the experiment and several branch es per plant received sleeve cages to ensure a sufficient supply of aphids. Lacewing rearing. Chrysoperla rufilabris eggs were ordered from Rincon Vitova insectaries (Ventura, CA). Eggs were examin ed under a microscope to check for signs of development and 400 eggs that appeared healthy we re placed singly into a cells of five 96 well plates. Wells were sealed with Parafilm that was pressed firmly against the top of the plate to contain first instar larvae and prevent cannibalism. Plates were examined every 24 h for first instar larvae, and upon discovery, the larvae we re placed into a 29.5-ml plastic Solo Cup (Urbana, IL) that had an organdy cl oth vented lid. As the larvae hatched from their eggs, they

PAGE 42

42 were randomly assigned to one of the 28 plants. Ten larvae were assigned to each plant (40 per cultivar). Lacewing larvae were reared in the 29.5-ml condiment cups and fed crapemyrtle aphids ad libitum by placing aphid-infested leaf material from the appropriate plant directly into each cup. Larvae were fed every 24 h, and the old leaf mate rial was removed and fresh material was placed into each cup. During the feeding procedures a small piece of wet cotton was placed into each cup to provide water and maintain a higher humidity than in the growth chamber. The larvae were kept in an environmental growth chamber at 25 1 C with a 16:8 light:dark photoperiod. For each lacewing, the dates of larval emergence, pupation and adult emergence were recorded. On the day of adult emergence, adult lacewi ngs were placed into the freezer at -4 C. Adults were dried in an oven at 60 C for 12 h and the dry mass was measured to an accuracy of 0.1 mg. The average of three measurements was used as the dry mass in statistical analyses. Statistics. The experimental design was originally constructed as a randomized complete block design with 10 larvae assigned to each plant and four plants per cultivar. However, due to the loss of some sleeve cages by branch breakage, weather conditions, farm equipment, or other accidental loss, some plants did not have enough aphids to support all ten developing larvae, and aphids from other plants within the same cultivar were used to ensure an adequate amount of prey for the developing larvae. All larvae were fed aphids from their assigned cultivar but plant was no longer a valid blocking criterion. Becau se plant was no longe r a blocking factor, survivorship data was binary with individual lacewings as replicat es instead of i ndividuals(plant) as the level of replic ation. Survivorship data were an alyzed using PROC GLIMMIX SAS 9.1.3 (SAS Institute 2000-2004) using the binomial di stribution option at the end of the model statement with the default logit link. Larval de velopment time and dry mass were analyzed using

PAGE 43

43 the lognormal distribution statement with defa ult link. Data were subjected to the same orthogonal contrasts for parentage, ma ture plant height, and source of L. fauriei germplasm outline in Chapter 2. Suboptimal Diet Experiment Aphid rearing. Crapemyrtle aphids were reared in the laboratory on the cultivars Byers Wonderful White, Carolina Beau ty, Apalachee, and Sioux. Cultivar selection was based on parentage. Carolina Beauty and Byers Wonderful White are from the pure L. indica parentage, and Apal achee, and Sioux are L. indica L. fauriei hybrids. Excised leaf tissue in the form of 40-mm leaf disks was used to rear crapemyrtle aphi ds and was kept inside 100 x 15 mm agar Petri-plates. Preparation of agar plat es followed the procedures outlined in chapter 2, but 20 ml of agar was placed into each 100 x 15 mm Pe tri-plate, instead of 5 ml of agar into a 40 x 10 mm Petri-plate. Leaf materi al was collected in the field by placing leaves into plastic bags, immediately placing the bags into a cooler with ice packs, and transporting the leaf material to the laboratory. In the laboratory, one 40-mm leaf disk was removed from each leaf and washed with cold tap water for several minutes. Wh ile washing, the disks were carefully rubbed between the thumb and forefinger to remove any fore ign debris, insects, or mites. Due to results from a previous feeding experiment, where none of the lacewing larvae survived to adulthood, leaf material used in this experiment was not disinfected with bleac h, and mold was not of concern in this experiment because the leaf disks were discarded after 7 d. Four leaf disks from the same cultivar were placed into each 100 x 15 mm Petri-plate with the adaxial surface of the leaf in contact with the agar. Sixty leaf disks (15 Petri-plates) from each cultivar were prepared daily for use in the experiment. The crapemyrtle aphid culture was started fr om adult virginoparae collected from the cultivar Tonto in field. Crapemyrtle branches that were infested with crapemyrtle aphids were

PAGE 44

44 brought into the labora tory and placed in a large container. Sixteen virginoparous aphids that were haphazardly aspirated from the entire pool of field-collected aphids were added to each of the 60 Petri-plates containing leaf material. To maintain a culture of nymphs of equal age and relatively equal density, the virginoparous aphids were rem oved every 24 h and transferred to a new plate containing freshly prepared leaf mate rial from the same cultivar. Aphid nymphs developed into adults within 6-7 d and the adul ts obtained after one generation were used to maintain the aphid culture for the cu ltivar that they developed upon. Petri-plates containing the aphid culture were kept on the laboratory bench at a temperature of 27 2 C and provided with supplemental lighting. Fl orescent lights were used to maintain a 16:8 light:dark photoperiod. Each light fixture co ntained two 40-W fluores cent light tubes and was hung 0.5 m from the top of the bench. Lacewing rearing. Lacewing and Ephestia kuehniella Zeller eggs were obtained from Beneficial Insectary (Re dding, CA). Upon arrival, E. kuehniella eggs were stored in the freezer until being used as prey for lacewings. Lacewing eggs were inspected for eggs that showed signs of advanced development, and 132 eggs that showed signs of advanced development were placed individually into -dram vials plugged with cotton. Each lacewing egg was randomly assigned to one of 11 treatments. Three treatm ents were used as a mortality control and lacewing larvae in these treatmen ts were fed 2.5, 5, or 10 mg of E. kuehniella eggs per day. Eight treatments tested for the effects of prey level and cultivar. Lacewing larvae were fed either 25 or 50 aphids per day that we re reared on one of the following crapemyrtle cultivars: Apalachee, Sioux, Carolina B eauty, or Byers Wonderful Wh ite. Larvae were fed once per day, and before placing new prey into a vial, the larva was rem oved and any remaining prey or excrement were removed from the vial. If a larv a was attached to the vial and molting, the larva

PAGE 45

45 was not removed from the vial and any remaini ng prey were carefully removed with a small camel hair brush without touchi ng or disturbing the larva. Most larvae molted on the cotton plug, which allowed them to be re moved without disturbing them. Aphids were delivered to the vials by first aspirating them into a 1.5-ml Eppendorf tube, inserting the vial into the tube until it made a seal, and tapping the aphids from the Eppendorf tube into the vial. This method of transfer wa s found to be non-lethal to the aphids and more efficient than using a brush. Ephestia kuehniella eggs were measured and delivered volumetrically by calibrating a 10l pipette tip to the three diet levels. The 10-mg E. kuehniella egg control was considered an ad libitum diet because there were always uneaten eggs present for all larval instars. Vials were kept on their si de in the lid of a large 150 x 20 mm Petri-plate. To maintain a high relative humidity of 6080%, two pieces of cotton were wetted with deionized water and the bottom of the plate was placed on the lid slightly ajar (Figure 3-2). Humidity was tested on several occasions with a digital hygrometer in each Petri-plate and found to be within 60-80%. Experimental parameters. Two samples consisting of 25 aphids were collected from each cultivar and placed in a pre-weighed 1.5ml Eppendorf tube. Fresh mass of the aphid sample was recorded by weighti ng the tube a second time. Afte r weighing, the samples were place into a freezer at -4 C. For each lacewing, I recorded th e date of larval eclosion, pupation and adult emergence. Pupal mass was measured 3 d into pupation and adult fresh mass was recorded on the day of adult emergence. Afte r measuring fresh mass, adult lacewings were stored in a freezer at -4 C. Adult lacewings and aphid samples were lyophilized by first freezing them at -80 C for 2 h, and immediately transferring them to a Labconco FreeZone 2.5 for 24 h. Dry mass of aphid and lacewing samp les were measured to an accuracy of 0.01mg.

PAGE 46

46 Statistics. Statistical analyses used PROC GLIMMIX SA S 9.1.3 (SAS Institute 20002004). The GLIMMIX procedure was chosen because it allows for generalized mixed models and can analyze data that conform to binom ial, Poisson, lognormal, or other non-normal distributions. Data transformati ons were performed on data that did not meet the assumptions for an analysis of variance. The most a ppropriate transformation was assessed by doing a residual analysis for normality and equal variance. Survivorship data were analyzed using the binomial distribution with the de fault logit link. Larval develo pment and adult dry mass were analyzed using the lognormal distribution, and p upal development was anal yzed using a square root transformation. Transformation of pupa l mass data was not necessary. Results Ad Libitum Experiment The cultivar or source of crap emyrtle aphid prey did not sign ificantly influence lacewing larval survivorship, but did influe nce percentage of pupae that emer ged as adults (F = 3.26; df = 6, 172; P = 0.0046) and total survivorship (F = 3.23; df = 6, 271; P = 0.0044; Table 3-1; Figure 3-3; Figure 3-4). Total survivorsh ip of lacewings from first instar larva to adult indicated there were differences associated with the source of L. fauriei germplasm (F = 7.68; df = 1, 271; P = 0.0060; Table 3-1; Figure 3-3) and differences be tween the cultivars Lipan and Tuscarora (F = 6.15l; df = 1, 271; P = 0.0137). Pupae of lacewings that were fed aphids reared on Apalachee had a lower emergence rate than lacewings that were fed aphids from the other sources of L. fauriei germplasm (F = 7.68; df = 1, 172; P = 0.0062; Ta ble 3-1; Figure 3-4). In the Bashams Party Pink source of L. fauriei germplasm, pupal emergence was lower for lacewings that were fed aphids reared on Lipan, which has a medi um mature plant height, than lacewings fed aphids reared on Tuscarora (F = 7.49; df = 1, 172; P = 0.0069; Table 3-1; Figure 3-3; Figure 34).

PAGE 47

47 Larvae that were fed aphids reared on Apalachee had a longer time to pupation than larvae that were fed aphids from the other sources of L. fauriei germplasm (Table 3-1; Figure 35). Cultivar mature plant height within source of germplasm influenced larval development in the pure L. indica L. fauriei seedlings, and L. fauriei Bashams Party Pink sources of germplasm. In each source of germplasm, lacewi ng larvae that were fed aphids reared on the cultivar with a medium mature plant height t ook longer to pupate (Table 3-1; Figure 3-5). Adult dry mass was greater for lacewings that were fed aphids reared on cultivars from the L. fauriei seedlings source of germplasm when compared to lacewings that were fed aphids from the L. fauriei Bashams Party Pink source of germplasm (Table 3-1; Figure 3-6). Differences in adult dry mass were associated with mature pl ant height, and lacewings that were fed aphids from cultivars with a tall mature plant height ha d a higher dry mass than lacewings that were fed aphids reared on cultivars with a medium matu re plant height (Table 3-1; Figure 3-6). Suboptimal Diet Experiment Aphid diets. Aphids reared on pure L. indica cultivars had a lower dry mass than aphids reared on the L. indica L. fauriei cultivars (Table 3-2; Figure 3-7). Aphid dry mass was linearly related to fresh mass for all cultivars (Tab le 3-3; Figure 3-8). Slopes for the regressions of dry mass vs. fresh mass were not found to be significantly different us ing an analysis of covariance. Chrysopidae. Larval developmental time was shor ter for larvae that were given higher levels of daily prey, and this effect was c onsistent for all crapemyrtle cultivars and the E. kuehniella egg controls (Table 3-4; Figure 3-9A). In addition to differences in daily prey levels, larvae reared on E. kuehniella eggs had a lower duration of larv al development in comparison to larvae reared on crapemyrtle aphi ds (Figure 3-9A). Pupal durat ion was shorter for larvae fed E.

PAGE 48

48 kuehniella eggs but did not differ in a ssociation with crapemyrtle cu ltivar (Table 3-4; Figure 39B). Pupal mass was greater for lacewing larvae that were fed higher levels of daily prey and this was consistent across all crapemyrtle cultivars and the E. kuehniella egg controls (Table 3-4; Figure 3-10). Differences in pupal mass were seen between lacewing larvae that were fed E. kuehniella eggs and larvae that were fed crapemyr tle aphids from Byers Wonderful White (Figure 3-10). Lacewing adult dry mass was different among the crapemyrtle and E. kuehniella egg treatments (F = 6.24; df = 4, 61; P < 0.001; Table 3-4; Figure 3-11). Lacewings provisioned with higher levels of daily prey had a greater dry mass (F = 61.34; df = 1, 61; P < 0.0001; Table 3-4), and males had a lower dry mass than females (F = 2.41; df = 1, 61; P < 0.001; Table 3-4). Discussion Ad Libitum Experiment Data from this experiment indicate that there are multitrophic interactions among crapemyrtle Lagerstroemia spp., crapemyrtle aphids, and the green lacewing. The pupal stage appears to be the most vulnerable stage with resp ect to survivorship, and the most critical stage with regards to developmental time was the larv al stage. Orthogonal contrasts indicated that lacewings were affected by broader plant attri butes like mature plant height, and source of L. fauriei germplasm. If crapemyrtle is evaluated for enhancing biological control in pecans or other economically important crops, as suggest ed by Mizell and Schiffhauer (1987b), this research may be helpful for selecting the most appropriate cultivars for use as an augmentation crop in pecans. Natchez and Tuscarora are tall cultivars from the L. indica L. fauriei parentage, and lacewings fed crapemyrtle aphids reared on thes e cultivars had the shortest development time,

PAGE 49

49 lowest pupal mortality, and highest adult dry mass. Previous studi es (Chapter 2) showed that cultivars from the L. indica L. fauriei parentage were more suitable as hosts and more preferred by crapemyrtle aphids. Natchez is rarely infest ed with crapemyrtle aphids in the field and has the lowest susceptibility rating among the cultiv ars tested in this experiment (Mizell and Knox 1993). Lower aphid populations in the field could be due to higher predation rates by lacewings or other aphid predators. This may be a result of either qualitative or qua ntitative differences in aphid prey. Qualitative or quantitative differences in aphid prey have been determined by other researchers through the use of suboptimal diets (G iles et al. 2000, Giles et al. 2002a, Giles et al. 2002b). The idea put forward by Giles was that wh en predators have differential survivorship based on the source of prey this difference may or may not be overcome by increasing the amount of prey consumed. If the predator ca n overcome a nutritional deficiency by consuming more prey, the difference is quantitative. If th ere is no increase in predator fitness with the increased consumption of prey, the difference is qu alitative. However, toxic chemicals within an aphid can affect aphid predators, and in the cont ext of toxins, a predator would likely be more adversely affected by consuming more prey (Price et al. 1980, Malcolm 1990, Van Emden 1995). Lagerstroemia indica has several alkaloids that are fo und in the leaves, bark, and seeds (Ferris et al. 1971a, 1971b, 1971c). Mizell et al. (2002) suggested that these chemicals could play a defensive role and help protect crapemyr tle aphids against insect parasitism. Further experiments examining more hom ogenous aphid populations derive d from different crapemyrtle cultivars and offered to lacewings at differing leve ls of prey will help in determining if lacewing mortality is associated with the number of crapem yrtle aphids eaten or the plant that the aphids developed upon.

PAGE 50

50 Suboptimal Diet Experiment Lacewing survivorship in this experiment was not found to be significantly affected by cultivar or the amount of food provisioned on a daily basis, but daily prey levels and food source did influence larval developmental time, pupa l mass and adult dry mass. Lacewing larvae developed faster and had a higher pupal mass when they were reared on hi gher daily prey levels, regardless of the food source. Stud ies on the diet and life history of C. rufilabris have indicated that faster larval developmental time is a ssociated with higher quality food (Hydorn and Whitcomb 1979, Legaspi et al. 1994, Legaspi et al. 1996). However, Hydorn and Whitcomb (1979) also showed that C. rufilabris larvae reared on Tribolium castaneum (Herbst) had a higher adult mass, but lower fecundity than C. rufilabris larvae that were reared on aphids or Phthorimaea operculella (Zeller) eggs. Hydorn and Whitcomb (1979) attributed the higher adult mass to a longer developmental ti me, and in this experiment, C. rufilabris that were reared on crapemyrtle aphids had a longer larval developm ent time and a larger dry mass than lacewings reared on E. kuehniella eggs. These results are interesti ng because lacewing larvae reared on E. kuehniella eggs had a higher pupal mass than larv ae reared on crapemyrtle aphids. The discrepancy between these findings may be due to pupal survivor ship, and although the relationship is not significant, pup ae that did not successfully emer ge as adults had a mean pupal mass of 6.5 mg whereas pupae that did emerge had a mean pupal mass of 6.23 mg. Future experiments using higher da ily prey levels can eval uate the hypothesis that C. rufilabris suffers higher mortality rates when consuming large quantities of crapemyrtle aphids. Lacewing Study Results from this study indicate that in the ad libitum diet experiment crapemyrtle cultivar influenced lacewing survivorship, but this eff ect was not detected in the suboptimal diet experiment. Differences between the two expe riments that were not controlled were the

PAGE 51

51 exposure to plant material, number of aphids c onsumed, and the rearing containers. In the ad libitum diet experiment, lacewing larvae were exposed to plant material and lacewing may have attempted to feed directly on the leaf material to obtain water or nutrients. This exposure cannot eliminate the possibility of a direct plant eff ect on lacewing fitness. However, under field conditions it is unlikely that lacewings would encounter a sufficient number of aphids to complete larval development without being present on the plant, which would expose the developing larvae to leaf materi al, sooty mold, or aphid honeydew. Thus, the results from the ad libitum diet experiment are the most applicable fo r considering what may happen in the field. Future experiments looking at survivorship with and without plant material can be conducted to test the hypothesis that the plan t is directly affecting the f itness of lacewing larvae. Aphid size, age, or instar may be important factors in determining prey suitability, and New (1984) suggested that studie s investigating the e ffects of aphid prey on the fitness or survival of lacewings should account for variation in size, instar, or age of aphid prey. In the suboptimal diet study, the age and relative dens ity of aphids was cont rolled and there was no significant difference in lacewing survivorship am ong the different sources of diet. However, there was an indication that ea ting more crapemyrtle aphids ma y decrease the survivorship of lacewings in the cultivars Apalachee, Sioux and Carolina Beauty Byers Wonderful White did not show this trend and may be the most appropriate cultivar, from the cultivars tested, to serve as an augmentation crop in pecans.

PAGE 52

52 Table 3-1. Analysis of variance results for the ad libitum diet experiment with contrast statements. Response variable (effect) Contrast Num df Den df F Pr > F Pupal survivorship (cultivar) 6 172 3.26 0.0046 L. indica vs. L. indica x L. fauriei 1 172 0.66 0.4162 Apalachee vs. All other L. indica x L. fauriei 1 172 7.68 0.0062 Bashams Party Pink vs. Seedlings 1 172 1.42 0.2347 Lipan vs. Tuscarora (Bashams Party Pink) 1 172 7.49 0.0069 Sioux vs. Natchez (Seedlings) 1 172 0.90 0.3431 Byers Wonderful White vs. Carolina Beauty 1 172 1.66 0.1997 Total Survivorship (cultivar) 6 271 3.23 0.0044 L. indica vs. L. indica x L. fauriei 1 271 0.61 0.4340 Apalachee vs. All other L. indica x L. fauriei 1 271 7.68 0.0060 Bashams Party Pink vs. Seedlings 1 271 3.31 0.0700 Lipan vs. Tuscarora (Bashams Party Pink) 1 271 6.15 0.0137 Sioux vs. Natchez (Seedlings) 1 271 0.20 0.6548 Byers Wonderful White vs. Carolina Beauty 1 271 1.91 0.1676 Larval Development Time (cultivar) 6 176 5.22 0.0001 L. indica vs. L. indica x L. fauriei 1 176 0.02 0.8929 Apalachee vs. All other L. indica x L. fauriei 1 176 4.10 0.0445 Bashams Party Pink vs. Seedlings 1 176 0.62 0.4304 Lipan vs. Tuscarora (Bashams Party Pink) 1 176 10.95 0.0011 Sioux vs. Natchez (Seedlings) 1 176 10.60 0.0014 Byers Wonderful White vs. Carolina Beauty 1 176 4.52 0.0348 Adult Dry Mass (cultivar) 6 113 9.31 0.0001 L. indica vs. L. indica x L. fauriei 1 113 3.38 0.0686 Apalachee vs. All other L. indica x L. fauriei 1 113 0.07 0.7866 Bashams Party Pink vs. Seedlings 1 113 4.16 0.0437 Lipan vs. Tuscarora (Bashams Party Pink) 1 113 22.91 0.0001 Sioux vs. Natchez (Seedlings) 1 113 24.71 0.0001 Byers Wonderful White vs. Carolina Beauty 1 113 0.45 0.5058 Table 3-2. Analysis of variance results with contrasts for the dry mass of crapemyrtle aphids reared on different cultivars of crapemyrtle. Response Contrast Num df Den df F Pr > F Dry Mass (cultivar) 3 40 11.42 <0.0001 L. indica vs. L. indica x L. fauriei 1 40 20.74 <0.0001 Medium vs. Tall 1 40 0.43 0.5175

PAGE 53

53 Table 3-3. Regression analyses re sults for the comparison of aphid dry mass to aphid fresh mass for different crapemyrtle cultivar s using samples of 25 aphids. Cultivar Model df Error df F Pr > F R2 Apalachee 1 17 21.38 0.0008 0.53 Byers Wonderful White 1 19 21.13 0.0002 0.50 Carolina Beauty 1 20 75.20 0.0001 0.78 Sioux 1 20 121.87 0.0001 0.85 Table 3-4. Analysis of variance results for C. rufilabris in the suboptimal diet experiment. Response Effect Num df Den df F Pr > F Larval Development Diet Source 4 86 46.69 <0.0001 Prey Level 1 86 44.93 <0.0001 Diet Source Prey Level 4 86 0.70 0.5962 Pupal Duration Diet Source 4 71 7.31 <0.0001 Prey Level 1 71 0.19 0.6640 Diet Source Prey Level 4 71 0.67 0.6171 Pupal Mass Diet Source 4 86 2.92 0.0257 Prey Level 1 86 30.36 <0.0001 Diet Source Prey Level 4 86 0.21 0.9341 Adult Dry Mass Diet Source 4 61 6.24 0.0003 Prey Level 1 61 61.34 <0.0001 Sex 1 61 23.41 <0.0001 Diet Source Prey Level 4 61 1.17 0.3324 Diet Source Sex 4 61 1.28 0.2872 Sex Prey Level 1 61 0.13 0.7189

PAGE 54

54 20 cm Figure 3-1. Sleeve cage attached to a br anch of crapemyrtle (Natchez). 15 cm Figure 3-2. Vials and Petr i-plates containing lacewing larvae in the suboptimal diet experiment.

PAGE 55

55 CBBWApaSioNatLipTus % Total survivorship 0 20 40 60 80 100 Seedlings BPP Cuttings Seedlings; BPP L. indica L. in dica L. fauriei ** **ab ab b ab a ab ab Med Tall Med Tall Med Tall Figure 3-3. Percentage of lacewi ng larvae that successfully devel oped from first instar larva to adult in the ad libitum diet experiment with orth ogonal contrasts for the cultivars Byers Wonderful White (BW), Carolin a Beauty (CB), Apalachee (Apa), Natchez (Nat), Sioux (Sio), Tuscaror a (Tus), and Lipan (Lip). Columns followed by different levels are different at P < 0.05 Tukeys HSD, and contrasts marked by asterisks such that *P < 0.05; **P < 0.01.

PAGE 56

56 CBBWApaSioNatLipTus % Emergence 0 20 40 60 80 100 Med Tall Seedlings BPP Cuttings Seedlings; BPP L. indica L. indica L. fauriei ** **ab a ab ab b ab ab Med Tall Med Tall Figure 3-4. Percentage of lacewi ng pupae emerging as adults in the ad libitum diet experiment with orthogonal contrasts for the cultivar s Byers Wonderful White (BW), Carolina Beauty (CB), Apalachee (Apa ), Natchez (Nat), Sioux (Sio), Tuscarora (Tus), and Lipan (Lip). Columns followed by di fferent levels are different at P < 0.05 Tukeys HSD, and contrasts marked by as terisks such that *P < 0.05; **P < 0.01.

PAGE 57

57 CBBWApaSioNatLipTus Sqrt. larval development days 0 1 2 3 4 abc ab ab bc a c ab Seedlings BPP Cuttings Seedlings; BPP L. indica L. indica L. fauriei ** ** Med Tall Med Tall Med Tall Figure 3-5. LS mean ( SE) for the squa re root of larval development for the ad libitum diet experiment with orthogonal contrasts fo r the cultivars Bye rs Wonderful White (BW), Carolina Beauty (CB), Apalachee (Apa), Natchez (Nat), Sioux (Sio), Tuscarora (Tus), and Lipan (Lip). Columns followed by different levels are different at P < 0.05 Tukeys HS D, and contrasts marked by asterisks such that *P < 0.05; **P < 0.01.

PAGE 58

58 CBBWApaSioNatLipTus Log drymass in mg 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 cb c abc a c ab c Seedlings BPP Cuttings Seedlings; BPP ** ** L. indica L. indica L. fauriei Med TallMed TallMed Tall Figure 3-6. LS mean log dry mass ( SE) for adult lacewings in the ad libitum diet experiment with orthogonal contrasts for the cultivar s Byers Wonderful White (BW), Carolina Beauty (CB), Apalachee (Apa ), Natchez (Nat), Sioux (Sio), Tuscarora (Tus), and Lipan (Lip). Columns followed by di fferent levels are different at P < 0.05 Tukeys HSD, and contrasts marked by as terisks such that *P < 0.05; **P < 0.01.

PAGE 59

59 Diet BWCBApaSio Mass in mg 0.0 0.2 0.4 0.6 0.8 1.0 1.2 b ab a a Figure 3-7. Mean dry mass ( SE) for aphid samples in the suboptimal diet experiment for crapemyrtle aphids reared on the pure L. indica cultivars Byers Wonderful White (BW), Carolina Beauty (CB), and the L. indica L. fauriei cultivars Apalachee (Apa), and Sioux (Sio). Columns with different letters are different at P < 0.05 using Tukeys HSD.

PAGE 60

60 Fresh mass mg 012345 Dry mass mg 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Apa (y = -0.0427 + 0.2334x) BW (y = 0.1245 + 0.1973x) CB (y = 0.1294 + 0.1877x) Sio (y = -0.0963 + 0.2726x) Figure 3-8. Relationship of dry mass to fresh mass for aphid samples taken from the cultivars Byers Wonderful White (BW), Carolina Beauty (CB), Apalachee (Apa), and Sioux (Sio).

PAGE 61

61 A EphBWCBApaSio 0 2 4 6 8 10 12 14 Ad lib High Low c a ab ab b BDiet EphBWCBApaSio Time in days 0 2 4 6 8 10 12 14 Ad lib High Low b a a a a Figure 3-9. Mean developmental time of lacewings ( SE) for the suboptimal diet experiment for A) lacewing larvae and B) lacewing pupae for the different sources of diet using E. kuehniella eggs (Eph) and crapemyrtle aphids from the cultivars Byers Wonderful White (BW), Carolina Beau ty (CB), Apalachee (Apa ), and Sioux (Sio). Combined paired columns (low and high pr ey levels) with different letters are different for the main effect of diet at P < 0.05 using Tukeys HSD. Low equals 25 aphids or 2.5 mg of E. kuehniella eggs daily and high equals 50 aphids or 5 mg of E. kuehniella eggs daily. Eph ad lib was not used in statistica l analyses, but presented to indicate relative trends.

PAGE 62

62 Diet EphBWCBApaSio Mass in mg 0 2 4 6 8 10 12 14 a b ab ab ab Ad lib High Low Figure 3-10. Mean pupal mass of lacewings ( SE ) in the suboptimal diet experiment for the treatments E. kuehniella eggs (Eph) and crapemyrtle ap hids from the cultivars Byers Wonderful White (BW), Car olina Beauty (CB), Apala chee (Apa), and Sioux (Sio). Combined paired columns (low and high prey levels) with different letters are different for the main effect of diet at P < 0.05 using Tukeys HSD. Low equals 25 aphids or 2.5 mg of E. kuehniella eggs daily and high equals 50 aphids or 5 mg of E. kuehniella eggs daily. Eph ad lib was not used in statistica l analyses, but presented to indicate relative trends.

PAGE 63

63 Diet EphBWCBApaSio Log mass in mg 0.0 0.1 0.2 0.3 0.4 0.5 0.6 b ab a a a Figure 3-11. LS mean log adult dry mass ( SE) for lacewings that successfully emerged without deformation in the suboptimal experime nt for the main effects of A) diet, B) daily prey levels, and C) sex. E. kuehniella eggs (Eph) and crapemyrtle aphids from the cultivars Byers Wonderful White (B W), Carolina Beauty (CB), Apalachee (Apa), and Sioux (Sio). Columns with different letters are different at P < 0.05 using Tukeys HSD.

PAGE 64

64 CHAPTER 4 LANDSCAPE ECOLOGY OF THE CRAPEMYR TLE APHID AND THE APHID PREDATOR Harmonia axyridis (PALLAS) Introduction Lady beetles in the family Coccinellidae are vora cious predators that are considered to be important for controlling aphids and other hemipterans pests (Obrycki and Kring 1998). Previous studies observing the response of lady b eetles to aphid prey indi cated that lady beetles respond to prey densities on a temporal and spa tial level, and because Coccinellidae do not search for prey at random, individuals tend to ag gregate in patches that have a high density of prey (Kareiva and Odell 1987, Kindlma nn and Dixon 1993, Kindlmann and Dixon 1999a, 1999b). Understanding the movement of cocci nellids on a landscape level can enhance the understanding and success of biol ogical control programs. To understand coccinellid movement through the landscape, researcher s have sought methods of qua ntifying the response of lady beetles while reducing time and cost of samp ling (Musser et al. 2004, Stephens and Losey 2004, Schmidt et al. 2008). Previous studies using stic ky traps in the form of sticky cards or small cylindrical traps have been used to track the movement, abundance, and distribution of Coccinellidae (Parajulee and Slosser 2003, Musse r et al. 2004, Stephens and Losey 2004, Hesler and Kieckhefer 2008, Schmidt et al. 2008). Mens ah (1997) discovered that Coccinellidae are highly attracted to hues of yellow, and Paraju lee and Slosser (2004) found that yellow sticky traps were an efficient method for sampling lady beetles in cotton. The multicolored Asian lady beetle, Harmonia axyridis (Pallas), was introduced into the United States as a biological control agent in arboreal and agricultural habitats (Tedders and Schaefer 1994, Nault and Kennedy 2003). Since its establishment in 1980, H. axyridis has become the dominant species of lady beetle am ong many agricultural and natural landscapes and is considered to be an important natural enem y for controlling aphid pe sts (Rutledge et al. 2004,

PAGE 65

65 Costamagna et al. 2007, Mize ll 2007). In crapemyrtle Lagerstroemia spp., H. axyridis is the most frequently observed aphid pr edator and is present in crap emyrtle from spring to autumn (Mizell 2007). The spatial a nd temporal distribution of H. axyridis has not been studied in crapemyrtle, and understanding how H. axyridis moves at a landscape level may help develop strategies to better use this predator in biologi cal control programs. Recent advances in spatial modeling and analys es have provided tools to detect spatial clusters, which are gaps or aggregations in the spatial distribution of an organism. Association tests can determine if the spatial distributions of two species or the distributions of one species over time are significantly associated at a spa tial level. Spatial Anal ysis by Distance IndicEs (SADIE) is a relatively new and novel way of measuri ng spatial aggregati ons (clusters) and associations, and uses ecological count data (P erry 1995, Perry et al. 1996, Perry 1998). SADIE uses the values of the experimental data set (e ntered as x, y, count) to generate data that are distributed at regular interval s, randomly, or completely clus tered (Perry 1995, Perry et al. 1996, Perry 1998, Perry et al. 1999, Perry and Dixon 2002) SADIE uses numerous permutations of the algorithms to asse ss how many randomly generated data sets are as clustered or more clustered than the original data set and uses this information to calculate the probability that the original data are more cluste red than by random chance. Cl ustering of data can indicate aggregations or gaps in the spatial distribu tion. Developers of the program describe the techniques used by SADIE to be biologically rele vant and under some conditions the program is more likely to detect clusters when compared to other geostatistical methods that assume a regular covariance matrix (Pe rry 1998, Korie et al. 2000). SADIE performs association analyses that comp are the distribution of two data sets over the same space. To test for spatial associations, SADIE uses information that is generated in the

PAGE 66

66 cluster analysis of a single set of data (P erry and Dixon 2002). A limitation of performing association analyses in SADIE is that the data in both data sets must have the same x, y coordinates. The association f unction in SADIE uses the clust er.dat files generated during a cluster analysis to calculate (Chi), which can be positive or negative in value (Perry et al. 1996, Perry and Dixon 2002). Positive values of indicate a tendency toward an association, while negative values of point toward disassociation. A ssociation tests in SADIE are twotailed and significance is reco rded as P < 0.025 for a signific ant association and P > 0.975 for two data sets that are si gnificantly disassociated. The purpose of this study was to observe the spatial and tempor al distribution of crapemyrtle aphids and H. axyridis in a large plot of crapemyrtle, and to determine if the distributions of aphids and H. axyridis were spatially or temporally associated. The second objective of this study was to determine if th ere were cultivar e ffects on populations of crapemyrtle aphids and H. axyridis The third objective of this study was to evaluate if the number of H. axyridis trapped on a yellow sticky trap was correlated with aphid or H. axyridis populations. Materials and Methods Experimental Setup The study was conducted in a large pl ot of 7-yr-old crapemyrtles Lagerstroemia spp. located at the North Florida Research and Educa tion Center in Quincy, FL (Figure 4-1A). The plot contained four blocks of crapemyrtle and each block had two rows of plants. Fourteen cultivars of crapemyrtle, Byers Wonderful Wh ite, Carolina Beauty, Biloxi, Lipan, Tuscarora, Tonto, Sioux, Natchez, Tuskeg ee, Miami, Yuma, Acoma, Apalachee, and Osage were represented in each block (sev en per row) and each cultivar was represented

PAGE 67

67 by four plants that were planted side by side (Fig ure 4-1A, B). Cultivars vary in their attributes of parentage, disease resistance, and crapem yrtle aphid susceptibility (Table 4-1). Sticky traps used in the experiment were c onstructed from 7.6 91 cm mailing tubes that were painted yellow (PPG, 11-34 yellow Pi ttsburg, PA) and covered with Tanglefoot (Grand Rapids, MI). Traps were placed over the top of bamboo poles fitted with a wire stop that held the trap such that the bottom of the trap wa s 1 m above the ground. The study used 56 sticky traps inside the plot and 44 stic ky traps outside the plot. Trap s inside the plot were placed between the first or second pair of plants from the same cultiva r, and this arrangement allowed the traps to be staggered so that the traps in ad jacent rows were not next to each other (Figure 42). Traps on the northern and southern periphery of the plot were arranged as if there was an additional block of crapemyrtle and used the same spacing dimensions as th e traps located inside the plot (Figure 4-2). Traps on the eastern and western periphery we re placed at varying intervals to allow farm equipment to pass through the plot area without moving traps. Three traps were placed in adjacent habitats. Trap s numbered 26 and 63 were placed in a plot of ornamental grasses, and trap 100 was placed in an adjacent plot of crapemyrtle (Figure 4-2). Sampling 2007. Sampling in 2007 began on 4-August and continued until 11-November. Crapemyrtle plants were sampled for crapemyr tle aphids and Coccinellidae. Sampling for aphids was conducted by counting the number of aphids on 10 haphazardly selected leaves, and sampling for lady beetles was conducted by visual sampling. To sample visually, one complete walk around the plant was perfor med while visually examining th e plant for lady beetles. All 224 plants were sampled on the same day, and plants were sampled once per week.

PAGE 68

68 Samples of Coccinellidae were collected from sticky traps two to four times per week. Sticky trap samples were collected the morning of plant sampling and then again the following morning to obtain a sample for the day of plant sampling. Traps were then checked every 3-6 d. While checking sticky traps, lady beetles and other insects were removed from the traps to keep the traps clean and increase their longevity in the field. Traps were exchanged with newly prepared traps as needed and all 100 trap s were changed within a 24-h period. 2008. Plants were sampled for aphids as in 2007, but because of results obtained in 2007, only the plants adjacent to sticky traps were sampled for aphids and H. axyridis Because the number of H. axyridis was very low in 2008, sampling for coccinellids on plants was performed by beat sampling. Four beat samples were take n per plant; one sample per ordinal direction. Beat sampling was conducted by placing a beat sheet 0.75 0.75 m under the plant and beating the foliage above twice with a pi ece of PVC pipe. Sticky trap sa mples were collected as in 2007 with two to four sampling days per week. Freque nt rain storms and higher levels of rain in 2008 caused some asynchrony with the collection of stic ky trap samples over the same time period as plant sampling. When an asynchrony occurred, the sticky trap samples were collected 24 h before or 24 h after the day of plant sampling. Trap cleaning and exchanging procedures were the same as the procedures in 2007. Statistics Notation. Data were recorded in three di fferent data sets that measured H. axyridis response on the level of plant, traps that were ex posed for 24 h in the field and traps that were exposed for several days in the field. To help eliminate confusion when describing statistical tests and results, the following terms are used to describe the different measurements of H. axyridis : 1) H. axyridis observed on a plant are referred to as H. axyridis per plant, 2) H. axyridis trapped during plant sampling are referred to as H. axyridis trapped in a 24-h period, 3)

PAGE 69

69 H. axyridis trapped in traps that were exposed for 3 d are referred to as H. axyridis trapped over a 3-d period, 4) when the number of H. axyridis in a trap was divided by the number of days the trap was in the field, the measurement is referred to as H. axyridis per trap day. Analysis of variance. Analyses of lady beetle data were restricted to H. axyridis because H. axyridis was the dominant lady beetle in the lands cape and the number of individuals from other species was extremely low. Data from 2007 and 2008 for the number of aphids per plant and the number of H. axyridis per plant were analyzed with a repeated measures analysis in PROC GLIMMIX SAS 9.1.3 (SAS Inst itute 2000-2004). The statistical model used to analyze the aphid and H. axyridis per plant data were the same a nd contained two random statements. Block was random and the statement _residual_ / sub=plant*cultivar*block type = AR(1); was used in the model to performed the repeated measure (SAS institute 2000-2004). Sticky trap data for H. axyridis were divided into two da ta sets per year. One of the data sets was the number of H. axyridis trapped in a 24-h period and the other data set had the number of H. axyridis trapped on all sampling dates re gardless of how long the trap was in the field. Repeated measures analyses for H. axyridis sticky trap data were perf ormed using PROC GLIMMIX SAS 9.1.3 (SAS Institute 2000-2004). The model for H. axyridis data contained the random statement random _residual_ /sub = cultivar*block type = AR(1) to perform the repeated measure (SAS Institute 2000-2004). Data analyzed using repeated measures anal ysis were count data, and to correct for the non-normality of count data, th e distribution function built into the PROC GLIMMIX procedure was used. To fit a l ognormal distribution in PROC GLIMMIX, the statement /dist = lognormal was used at th e end of the model statement (SAS Institute 20002004). All repeated measures analyses were co nducted using the lognorm al distribution function in SAS and the default identity link. The main effects and interactions among cultivar, edge, and

PAGE 70

70 time were tested in all analyses, but the interactio ns of cultivar edge and cultivar edge time could not be tested because the edge does not contain all 14 cultivars. Repeated measures analyses for plant data did not us e sampling dates that were more th an 2 wk before or 2 wk after the peak in aphid populations because the data were mostly zeros and caused overdispersion of the data. Means were separated, when appropria te, using a Tukeys HSD family-wise error rate for all pair-wise comparisons (SAS Institute 2000-2004). Regression analyses. Regression analyses were conduc ted using SigmaPlot 11 (Systat Software Inc.). Data used in regression analyses were transformed as necessary to meet the assumptions of the model and pooled by samp ling date. The following regressions were performed and are listed as y vs. x: 1) H. axyridis per plant vs. aphids per plant, 2) H. axyridis trapped in a 24-h period vs aphids per plant, 3) H. axyridis trapped in a 24-h period vs. H. axyridis per plant 4) the number of H. axyridis trapped on traps outside th e plot vs. the number of H. axyridis trapped inside the plot (all sa mpling dates), 5) the sum total of H. axyridis trapped within the plot was divided by the total number of H. axyridis trapped and observed on plants in the plot yielding the proportion of H. axyridis trapped, and a regression was performed on the proportion of H. axyridis trapped vs. mean number of aphids per plant. Spatial statistics. Data were analyzed for spatial a ggregations or clusters using SADIE shell. All response variables from all samp ling dates in 2007 were subjected to a cluster analysis, but because of results at low aphid populations, the dates of testing were restricted in 2008. The cluster indices produced by SADIE in the c luster.dat file were used in analyses that tested for spatial associations. Plants and stic ky traps were not in the exact same place, which caused a problem with association analyses that require the exact same x, y coordinates. To remedy the coordinate problem, data from the tw o plants adjacent to a sticky trap were summed

PAGE 71

71 and assigned the coordinates of the sticky trap. Data created from su mming the response of H. axyridis or aphids on the plants adjacent to sticky traps were subjected to a cluster analysis to obtain a proper cluster.dat file that could then be used for testing spatial associations. Numerous association tests were performed in an effort to detect spatial patterns among crapemyrtle aphids and H. axyridis To test for a spatial associ ation between the distribution of aphids and H. axyridis per plant, the aphid and H. axyridis per plant data were paired by the date of sampling (e.g., 06-August). The procedure of pairing data by sampling date was repeated on aphid and H. axyridis per plant data for all sampling dates of interest. The spatial association of aphids and H. axyridis trapped in a 24-h period was eval uated by repeating the previous procedures and using the summed aphid data. A third set of association analyses for H. axyridis plant data and H. axyridis trapped in a 24-h period were perf ormed using the previous procedures and the summed H. axyridis data. Predators often show a delayed populationlevel response in which the number of predators observed lags behind the number of prey. To test if the number of H. axyridis on a plant was lagging behind the number of aphids on a plant, an association analysis was performed by pairing aphid data and H. axyridis per plant data that were separated by 1 wk (e.g., aphids on 06-August and H. axyridis on 13-August). Two additional sets of association analyses were performed to assess if the capture of H. axyridis trapped in a 24-h peri od was lagging behind the number of aphids or H. axyridis on plants that were adjacent to the sticky traps. Spatial association tests were performed on data from the H. axyridis trapped over a 3-6 d period to investigate the hypothe sis that traps are capturing H. axyridis 3 d before or 3 d after visiting a plant. Sets of spatial analyses were performed by pairing H. axyridis per plant (summed) with H. axyridis trapped over a 3-d period for the sampling dates that represent the

PAGE 72

72 number of H. axyridis trapped 3 d before or 3 d after plant sampling. The aphid (summed) data and the H. axyridis trapped in a 24-h period data were also compared with the number of H. axyridis trapped over a 3-d period for the dates before and after plant sampling. Results from the spatial association tests were compiled in to tables that contain the value of (Chi) for each test and the probability of association P(a). The values of were included in the results because trends in the value of over time can indicate broader spatial trends. Results 2007 Data Phenology. Phenology data for crapemyrtle aphids and H. axyridis per plant were plotted over time and pooled by sampling date. The mean number of aphids and H. axyridis per plant peaked in early August, but peak H. axyridis populations occurred one week after peak aphid populations (Figure 4-3). Phenology for aphids and H. axyridis was similar on all 14 cultivars of crapemyrtle Lagerstroemia spp. (Figure 4-4A-N). Sticky trap data for H. axyridis was pooled by sampling date and divided by the number of days the trap was in the field. The number of H. axyridis trapped per trap/day was the greatest in la te August and early September, and during the peak in trap catch, the number of H. axyridis captured on traps inside th e plot was greater than the number of H. axyridis caught in traps outside the plot (F igure 4-5). Although species of lady beetles other than H. axyridis were low in numbers, sticky traps captured specimens of: Hippodamia convergens Gurin-Mneville, Coccinella septempunctata (L.), Olla v-nigrum (Mulsant), Coleomegilla maculata (Mulsant), Cycloneda spp., and Neoharmonia spp. Analysis of variance. Results from the repeated measures analysis for the aphids per plant data indicated that the number of aphids per plant differed accordi ng to cultivar (F = 3.31; df = 13, 1485; P < 0.001), time (F = 213; df = 1, 1485; P < 0.001), and time cultivar (F = 2.91, df = 13, 1475; P < 0.001; Table 4-2, Table 4-3). The resu lts from the repeated measures analysis

PAGE 73

73 using H. axyridis per plant data indicated that time was the only factor in the an alysis (F = 70; df = 1, 779; P < 0.001; Table 4-4). Numbers of H. axyridis trapped in a 24-h period differed according to time (F = 12.20; df = 1, 379; P < 0.001; Table 4-5), and the number of H. axyridis trapped in sticky traps for all trapping dates di ffered among the factors of time (F = 173; df = 1, 1760; P < 0.001;), cultivar (F = 2.65; df = 13, 41; P = 0.009), and time cultivar (F = 2.65; df = 13, 1760; P = 0.001; Table 4-6; Table 4-7). Regression analyses. Regression analyses used data th at were pooled by sampling date. When comparing trap data vs. plant data, the an alysis used data from the 56 sticky traps that were located inside the plot. Data for aphids per plant and H. axyridis per plant were square root transformed to satisfy the assumption of equal variance, and the regre ssion analysis results indicated a strong positive correl ation for the square root of H. axyridis per plant vs. the square root of aphids per plant (R2 = 0.87; Figure 4-6). Data for H. axyridis trapped in a 24-h period vs. aphids per plant were square root transforme d, and the correlation of the square root of H. axyridis trapped in a 24-h period and the square root of aphids per plant had a positive correlation (R2 = 0.41; Figure 4-7). A qua dratic relationship explai ned the variation in the relationship between the square root of H. axyridis per plant and the square root of H. axyridis trapped in a 24-h period be tter than a linear model (R2 = 0.68; Figure 4-8). To better understand the quadratic nature of the prev ious comparison, the proportion of H. axyridis trapped was calculated using the following equation (sum total number of H. axyridis trapped) / (sum total number of H. axyridis trapped + sum total number of H. axyridis observed on plants). When the relationship between the number of a phids per plant and the proportion of H. axyridis on traps was evaluated, the correlation was negative (Figure 4-9; R2 = 0.61). Regression analysis for the relationship of the number of H. axyridis trapped inside the plot vs. the number of H. axyridis

PAGE 74

74 trapped outside the plot used a log transformation, and the two fact ors were positively correlated (R2 = 0.51; Figure 4-10). Cluster analyses. Results from cluster analyses indi cated that the distribution of aphids per plant was clustered on 06-August, 01-Oct ober, and 05-November (Table 4-8). The distribution of H. axyridis per plant was significantly clus tered on 03-September, 01-October, 08-October, 15-October, and 22-October (Table 4-8). Harmonia axyridis trapped in a 24-h period had a spatial distribution that was clustered on 13-August, 24-September, 22-October, and 12-November (Table 4-8). When the aphids per plant data were pooled by trap and assigned the coordinates of the trap, the sp atial distribution of aphids wa s clustered on 15-October and 05November (Table 4-9). Data for H. axyridis per plant pooled by trap were not clustered for any of the sampling dates (Table 4-9). Cluster analysis results for the H. axyridis trapped over a 3-d period indicated that the distribution of H. axyridis trapped was clustered for the sampling periods of the 11-13-September and all 3 d sampli ng dates between 12-28-Oct ober (Table 4-10). Association analyses. The distribution of aphids and H. axyridis on plants were spatially associated on 12-November, and the values of for all comparisons did not show a noticeable trend over time (Table 4-11). Spatial associat ion tests for the distri bution of aphids and H. axyridis trapped in a 24-h period indicat ed that the di stribution of H. axyridis trapped in a 24-h period was not associated with the distribution of aphids for any sampling date. The distribution of H. axyridis on plants was not associated with the distribution of H. axyridis trapped in a 24-h period for any sampling date, but the values of had a trend where increased in value between 06-August and 03-September (Table 4-12). Analyses to test for lags in the data from w eekly sampling indicated th at the dist ribution of aphids on 06-August was associat ed with the distribution of H. axyridis per plant on 13-August

PAGE 75

75 (Table 4-13). The values of for the previous set of associa tion tests did not show an obvious temporal trend. When the distribution of H. axyridis trapped in a 24-h period was paired with the distribution of aphids from the previous w eek, there was a significant association between the distributions of aphids on 01-October vs. H. axyridis trapped in a 24-h period on 08-October (Table 4-14). The values of for the association tests of H. axyridis trapped in a 24-h period and aphids from the previous week decrease d from 06-August through 03-September (dates represented by aphid sampling date ) (Table 4-14). Results from the spatial analyses assessing the delayed trapping of H. axyridis trapped in a 24-h period wi th respect to the number of H. axyridis on a plant (summed) showed a significant association between H. axyridis on plants on 03-September and H. axyridis trapped in a 24-h period on 10-Se ptember (Table 4-15). The values of for the data of H. axyridis trapped in a 24-h period with respect to the number of H. axyridis on a plant (summed) did not show an obvious temporal trend. Tests assessing the spatial association of H. axyridis trapped over a 3-d period with the distribution of aphids indicated that the distribution aphids and H. axyridis trapped over a 3-d period were not spatially associ ated for any pairing before or after aphid sampling(data not shown). Harmonia axyridis trapped over a 3-d period for 1416-Septermber were spatially associated with H. axyridis on plants for 17-September (Table 4-16). H. axyridis trapped over a 3-d period was spatially associated to the H. axyridis trapped in a 24-h period for the pairings of: 1) 24-h period on 24-September and 3-d period fr om the 25-27-September; 2) 24-h period on 15October and 3-d period from 16-18-October (Table 4-17). The values of did not have a temporal trend for any of the sets of association analyses using the H. axyridis trapped over a 3-d period data.

PAGE 76

76 2008 Data Phenology. In 2008, the peak number of crapemyrtle aphids per plant occurred on 18August, and declined rapidly between the samp ling dates of the 18-25-August (Figure 4-11). The number of H. axyridis observed on a plant through beat sampling had a peak on 18-August, but the number of H. axyridis sampled on a plant through be ating was low throughout the 2008 season (Figure 4-11). Trap catch measured as H. axyridis per trap/day was the highest on 25August, and the number of H. axyridis per trap/day was greater for traps located inside the plot as compared to traps located out side the plot (figure 4-12). Analysis of variance. Results from the repeated measur es analyses indicated that time was the only significant factor for the analyses of: aphids per pl ant (F = 260; df = 1, 733; P < 0.01; Table 4-18), H. axyridis per plant; (F = 5.99; df = 1, 194; P = 0.015; Table 4-19), and H. axyridis trapped in a 24-h period (F = 24.8; df = 1, 176; P < 0.001; Table 4-20). Results from the repeated measures analysis of H. axyridis trap data for all days of trapping indicated differences in time (F= 14.97, df = 1, 789, P < 0.001), cultivar, where Tuskegee had the greatest number of H. axyridis per plant and Carolina Beau ty had the least number of H. axyridis per plant, (F = 3.11; df = 13, 41; P = 0.03), and there was a time cultivar interacti on (F = 3.12, df = 13, 789, P < 0.001; Table 4-21; Table 4-22). Regression analyses. Regression analyses for the 2008 data indicated that there was a correlation between the number of H. axyridis per plant and the number of aphids per plant (R2 = 0.49; figure 4-13). Results from the regression of H. axyridis trapped in a 24-h period vs. H. axyridis per plant showed a strong corre lation between the two factors (R2 = 0.74; Figure 4-14). Data for traps located inside and outside the plot were square root tran sformed and the square root of H. axyridis trapped outside the plot had a positive correlation with the square root of H. axyridis trapped inside the plot (R2 = 0.73; Figure 4-15). Analyses testing the rela tionships of:

PAGE 77

77 H. axyridis trapped in a 24-h period vs. H. axyridis per plant, H. axyridis trapped in a 24-h period vs. aphids per plant; and the proportion of H. axyridis trapped in a 24-h period vs. aphids per plant were not significant (data not shown). Cluster analyses. In 2008, the distribution of aphids wa s spatially clustered for all weekly sampling dates between 24-June and 23-Sept ember except for 11-August and 03-September (Table 4-23). Results from the cluster analyses on H. axyridis indicated that the H. axyridis per plant data were clustered on 19-August and 26-A ugust (Table 4-23). Clus ter analyses conducted on the data for H. axyridis trapped in a 24-h period indi cate that the distribution of H. axyridis was not clustered for any of the weekly sampli ng dates (data not shown) The distribution of H. axyridis trapped over a 3-d period was clustered for the trapping period of 29-August through 01-September (Table 4-24). When the aphid data was summed for plants adjacent to a trap and assigned the trap coordinates, th e distribution of aphids was cl ustered for the weekly sampling dates of 08-22-July, 25-August, a nd 10-September (Table 4-25). Harmonia axyridis data that were summed and pooled by trap showed si gnificant clustering on 11-August and 19-August (Table 4-25). Association analyses. Association analyses to test for a spatial association of aphids and H. axyridis on a plant indicated that the distributions of aphids and H. axyridis were significantly associated for the following sampling dates: 02, 08-July, 19-August th rough 03-September, and the sampling dates between 23-September and 06October (Table 4-26). The values of in the association analyses for aphids and H. axyridis per plant did not have an obvious trend over time. Because of the problem with weather that affect ed the ability to check sticky traps on the exact day of plant sampling, results for the association analyses compari ng the distributions of aphids and H. axyridis to H. axyridis trapped in a 24-h period are refe renced using the plant sampling

PAGE 78

78 date. The distribution of aphids on a plant and H. axyridis trapped in a 24-h period was spatially associated for the sampling dates of: 14, 22-July 19-August, and 10-September, and the values of did not show an obvious temporal trend (Table 4-27). Results from the association analyses of H. axyridis on plants adjacent to traps and H. axyridis trapped in a 24-h period indicated that the spatial distributions were associat ed on 03-September, and the values of showed a tendency to increase in value from 22-J uly through 03-September (Table 4-28). Analyses that tested for lags in the 2008 data from weekly sampling of the plot indicated that there was a significant association between the distributions of aphids and H. axyridis per plants for the following pairs of data: 1) aphids on 24-June vs. H. axyridis on 02-July, 2) aphids on 02-July and H. axyridis on 08-July, 3) aphids on 11-August and H. axyridis on 19-August, 4) aphids on 19-August and H. axyridis on 26-August, 5) aphids on 26-August and H. axyridis 03September, 6) aphids on 16-September and H. axyridis on 23-September, 7) aphids on 01October and H. axyridis on 06-October, 8) aphids on 13-October and H. axyridis on 20-October (Table 4-29). Aphids were spatially associated with H. axyridis trapped in a 24-h period for the tests pairing aphids on 28-July with H. axyridis on 04-August and aphids on 04-August with H. axyridis on 11-August (Table 4-30). Re sults from association analyses testing lags in the spatial association of H. axyridis per plant and H. axyridis trapped in a 24-h period indicated that the distribution of H. axyridis on a plant on 08-July was associ ated with the distribution of H. axyridis trapped in a 24-h period on 14-July (Table 4-31). The sampling of traps over 3-d periods was interrupted in 2008 by the weather and tests fo r lags in the data fo r sampling dates between plant sampling were not performed. 2007 and 2008. The number of crapemyrtle aphids and H. axyridis observed between the years 2007 and 2008 was different. Peak crapemyrtl e aphid populations measured as aphids per

PAGE 79

79 10 leaves in 2008 were more than five times grea ter than the peak number of aphids per sample in 2007 (Figure 4-16). The number of H. axyridis recorded on a plant cannot directly be compared between the two years of sampling becau se the sampling techniques were not equal. Harmonia axyridis abundance measured in sticky trap s was lower in 2008 (Figure 4-17). Discussion Habitat Level Effects It is unknown if the low number of crapemyr tle aphids in 2007 was due to the abundance of H. axyridis or if the large number of aphids in 2008 was due to the absence of H. axyridis because other factors such as climatic conditi ons, other aphid populations in the landscape and overwintering survival are lik ely to affect the number of H. axyridis present in any given year. Climatic conditions between the two years were different, and average rainfall and average temperature data reported by the Florida Automa ted Weather Network (FAWN) site located at the research station in Quincy recorded that the average ra infall in 2007 from 01-May to 20August was 2.4 mm per day but was 3.7 mm per day in 2008. Rainfall data were averaged for the dates of 01-May through 20-A ugust to exclude the effects of Tropical Storm Fay. Aphid populations dramatically decrea sed during Tropical Storm Fay, which was likely caused by the wind and the more than 300 mm of rain that th e station received over a 3-d period. However, H. axyridis activity increased following Fay, which ma y have been caused by declining aphid populations in crapemyrtle and othe r habitats. Increased movement with decreasing prey density has been shown in the coccinellid species C. septempunctata and the large number of H. axyridis observed in the crapemyrtle plot in 2007 may al so be related to the low density of aphid prey (Bianchi and Werf 2004). Temperature may have also influenced aphid and H. axyridis populations. Models investigating the influence of increasing temperature on the pr edator prey dynamics of lady

PAGE 80

80 beetles have indicated that a rise in temperat ure favors coccinellid populations and leads to a decrease in aphid populations (Ski rvin et al. 1997). In the Skir vin et al. (1997) model, aphid reproduction reaches a maximum, but coccinellid developmental time continues to decrease resulting in higher numbers of beetles and lower aphid popula tions. Crapemyrtle aphid and H. axyridis data from my study support the findings of th e Skirvin et al. (1997) model in that the average daily temperature between 01-May and 20-August was 0.5 C greater in 2007 than in 2008, and there was a greater number of H. axyridis per plant in 2007 and a lower number of aphids per plant in 2007 in compar ison to the data for aphids and H. axyridis per plant in 2008. In both years, aphid populations were low when compared to the field susceptibility data of Mizell and Knox (1993). Mizell an d Knox (1993) recorded the peak in aphid populations for the years 1990 and 1991 at approximately 150 and 175 aphi ds per leaf, which translates to 1500 and 1750 aphids per sample. Data from Mizell and Knox (1993) were recorded before the widespread establishment of H. axyridis and H. axyridis has since become the dominant aphid predator in crapemyrtle and pecan (Mizell 2007). Harmonia axyridis and crapemyrtle aphids are from Southeast Asia, an d the introduction of H. axyridis would be equivale nt to a classical biological control program and may have reduced the abundance of crapemyrtle aphids in recent years. Results from this study indicate that H. axyridis per plant had a numerical response to the mean number of aphids per plant for both years of sampli ng. Although the values of the response (slopes of the regressions) cannot be co mpared directly because they represent two different methods of sampling, the mean number of H. axyridis on a plant was correlated to the mean number of aphids on a plant in 2007 and 2008. The re ason for the lower number of H. axyridis observed in 2008 is not know n, but may have been influenced by a number of factors

PAGE 81

81 such as aphid populations in other habitats, temperature, rainfall, or the number of H. axyridis that survived the winter from the previous years population. Cultivar Level Effects The number of H. axyridis trapped in sticky traps in a 24-h period was not different according to cultivar, but when the number of H. axyridis trapped was considered for all trapping dates, the effect of cultivar was significant for bo th years. Traps placed between plants from the cultivars Natchez and Tuskegee cons istently trapped higher numbers of H. axyridis The capture rate of traps between pl ants of Natchez and Tuskegee was not apparently related to the number of aphids or H. axyridis on plants immediately adjacent to the trap, and plant cues or signals may play a role in H. axyridis attraction or increased moveme nt in association with these cultivars. Both Natchez and Tuskegee are tall cultivars from the L. indica L. fauriei parentage and have a similar architecture when growing from pruned trunks. Shape may play a factor in attracting coccinellids and has not b een looked at in crapemyrtle. Mensah (1997) demonstrated that lady beetles respond to wavele ngths that correspond to foliage color, and that Coccinellidae were more attracted to hues of yellow. Harmonia axyridis foraging behavior may also change in response to the coloration of cr apemyrtle foliage. Sticky trap data must be considered carefully because they represent th e total number of visits during a time period, but plant counts represent th e number of insects present at a pr ecise moment in time. The sticky traps used in this experiment were not c onsistently reliable at correlating the number H. axyridis on a trap and the number of aphids or H. axyridis per plant, and this problem has also occurred for the use of sticky cards (Evans 2003, Stephens and Losey 2004). Spatial Relationships Spatial clustering and spatial associations differed in the number of occurrences between the two seasons, and these differenc es are likely to be caused by de nsity dependent interactions.

PAGE 82

82 Several models exploring density dependence an d predation rates of lady beetles have been proposed to account for the spatial distribution of Coccinellidae and their aphid prey (Kareiva and Odell 1987, Kindlmann and Dixon 1993, Kindl mann and Dixon 1999b). Kareiva and Odell (1987) proposed a model that used predator satiati on and turning rate to de scribe the spatial and temporal distribution of Coccinel lidae and their prey. In the Kareiva and Odell (1987) model, the speed of movement for each co ccinellid was held constant, but the turning rate while moving through the habitat increased with satiation. Thus, Coccinellids that were well fed turned a lot and did not move out of the aphid patch, while those that consumed a marginal amount of food turned less and dispersed a greater distance from the food source (Kareiva and Odell 1987). The data from 2007 represent a case where prey de nsity was low and under such conditions it would be difficult for a predator to become satiated, wh ich could explain why th e spatial distributions were not associated. Because non clustered data can still be associated using SADIE, predators should still be associated with th eir aphid prey if the prey density is large enough to retain them through satiation. Spatial distribution of prey coupled with prey abundance has been shown to influence the effectiveness of generalist natural enemies, and when prey are spatially clustered, they are more likely to show population growth in the presence of pred ators (Bommarco et al. 2007). Data from my experiment agree with th e Bommarco et al. (2007) model when comparing the number of times aphid populations were clus tered in 2007 compared to 2008. However, the absence of lady beetles may have allowed for more spatial clustering. Cause and effect relationships are difficult to prove and experime nts that manipulate aphi d populations and beetle abundance will be useful for assessing these re lationships. Future experiments examining a variety of aphid species over multiple habitats will also help determine if aphid populations

PAGE 83

83 outside of the area being studi ed are interacting and affec ting the number of lady beetles observed inside the crapemyrtle plot. Field Study Harmonia axyridis demonstrated a numerical respons e to aphid populations in 2007 and 2008, but the response was on a population or habita t level. Temporal and spatial responses were not correlated on a plant by plant level, and the number of H. axyridis trapped in a 24-h period was not consistently corr elated with the number of H. axyridis or aphids present on plants. However, when aphid populations were low, th ere was a marginal positive correlation between the number of H. axyridis trapped in a 24-h period and the nu mber of aphids on a plant for both years of sampling. Sticky trap results may be more useful when aphid populations are low. Sticky traps used in this study were useful for monitoring the species diversity of Coccinellidae in the landscape and captured lady beetle species that were not commonly recorded from visual or beat sampling of plants. In a addition to H. axyridis the sticky traps captured specimens of: H. convergens C. septempunctata O. v-nigrum C. maculata Cycloneda spp., and Neoharmonia spp. Future experiments investigating aphid natu ral enemies in crapemyrtle should consider the interactions among the different sp ecies of Coccinellidae to assess possible negative interactions such as intraguild predation or competition.

PAGE 84

84 Table 4-1. Crapemyrtle cultivars planted in th e block of crapemyrtle used for aphid and lady beetle sampling. Cultivar Mature plant height Parentage and pedigree Erisyphe lagerstroemiae susceptibility Sarucallis kahawaluokalani susceptibility Acoma Semidwarf ( L. indica ) 'Pink Ruffles' ( L. indica L. fauriei ) 'Basham's Party Pink' Resistant Moderately Susceptible Apalachee Medium ( L. indica ) 'Azuka Dwarf Hybrid' L. fauriei Resistant Susceptible Biloxi Tall [ L. indica 'Dwarf Red' L. fauriei ] [ L. indica 'Low Flame' L. fauriei ] Resistant Susceptible Byers Wonderful White Tall Pure L. indica Susceptible Moderately Susceptible Carolina Beauty Medium Pure L. indica Susceptible Moderately Susceptible Lipan Medium [ L. indica L. fauriei ) 'Pink Lace' L. fauriei ] [( L. indica 'Red' L. indica 'Carolina Beauty') ( L. indica L. fauriei ) Basham's Party Pink] Resistant Moderately Susceptible Miami Tall [ L. indica 'Pink Lace L. fauriei ] [ L. indica 'Firebird' ( L. indica L. fauriei ) seedling unknown cultivar] Tolerant Moderately Resistant Natchez Tall L. indica 'Pink Lace' L. fauriei Tolerant Moderately Resistant Osage Medium [ L. indica 'Dwarf Red' L. fauriei ] [ L. indica 'Pink Lace' L. fauriei ] Resistant Moderately Susceptible Sioux Medium [ L. indica 'Tiny Fire' ( L. indica L. fauriei ) seedling] [( L. indica 'Pink Lace' L. fauriei ) ( L. indica ) 'Catawba'] Resistant Moderately Resistant Tonto Semidwarf [( L. indica 'Pink Lace' L. fauriei ) L. indica 'Catawba'] [( L. indica ) 'Tuscarora' ( L. indica L. fauriei ) 'Basham's Party Pink' ( L. indica ) 'Cherokee'] Resistant Moderately Susceptible Tuscarora Tall ( L. indica L. fauriei ) 'Basham's Party Pink' ( L. indica ) 'Cherokee' Tolerant Moderately Susceptible Tuskegee Tall ( L. indica ) 'Dallas Red' ( L. indica L. fauriei ) 'Bashams Party Pink' Resistant Moderately Resistant Yuma Medium [( L. indica ) 'Pink Lace' ( L. fauriei )] [ L. amabilis 'Makino' L. indica 'Hardy Light Pink' L. indica 'Red'] Resistant Moderately Susceptible

PAGE 85

85 Table 4-2. Analysis of variance results for the 2007 repeated meas ures analysis for aphids per plant. Effect Num df Den df F Pr > F Time 1 1485 213.44 0.0001 Edge 1 1485 3.21 0.0735 Cultivar 13 1485 3.31 0.0001 Time Edge 1 1485 2.32 0.1279 Time Cultivar 13 1485 2.91 0.0003 Table 4-3. LS means and mean separation for th e effect of cultivar in the 2007 analysis of variance for aphids per plant. Tukey groupings with different letter s are different at P < 0.05. Cultivar LS meanSE Tukey Kramer grouping Apalachee 1.1654 0.06923 A Tuscarora 0.9061 0.07429 AB Lipan 0.8894 0.07241 AB Carolina Beauty 0.8485 0.07467 B Tuskegee 0.8463 0.07402 B Byers Wonderful White 0.8104 0.08748 BC Biloxi 0.7429 0.07178 BC Acoma 0.7412 0.07891 BC Natchez 0.6965 0.07799 BC Miami 0.6844 0.07241 BC Tonto 0.6831 0.07731 BC Osage 0.6727 0.07393 BC Sioux 0.6297 0.08255 BC Yuma 0.5073 0.08021 C Table 4-4. Analysis of variance results fo r the 2007 repeated m easures analysis for H. axyridis per plant. Effect Num df Den df F Pr > F Time 1 779 70.21 0.0001 Edge 1 779 2.23 0.1358 Cultivar 13 779 1.04 0.4117 Time Edge 1 779 2.28 0.1312 Time Cultivar 13 779 0.90 0.5494

PAGE 86

86 Table 4-5. Analysis of variance results fo r the 2007 repeated m easures analysis for H. axyridis trapped in a 24-h period. Effect Num df Den df F Pr > F Time 1 379 12.20 0.0005 Edge 1 41 0.08 0.7734 Cultivar 13 41 1.71 0.0958 Time Edge 1 379 0.30 0.5832 Time Cultivar 13 379 1.72 0.0551 Table 4-6. Analysis of variance results fo r the 2007 repeated m easures analysis for H. axyridis in traps for all sampling dates regardless of trapping duration. Effect Num df Den df F Pr > F Time 1 1760 173.23 0.0001 Edge 1 41 0.28 0.5993 Cultivar 13 41 2.65 0.0087 Time Edge 1 1760 0.27 0.6010 Time Cultivar 13 1760 2.65 0.0011 Table 4-7. LS means and mean separation for effect of cultivar in the 2007 analysis of H. axyridis trapped for all sampling dates regard less of trapping duration. Tukey groupings with different lett ers are different at P < 0.05. Cultivar LS mean SE Tukey Kramer grouping Byers Wonderful White 1.0099 0.05426 A Tuskegee 0.9046 0.05984 AB Natchez 0.8949 0.05806 AB Yuma 0.8856 0.05914 AB Apalachee 0.8733 0.05723 AB Acoma 0.8523 0.0577 AB Tonto 0.8016 0.05846 AB Osage 0.8004 0.056 AB Miami 0.7588 0.05839 AB Tuscarora 0.7368 0.06493 AB Carolina Beauty 0.7256 0.05779 AB Lipan 0.7075 0.05763 B Biloxi 0.7034 0.06033 B Sioux 0.6086 0.06147 B

PAGE 87

87 Table 4-8. Probability of spatia l clustering for the response vari ables sampled once per week in 2007. Date Aphids H. axyridis per plant H. axyridis trapped in a 24-h period 06-Aug 0.0099 0.2241 0.7293 13-Aug 0.2164 0.0704 0.0357 20-Aug 0.2093 0.1152 0.6167 27-Aug 0.0663 0.8120 0.2616 03-Sep 0.5418 0.0471 0.4295 10-Sep 0.2176 0.3892 0.3248 17-Sep 0.1533 0.3166 0.0900 24-Sep 0.0582 0.5953 0.0435 01-Oct 0.0481 0.0039 0.7169 08-Oct 0.2606 0.0101 0.6626 15-Oct 0.4059 0.0217 0.1994 22-Oct 0.2957 0.0377 0.0002 29-Oct 0.5140 0.0544 0.1386 05-Nov 0.0288 0.5937 0.4454 12-Nov 0.0739 0.2282 0.0154 Table 4-9. Probability of spa tial clustering for the 2007 data created by summing the number of aphids and H. axyridis sampled on plants adjacent to traps using trap coordinates. Date Aphids trap H. axyridis trap 06-Aug 0.1374 0.7329 13-Aug 0.2634 0.1664 20-Aug 0.3573 0.7776 27-Aug 0.1076 0.4309 03-Sep 0.2862 0.2095 10-Sep 0.7400 0.5234 17-Sep 0.8336 0.3865 24-Sep 0.2693 0.6388 01-Oct 0.2911 0.0828 08-Oct 0.1188 0.1699 15-Oct 0.0384 0.2732 22-Oct 0.6345 0.5355 29-Oct 0.0871 NA 05-Nov 0.0122 0.8974 12-Nov NA NA NA not applicable because the data set c ontained all zeros or only one response.

PAGE 88

88 Table 4-10. Probability of spatial clustering for H. axyridis trapped in sticky traps over a 3-d period in 2007. Dates of trapping Pr 07-09-Aug 0.4731 10-12-Aug 0.6965 14-16-Aug 0.0680 17-19-Aug 0.2782 21-23-Aug 0.0836 24-26-Aug 0.6626 28-30-Aug 0.7302 31-Aug-02-Sep 0.4847 04-06-Sep 0.1084 07-09-Sep 0.5006 11-13-Sep 0.0288 14-16-Sep 0.4248 18-20-Sep NA 21-23-Sep NA 25-27-Sep 0.0784 28-30-Sep 0.1269 02-04-Oct 0.1200 05-07-Oct 0.2842 09-11-Oct 0.1342 12-14-Oct 0.0059 16-18-Oct 0.0055 19-21-Oct 0.0040 23-25-Oct 0.0007 26-28-Oct 0.0171 30-Oct-01-Nov 0.0514 02-04-Nov 0.1284 06-08-Nov 0.2604 09-11-Nov 0.8726 NA Not applicable, traps were not used due to abundance of love bugs, Plecia nearctica

PAGE 89

89 Table 4-11. Spatial associati ons for the distributions of H. axyridis per plant and aphids per plant in 2007. Date Dutilleul adjusted Pr 06-Aug 0.1122 0.0459 13-Aug 0.0843 0.1437 20-Aug 0.0221 0.4198 27-Aug -0.0318 0.6565 03-Sep 0.0318 0.4198 10-Sep -0.0684 0.8626 17-Sep -0.0669 0.8244 24-Sep -0.0439 0.6718 01-Oct 0.0978 0.2061 08-Oct -0.1329 0.8015 15-Oct 0.1241 0.1182 22-Oct -0.0174 0.6107 29-Oct -0.0979 0.8550 05-Nov -0.0141 0.4504 12-Nov 0.4429 0.0153 Table 4-12. Spatial association of H. axyridis sampled on plants adjacent to sticky traps and H. axyridis trapped in a 24-h period in 2007. Date Dutilleul adjusted Pr 06-Aug -0.0205 0.5563 13-Aug 0.0099 0.4768 20-Aug 0.0763 0.2848 27-Aug 0.1002 0.2649 03-Sep 0.1940 0.1126 10-Sep -0.1467 0.8675 17-Sep 0.1528 0.1589 24-Sep 0.0305 0.0450 01-Oct 0.1085 0.2252 08-Oct 0.2595 0.1457 15-Oct 0.0858 0.2781 22-Oct 0.0282 0.4570 29-Oct NA NA 05-Nov 0.1056 0.2450 12-Nov NA NA NA Not applicable because one or more of the datasets contains all zeros

PAGE 90

90 Table 4-13. Spatial association of H. axyridis per plant and aphids per plant using sampling dates separated by 1 wk in 2007. Aphid date H. axyridis plant date Dutilleul adjusted Pr 06-Aug 13-Aug 0.1971 0.0065 13-Aug 20-Aug 0.0199 0.3896 20-Aug 27-Aug 0.1054 0.0390 27-Aug 03-Sep -0.0616 0.8052 03-Sep 10-Sep 0.0349 0.3896 10-Sep 17-Sep 0.0205 0.4091 17-Sep 24-Sep -0.0736 0.8506 24-Sep 01-Oct -0.0504 0.6039 01-Oct 08-Oct -0.1391 0.7922 08-Oct 15-Oct -0.1870 0.9026 15-Oct 22-Oct -0.0827 0.7468 22-Oct 29-Oct 0.0212 0.3961 29-Oct 05-Nov -0.0972 0.8117 05-Nov 12-Nov 0.4173 0.0844 Table 4-14. Spatial association of H. axyridis trapped in a 24-h pe riod and aphids using sampling dates separated by 1 wk in 2007. Aphid date H. axyridis trap date Dutilleul adjusted Pr 06-Aug 13-Aug 0.2184 0.1364 13-Aug 20-Aug 0.1421 0.1763 20-Aug 27-Aug 0.1335 0.1623 27-Aug 03-Sep 0.0210 0.4221 03-Sep 10-Sep -0.0058 0.5455 10-Sep 17-Sep 0.1555 0.0844 17-Sep 24-Sep 0.0320 0.4610 24-Sep 01-Oct -0.0485 0.6234 01-Oct 08-Oct 0.2975 0.0195 08-Oct 15-Oct -0.0464 0.5779 15-Oct 22-Oct 0.0939 0.3182 22-Oct 29-Oct -0.0725 0.7143 29-Oct 05-Nov 0.0586 0.4545 05-Nov 12-Nov 0.0937 0.2208

PAGE 91

91 Table 4-15. Spatial association of H. axyridis trapped in a 24-h period and H. axyridis on plants using sampling dates separated by 1 wk in 2007. H. axyridis plant date H. axyridis trap date Dutilleul adjusted Pr 06-Aug 13-Aug -0.2012 0.8636 13-Aug 20-Aug -0.2193 0.9481 20-Aug 27-Aug -0.0201 0.5714 27-Aug 03-Sep -0.0546 0.6688 03-Sep 10-Sep 0.3723 0.0130 10-Sep 17-Sep 0.1913 0.0844 17-Sep 24-Sep -0.0927 0.7143 24-Sep 01-Oct -0.1261 0.7597 01-Oct 08-Oct 0.1248 0.1429 08-Oct 15-Oct 0.1314 0.2597 15-Oct 22-Oct 0.0917 0.3052 22-Oct 29-Oct 0.0836 0.2857 29-Oct 05-Nov NA NA 05-Nov 12-Nov 0.2654 0.1104 Table 4-16. Spatial association of H. axyridis on plants and H. axyridis trapped over a 3-d period before plant sampling in 2007. H. axyridis plant date H. axyridis trap dates Dutilleul adjusted Pr 13-Aug 10-12-Aug -0.0330 0.6299 20-Aug 17-19-Aug -0.0007 0.5260 27-Aug 24-26-Aug 0.2231 0.0390 03-Sep 31-Aug-02-Sep 0.1137 0.2273 10-Sep 07-09-Sep 0.2319 0.0260 17-Sep 14-16-Sep 0.3072 0.0130 24-Sep 21-23-Sep NA NA 01-Oct 28-30-Sep 0.2269 0.0455 08-Oct 05-07-Oct 0.1205 0.2532 15-Oct 12-14-Oct 0.3608 0.0325 22-Oct 19-21-Oct 0.0385 0.3052 29-Oct 26-28-Oct NA NA 05-Nov 02-04-Nov 0.2681 0.2143 12-Nov 09-11-Nov NA NA NA Not applicable Analysis not valid due to missing data or all zeros.

PAGE 92

92 Table 4-17. Spatial association of H. axyridis on plants and H. axyridis trapped over a 3-d period after plant sampling in 2007. H. axyridis plant date H. axyridis trap dates Dutilleul adjusted Pr 13-Aug 14-16-Aug -0.0261 0.5325 20-Aug 21-23-Aug -0.0312 0.6299 27-Aug 28-30-Aug 0.1291 0.1753 03-Sep 04-06-Sep 0.2883 0.0584 10-Sep 11-13-Sep -0.0197 0.6104 17-Sep 18-20-Sep NA NA 24-Sep 25-27-Sep 0.3400 0.0065 01-Oct 02-04-Oct 0.1821 0.0909 08-Oct 09-11-Oct 0.2763 0.1429 15-Oct 16-18-Oct 0.3803 0.0065 22-Oct 23-25-Oct 0.3801 0.0260 29-Oct 30-Oct-01-Nov NA NA 05-Nov 06-08-Nov 0.0981 0.2273 NA Not applicable Analysis not valid due to missing data or all zeros. Table 4-18. Analysis of variance results for the 2008 repeated meas ures analysis for aphids per plant. Effect Num df Den df F Pr > F Time 1 733 260.10 0.0001 Edge 1 733 2.78 0.0960 Cultivar 13 733 1.54 0.0990 Time Edge 1 733 2.78 0.0958 Time Cultivar 13 733 1.54 0.0991 Table 4-19. Analysis of variance results fo r the 2008 repeated m easures analysis for H. axyridis per plant. Effect Num df Den df F Pr > F Time 1 194 5.99 0.0153 Edge 1 194 0.09 0.7586 Cultivar 13 194 1.51 0.1178 Time Edge 1 194 0.09 0.7587 Time Cultivar 13 194 1.51 0.1176

PAGE 93

93 Table 4-20. Analysis of variance results fo r the 2008 repeated m easures analysis for H. axyridis trapped in a 24-h period. Effect Num df Den df F Pr > F Time 1 176 24.18 0.0001 Edge 1 41 0.01 0.9252 Cultivar 13 41 0.89 0.5686 Time Edge 1 176 0.01 0.9250 Time Cultivar 1 176 0.01 0.9250 Table 4-21. Analysis of variance results fo r the 2008 repeated m easures analysis for H. axyridis trapped on all sampling dates, re gardless of trapping duration. Effect Num DF Den DF F Value Pr > F Time 1 789 14.97 0.0001 Edge 1 41 0.36 0.5500 Cultivar 13 41 3.11 0.0027 Time Edge 1 789 0.36 0.5462 Time Cultivar 13 789 3.12 0.0001 Table 4-22. LS means and mean separation for th e effect of cultivar in the 2008 analysis of H. axyridis trapped for all sampling dates, rega rdless of trapping duration. Tukey groupings with different lett ers are different at P < 0.05. Cultivar LS mean SE Tukey Kramer grouping Tuskegee 1.0752 0.08491 A Osage 1.0285 0.0786 AB Natchez 0.9542 0.08041 ABC Biloxi 0.9452 0.08108 ABC Apalachee 0.8632 0.08613 ABC Acoma 0.7711 0.09044 ABC Byers Wonderful White 0.7252 0.08889 ABC Tonto 0.6764 0.09147 ABC Tuscarora 0.6728 0.09316 ABC Miami 0.6721 0.09597 ABC Lipan 0.6376 0.09011 ABC Sioux 0.6369 0.08751 ABC Yuma 0.6214 0.09134 BC Carolina Beauty 0.5728 0.08731 C

PAGE 94

94 Table 4-23. Probability of spatia l clustering for the response vari ables sampled once per week in 2008. Date Aphid cluster Pr H. axyridis plant cluster Pr H. axyridis in a 24-h period Pr H. axyridis in a 24-h period cluster Pr 18-Jun 0.6204 NA NA NA 24-Jun 0.0193 NA NA NA 02-Jul 0.0040 0.8679 NA NA 08-Jul 0.0026 0.4718 07-Jul 0.3072 14-Jul 0.0208 0.3238 14-Jul 0.8884 22-Jul 0.0070 0.4689 21-Jul 0.6526 28-Jul 0.0208 0.6838 28-Jul 0.1042 04-Aug 0.0788 0.6677 04-Aug 0.9621 11-Aug 0.1357 0.6624 11-Aug 0.6967 19-Aug 0.0471 0.0003 18-Aug 0.2943 26-Aug 0.0176 0.0005 25-Aug 0.6942 03-Sep 0.2128 0.1197 02-Sep 0.5157 10-Sep 0.0206 0.2076 09-Sep 0.4424 16-Sep 0.0736 0.3498 15-Sep 0.0685 23-Sep 0.0412 0.0885 NA NA 01-Oct 0.1456 0.5894 NA NA 06-Oct 0.1446 0.9703 NA NA 13-Oct 0.2789 0.1880 NA NA 20-Oct 0.0823 0.2016 NA NA NA Data not collected Table 4-24. Probability of spa tial clustering for the 2008 data created by summing the number of aphids and H. axyridis sampled on plants adjacent to traps using trap coordinates. Date Aphid cluster Pr H. axyridis cluster Pr 08-Jul 0.0153 0.7893 14-Jul 0.0419 0.4347 22-Jul 0.0315 0.4609 28-Jul 0.0865 0.6035 04-Aug 0.1306 0.6159 11-Aug 0.1753 0.6890 19-Aug 0.0848 0.0023 26-Aug 0.0236 0.0025 03-Sep 0.4046 0.1438 10-Sep 0.0471 0.1899 16-Sep 0.1371 0.4801

PAGE 95

95 Table 4-25. Probability of spatial clustering for H. axyridis sampled in traps on dates other than weekly sampling in 2008. Dates of trapping Pr 17-23-Jun 0.7717 24-29-Jun 0.6647 01-06-Jul 0.2046 08-13-Jul 0.3727 15-17-Jul 0.5284 18-20-Jul 0.0908 22-24-Jul 0.4171 25-27-Jul 0.7099 29-31-Jul 0.4409 01-03-Aug 0.4257 05-07-Aug 0.2950 08-10-Aug 0.2269 12-17-Aug 0.1473 19-24-Aug 0.2531 26-28-Aug 0.2224 29-Aug-01-Sep 0.0421 03-08-Sep 0.3635 10-11-Sep 0.3446 12-14-Sep 0.3064 16-22-Sep 0.0714 23-29-Sep 0.9365 30-Sep-05-Oct 0.9279 6-12-Oct 0.7074 13-19-Oct 0.3442

PAGE 96

96 Table 4-26. Spatial associati ons for the distributions of H. axyridis per plant and aphids per plant in 2008. Date Dutilleul adjusted Pr 02-Jul 0.2881 0.0065 08-Jul 0.2528 0.0065 14-Jul -0.2823 0.8831 22-Jul -0.0983 0.8506 28-Jul 0.1772 0.0649 04-Aug -0.0057 0.5909 11-Aug 0.0556 0.2987 19-Aug 0.5161 0.0065 26-Aug 0.2328 0.0065 03-Sep 0.3939 0.0195 10-Sep 0.0409 0.3247 16-Sep -0.1165 0.8182 23-Sep 0.2168 0.0455 01-Oct 0.3880 0.0065 06-Oct 0.2122 0.0260 13-Oct 0.2248 0.1169 20-Oct 0.3939 0.0455 Table 4-27. Spatial associations for the distribution of aphids sampled on plants adjacent to sticky traps and H. axyridis trapped in a 24-h period in 2008. Aphid date Trap date Dutilleul adjusted Pr 08-Jul 07-Jul 0.0238 0.4675 14-Jul 14-Jul 0.3907 0.0195 22-Jul 21-Jul 0.3733 0.0130 28-Jul 28-Jul 0.2263 0.0909 04-Aug 04-Aug 0.0721 0.0519 11-Aug 11-Aug 0.0531 0.4351 19-Aug 18-Aug 0.2874 0.0195 26-Aug 25-Aug 0.0106 0.4146 03-Sep 02-Sep -0.0050 0.5065 10-Sep 09-Sep 0.3223 0.0130 16-Sep 15-Sep 0.1795 0.1948

PAGE 97

97 Table 4-28. Spatial associations of H. axyridis sampled on plants adjacent to sticky traps and H. axyridis trapped in a 24-h period in 2008. H. axyridis plant date H. axyridis trap date Dutilleul adjusted Pr 08-Jul 07-Jul 0.1109 0.1948 14-Jul 14-Jul 0.1336 0.2013 22-Jul 21-Jul -0.4531 0.9945 28-Jul 28-Jul -0.1243 0.7987 04-Aug 04-Aug -0.0776 0.7338 11-Aug 11-Aug -0.0969 0.7662 19-Aug 18-Aug 0.1358 0.1304 26-Aug 25-Aug 0.1537 0.1688 03-Sep 02-Sep 0.2309 0.0195 10-Sep 09-Sep 0.1553 0.1364 16-Sep 15-Sep -0.0212 0.5455 Table 4-29. Spatial associations of H. axyridis per plant and aphids per plant using sampling dates separated by 1 wk in 2008. Aphid date H. axyridis plant date Dutilleul adjusted Pr 24-Jun 02-Jul 0.2287 0.0195 02-Jul 08-Jul 0.2243 0.0195 08-Jul 14-Jul -0.1368 0.7078 14-Jul 22-Jul -0.1931 0.8636 22-Jul 28-Jul 0.0873 0.2013 28-Jul 04-Aug -0.0092 0.5325 04-Aug 11-Aug -0.0966 0.7468 11-Aug 19-Aug 0.5272 0.0065 19-Aug 26-Aug 0.3972 0.0065 26-Aug 03-Sep 0.1594 0.0390 03-Sep 10-Sep 0.0108 0.4675 10-Sep 16-Sep 0.0062 0.4545 16-Sep 23-Sep 0.3278 0.0065 23-Sep 01-Oct 0.1730 0.1688 01-Oct 06-Oct 0.2128 0.0130 06-Oct 13-Oct 0.2231 0.0519 13-Oct 20-Oct 0.4101 0.0130

PAGE 98

98 Table 4-30. Spatial associations of H. axyridis trapped in a 24-h pe riod and aphids using sampling dates separated by 1 wk in 2008. Aphid date H. axyridis trap date Dutilleul adjusted Pr 08-Jul 14-Jul 0.3120 0.0325 14-Jul 21-Jul -0.0853 0.3388 22-Jul 28-Jul 0.3916 0.0130 28-Jul 04-Aug 0.2726 0.0130 04-Aug 11-Aug 0.1351 0.1753 11-Aug 18-Aug -0.0603 0.7143 19-Aug 25-Aug 0.1290 0.2243 26-Aug 02-Sep -0.0782 0.5584 03-Sep 09-Sep 0.1166 0.1818 10-Sep 15-Sep 0.0602 0.2857 Table 4-31. Spatial associations of H. axyridis trapped in a 24-h period and H. axyridis on plants using sampling dates separated by 1 wk in 2008. H. axyridis plant date H. axyridis trap date Dutilleul adjusted Pr 02-Jul 07-Jul 0.0921 0.2597 08-Jul 14-Jul 0.3404 0.0065 14-Jul 21-Jul -0.2465 0.9545 22-Jul 28-Jul -0.2801 0.9675 28-Jul 04-Aug 0.1351 0.1753 04-Aug 11-Aug 0.0927 0.2792 11-Aug 18-Aug -0.1524 0.8961 19-Aug 25-Aug 0.1216 0.1364 26-Aug 02-Sep 0.1233 0.2013 03-Sep 09-Sep 0.2708 0.0325

PAGE 99

99 02550 12.5Meters Bilo x i Carolina Beauty Lipan Siou x ApalacheeOsage Miami Byers Wonderful White Acoma Tuscarora Tuskegee Natchez Yuma Tonto Yuma Acoma Osage Tonto Bilo x i Carolina Beauty Acoma Siou x Apalachee Miami Lipan Tuscarora Tuskegee Byers Wonderful White Tuskegee Acoma Bilo x i Miami Siou x Natchez Apalachee Byers Wonderful White Tonto Tuscarora Lipan Carolina Beauty Osage Yuma Carolina Beauty Yuma Tuskegee Siou x Acoma Tuscarora Byers Wonderful White Natchez Lipan Miami ApalacheeBilo x i Osage Tonto Figure 4-1. Maps of the crapemyr tle plot used for insect sampli ng: A) satellite photo and B) cultivar placement with each cultivar na me representing four consecutive plants. B A

PAGE 100

100 02550 12.5Meters Figure 4-2. Trap locations of the 100 sticky traps used for sa mpling Coccinellidae. Traps 26, 63, and 100 were placed in habitats adjacent to the plot. Date in 2007 AugSepOctNovDec Insects per sample 0 1 2 3 4 5 6 Crapemyrle aphids H. axyridis Figure 4-3. Mean number of insects per sample ( SE) for crapemyrtle aphids and H. axyridis sampled on plants for all dates of sampling in 2007. 26 63 100

PAGE 101

101 Aug Sep Oct Nov Dec Insects per sample 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphids H. axyridis Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphids H. axyridis AB C Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis D Date in 2007 Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis E Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis F Figure 4-4. Mean number of insects per sample ( SE) on crapemyrtle plants for crapemyrtle aphids and H. axyridis in 2007 for the cultivars A) A coma, B) Apalachee, C) Byers Wonderful White, D) Biloxi, E) C arolina Beauty, F) Lipan, G) Miami, H) Natchez, I) Osage, J) Sioux, K) Tonto, L) Tus carora, M) Tuskegee, N) Yuma.

PAGE 102

102 Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis I Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis Aug Sep Oct Nov Dec Insects per sample 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis J K Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis Date in 2007 Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis L MN Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis G Aug Sep Oct Nov Dec 0 2 4 6 8 10 12 Crapemyrtle aphid H. axyridis H Figure 4-4. Continued

PAGE 103

103 Date in 2007 Aug Sep Oct Nov Dec H. axyridis per trap day 0 1 2 3 4 5 H. axyridis inside plot H. axyridis outside plot Figure 4-5. Mean number of H. axyridis ( SE) per trap day trapped inside and outside the crapemyrtle plot in 2007. Sqrt. aphids per plant 0.00.51.01.52.02.5 Sqrt. H. axyridis per plant 0.0 0.5 1.0 1.5 2.0 2.5 y = -0.94 + 0.63x R 2 = 0.87 Figure 4-6. Relationship of the square root of H. axyridis per plant and the square root of aphids per plant in 2007.

PAGE 104

104 Sqrt. aphids per plant 0.00.51.01.52.02.5 Sqrt. H. axyridis per trap 0.0 0.5 1.0 1.5 2.0 2.5 y = 0.88 +0.62x R2 = 0.43 Figure 4-7. Relationship of the square root of H. axyridis per trap and the square root of aphids per plant in 2007 Sqrt H. axyridis per plant 0.00.51.01.52.02.5 Sqrt H. axyridis per trap 0.0 0.5 1.0 1.5 2.0 2.5 y = 0.37 + 3.89x 2.23x2R2 = 0.68 Figure 4-8. Relationship of the square root of H. axyridis per trap and H. axyridis per plant in 2007.

PAGE 105

105 Mean number of aphids per plant 012345 Proportion of H. axyridis trapped 0.0 0.2 0.4 0.6 0.8 1.0 y = 0.76 0.15x R2 = 0.61 Figure 4-9. Relationshi p of the proportion of H. axyridis trapped inside the plot and aphids per plant in 2007; (mean H. axyridis on traps in plot / (mean H. axyridis on plants + mean H. axyridis on traps in plot) and th e mean number of aphids per plant in 2007.

PAGE 106

106 Log H. axyridis per trap inside -1.5-1.0-0.50.00.51.01.52.02.5 Log H. axyridis per trap outside -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 y = -0.19 + 0.65x R2 = 0.51 Figure 4-10. Relationship of the log H. axyridis trapped outside the plot and the log H. axyridis trapped inside the plot in 2007 Date in 2008 Jun Jul Aug Sep Oct Nov Number of insects per sample 0 5 10 15 20 25 30 35 Crapemyrtle aphids H. axyridis Figure 4-11. Mean number of insects per sample ( SE) for crapemyrtle aphids and H. axyridis sampled on plants for all dates of samplings in 2008.

PAGE 107

107 Date in 2008 May Jun Jul Aug Sep Oct Nov H. axyridis per trap day 0.0 0.5 1.0 1.5 2.0 2.5 3.0 H. axyridis inside plot H. axyridis outside plot Figure 4-12. Mean number of H. axyridis ( SE) per trap day trappe d inside and outside the crapemyrtle plot in 2008. Aphids per plant 051015202530 H. axyridis per plant 0.0 0.5 1.0 1.5 2.0 2.5 3.0 y = 0.09 + 0.026x R 2 = 0.49 Figure 4-13. Relationship of H. axyridis per plant and aphids per plant in 2008

PAGE 108

108 H. axyridis per plant 0.00.51.01.52.02.53.0 H. axyridis per trap 0.0 0.5 1.0 1.5 2.0 2.5 3.0 y = 0.27 + 2.49x R2 = 0.74 Figure 4-14. Relationship of H. axyridis per trap and H. axyridis per plant in 2008. Sqrt. H. axyridis per trap inside 0.00.51.01.52.02.5 Sqrt. H. axyridis per trap outside 0.0 0.5 1.0 1.5 2.0 2.5 y = 0.07 + 0.38x R 2 = 0.73 Figure 4-15. Relationship of the square root of H. axyridis trapped outside the plot and the square root of H. axyridis trapped inside the plot in 2008

PAGE 109

109 Date Jul Aug Sep Oct Nov Mean aphids per sample 0 5 10 15 20 25 30 35 2007 2008 Figure 4-16. Mean number of crapemyrtle aphids per sample ( SE) in 2007 and 2008. Date May Jun Jul Aug Sep Oct Nov Dec H. axyridis per trap day 0 1 2 3 H. axyridis 2007 H. axyridis 2008 Figure 4-17. Mean number of H. axyridis trapped on all 100 sticky traps ( SE) in 2007 and 2008.

PAGE 110

110 CHAPTER 5 GENERAL CONCLUSIONS Host Suitability and Host Preference Host suitability of crapemyrtle, Lagerstroemia spp., to crapemyrtle aphid attack was influenced by cultivar, parentage, and the mature plant height w ithin the sources of L. fauriei germplasm. Pure L. indica cultivars such as Carolina B eauty and Byers Wonderful White were the least suitable as hosts for the longevity and total fecundity of crapemyrtle aphids but are vulnerable to powdery mildew. In areas wher e powdery mildew is a concern, Natchez or Tuscarora may be a better choice for use in la ndscapes where powdery mildew is a concern because they are powdery mildew resistant and the least suitable hosts for crapemyrtle aphids among the L. indica L. fauriei cultivars that were tested in the host suitability experiment. During host suitability testing, I evaluated Lagerstroemia chekiangensis Cheng, Lagerstroemia speciosa L., and Lagerstroemia limii Merr. for their ability to support the reproduction and development of the crapemyr tle aphid (Appendix C). It was found that L. speciosa and L. limii were not suitable hosts for th e development and reproduction of crapemyrtle aphids, and when crap emyrtle aphids were placed on leaf disks from these species, there was no parturition, no nymphal development, and all aphids (nymphs and adults) were dead within 3 d. However, the crapemyrtle cu ltivar Princess is a hybridization of L. speciosa and L. indica and was suitable for crapemyrtle aphid de velopment and reproduction. Future studies examining the differences between L. speciosa and L. speciosa L. indica hybrids would be useful for determining what factors are used by crapemyrtle aphids during host discrimination. While studies on the constituents in the phloem are very useful, recent studies have shown that feeding and parturition stimulants may be present in other plant tissues besides the phloem, but research in this area is new and the mechanis ms behind parturition occu rring before an aphids

PAGE 111

111 stylets reach the phloem is not completely understood (Powell et al. 2006). Research on secondary plant compounds may be able to explore whole plant extracts adde d to artificial diets to test if a particular compound is required to in itiate feeding or parturit ion. Alternatively, amino acid profiles have been shown to influence hos t discrimination by aphids, and therefore should also be considered when examining the differences between L. speciosa and L. speciosa L. indica (Tosh et al. 2003). Tritrophic Interactions Data from the lacewing studies indicate that several lacewing li fe history attributes were associated with the crapemyrtle cultivar. Lacewi ng survivorship was associated with cultivar in the ad libitum experiment but was not found to be si gnificantly different among the cultivars used in the suboptimal diet experiment. Both lacewing experiments foun d that lacewing larval development time and adult dry mass were affect ed by the source of crapemyrtle aphids that were provisioned to the larvae. Lacewings f eeding on aphids with a higher mass from the L. indica L. fauriei parentage had a shorter larval deve lopmental time and higher pupal mass. However, pupal and total mortality were grea ter among larvae that were fed aphids from cultivars with a L. indica L. fauriei parentage, which may indicat e that there is a trade-off associated with larval development, pupal mass, and survivorship. If the effects on lacewing survivorship are due to toxic al lelochemicals as postulated by Mi zell et al. (2002), then I would expect to see mortality increase with increa sing prey levels. Although the results are not significant, a trend for increasing mortality with increased daily prey levels was seen for lacewings that were fed aphids from the cultivars Sioux, Carol ina Beauty, and Apalachee. If prey toxicity is additive over time, it is possible that the restricted diets provided enough nutrition for development but did not contain enough toxins to cau se significant mortality.

PAGE 112

112 Qualitative and quantitative differences in prey suitability may also be affected by aphid density, stage of development or factors asso ciated with induced resistance. New (1984) suggested that studies on aphi d predators should account for variables like population density and stage of development, and in the suboptimal di et experiment, crapemyrtle aphids were of the same age and relatively the same density. Absolute control of density was not possible because aphids have the ability to leave or move between disks, but aphids within a cultivar appeared to be uniform in size. Absolute age of the plant in years or seasonal age or the plant could also affect predators at the th ird trophic level. The ad libitum experiment was conducted in July, 2006 and the suboptimal diet experiment was conduc ted in September, 2008. Other factors that could affect predators, but were not tested, include: flowering stage, seed filling, pruning practices, drought stress, soil cond itions, or temperature. In addition to affecting predators through aphid prey, plants can affect predators directly through morphological or chemical attributes (P rice et al. 1980, Van Emden 1995, Francis et al. 2001). Lacewings and other insect predators have been found to pierce plant tissues in an effort to receive nutrition or water, and direct effects from the exposur e to plant material may have affected lacewing survivorship (Stoner 1970, Stoner et al. 1974, Ruberson et al. 1986, Legaspi et al. 1994). Alkaloids are known to ex ist in crapemyrtle leaf tissue, and attempts to feed or gain nutrition from leaves could resu lt in exposure to these toxins (Ferris et al. 1971b). Future experiments can be designed to test if exposure to plant material was a contributing factor to the higher mortality seen in the ad libitum diet experiment as compared to the restricted diets. Multitrophic interactions among crapemyrtle, crapemyrtle aphids, and lacewings are an important consideration if crapemyrtle is to be used as an augmentation crop in pecans. Future work on assessing the quality of a cultivar for use as an augmentation crop for the enhancement

PAGE 113

113 of aphid predators should account for prey abund ance and prey suitability. Results from this study indicate that crapemyrtle cu ltivars that produce large quanti ties of aphids may also produce aphids that are lower quality or toxic to natu ral enemies. Apalachee was a suitable and preferred host for crapemyrtle aphids and is co mmonly infested with aphids in the mid summer months. Although, lacewing larvae that were fed aphids from A palachee had a relatively short period of larval development and high pupal mass, th e total survivorship of lacewings that were fed aphids from Apalachee was among the lowest seen in the ad libitum diet experiment. In contrast, Byers Wonderful White was less suitab le, and lacewing larvae reared on aphids from this cultivar had one of the longest larval deve lopmental times and lowest pupal weights, but the survivorship of larva to adult was higher for B yers Wonderful White than for Apalachee. The use of Apalachee as an augmentative crop ma y be compromised by adverse effects on natural enemies. Results from the lacewing study indicate that there may be possible trade-offs in the developmental time, survivorship, and repr oductive potential of lacewings feeding on crapemyrtle aphids from different cultivars. New (1984) stated that an efficiency rating of lacewings (E) can be calculated by taking the pupal mass (Pm) and dividing by larval developmental time (Ld), which is then multiplied by 100 (Equation 5-1). E = Pm/Ld 100 (5-1) However, the efficiency rating does not account for the effects of mortality when multiple predators are used. The previous equation can be modified to account for larval mortality by using the mean pupal mass (Mpm) and mean larval development time (Mld), which is then multiplied by the percent survivorship (%Ls) inst ead of the arbitrary value of 100 (Equation 5-2). E = Mpm/Mld % Ls (5-2)

PAGE 114

114 Equation 5-2 expresses the efficiency in terms of how well a diet supports larval development, but augmentation crops need to su pport reproductive potential as well as larval development. Lacewings are not reproducti vely active upon emergence because of an incompletely developed reproductive system. Matu rity of the reproductive system is associated with mass of the adult and larger heavier adu lts can produce gametes sooner than their smaller counterparts (New 1984). Perhaps the reproductiv e potential (R) of an adult lacewing can be assessed by an equation similar to the efficiency equation by using the mean dry mass of adults (Mam), dividing the dry mass by the mean pupal developmental time (Mpd) and multiplying by the survivorship from pupa to adult (%Ps) (Equation 5-3). R = Mam/Mpd % Ps (5-3) Having a high efficiency index does not transl ate into having a high reproductive potential (Table 5-1). Ideally an adjustment can be us ed to combine Equation 52 and Equation 5-3 that could be used to express a cultivars potential to support the developm ent and reproduction of C. rufilabris but data concerning the dr y mass and its influence on C. rufilabris fecundity are needed to weight the effects of a combined equation. Spatial and Temporal Interactions Harmonia axyridis had a numerical response to aphid populations on a plot level and was spatially associated with crap emyrtle aphids when aphid populations were high in 2008. Temporal associations on a plot level are likely to be influenced by regional aphid populations. Understanding how aphid populations over an entire landscap e interact with H. axyridis populations is needed to assess if behavior or general movements of H. axyridis can be exploited to enhance biological control of aphids. Results from the spatial experiment indicate th at Natchez may have a ttributes that attract more lady beetles than other cultivars. Sticky traps between Natchez plants had one of the

PAGE 115

115 highest catches of H. axyridis for both 2007 and 2008. Aphid popul ations between the two years of testing were very different indicating that the attractive effe ct of Natchez may be density independent. Future experiments of the attraction of Natchez to H. axyridis could be conducted at the North Florida Research and Education Center because it contains two monoculture plantings of Natchez arranged in concentric circles. If Natch ez attracts lady beetle predators even when prey are absent, it could make a good banker plant for use in landscapes around homes allowing home owners the benefits of aphid predators without sacrific ing aesthetics. Multitrophic Interactions Crapemyrtle breeding programs have focuse d on producing cultivars that are disease resistant and have not considered insect resistan ce in the selection process. Powdery mildew resistant cultivars that are L. indica L. fauriei hybrids are more suitable, preferred and susceptible to aphid attack, but hybridized cultiv ars are the most commonly planted cultivars due to their resistance to powdery mildew. Th e widespread use of cultivars that are L. indica L. fauriei hybrids may increase the number of aphids in the landscape leading to more frequent and severe aphid outbreaks in the fu ture. Controlling aphid outbrea ks may also be dependent upon tritrophic interactions with pr edators through direct or indirect mechanisms. Apalachee is highly suitable, susceptible, and preferred as an aphid host, but lacewings that were fed aphids reared on Apalachee had the lowest survivorship in the ad libitum experiment. Because cultivar has the potential to aff ect insect natural enemies, trit rophic interactions should be considered when choosing a crap emyrtle cultivar for use as an augmentation crop. Cultivars used an augmentation crop should be attractive to insect natural enemies and some cultivars may be attractive to H. axyridis regardless of the aphid population. Based on the experiment in this study, I recommend the cultivars Byers Wonderful White, Natchez, or Sioux for use as an a ugmentative crop in pecans. Byers Wonderful

PAGE 116

116 White and Sioux are recommended because they were moderately susceptible to aphid attack but did not induce high levels of mortality, relativ e to the other cultivars, to lacewing larvae and pupae in the ad libitum experiment. Natchez and Byers Wonderful White may be useful as augmentation crops regardless of crapemyr tle aphid populations and could affect H. axyridis populations directly. Sticky traps placed between two plants of Natchez or Byers Wonderful White had some of the highest tr ap catches in both years of sa mpling. Results from this study indicate that crapemyrtle cultivar selection is an important consideration for homeowners or growers that wish to use crapemyrtle as an or namental or augmentation crop. Future breeding programs can use results from this study to create cu ltivars that are more re sistant to insect pests and more beneficial to in sect natural enemies.

PAGE 117

117 Table 5-1. Adjusted efficiency ratings and repr oductive potential of lacewings for the cultivars used in the restricted diet experiment and the values for the factors used in equations 3-2; 3-3 to generate the ratings. Cultivar Mpm Mld % Ls E Mam Mpd % Ps R Apalachee 6.50 9.89 90 59.17 1.50 8.33 88 15.83 Sioux 6.21 9.17 100 67.77 1.47 9.5 66 10.22 Carolina Beauty 6.17 9.85 100 62.59 1.50 9.23 85 13.86 Byers Wonderful White 5.83 10.35 100 56.36 1.40 9.57 95 13.83

PAGE 118

118 APPENDIX A GIS MAPS To help visualize the spatial di stribution of crapemyrtle aphids Sarucallis kahawaluokalani (Kirkaldy), Harmonia axyridis Pallas, spatial maps were generated by the date of sampling and response variable. Raw data for 2007 were mapped using ArcMap (ArcView version 9.1; Figures A-1-A-15). Data values are represented by using symbols th at have a circumference that is relative to the number of insects per sample and 0.3 m is equal to one insect. Data for H. axyridis and crapemyrtle aphids was mapped using th e same procedures as in 2007, except, the value shown by the circles is the square root of the number of aphids per sample (Figures A-16A-27). The square-root scale was used for aphids because the range of values was 0-423 aphids per sample, and using a linear scale would have resulted in circles that covered the entir e map.

PAGE 119

119 Figure A-1. Distribution of crapemyrtle aphids and H. axyridis on 06-August 2007. Circle diameters ar e relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 120

120 Figure A-2. Distribution of crapemyrtle aphids and H. axyridis on 13-August 2007. Circle diameters ar e relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 121

121 Figure A-3. Distribution of crapemyrtle aphids and H. axyridis on 20-August 2007. Circle diameters ar e relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 122

122 Figure A-4. Distribution of crapemyrtle aphids and H. axyridis on 27-August 2007. Circle diameters ar e relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 123

123 Figure A-5. Distribution of crapemyrtle aphids and H. axyridis on 03-September 2007. Circle diam eters are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 124

124 Figure A-6. Distribution of crapemyrtle aphids and H. axyridis on 10-September 2007. Circle diam eters are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 125

125 Figure A-7. Distribution of crapemyrtle aphids and H. axyridis on 17-September 2007. Circle diam eters are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 126

126 Figure A-8. Distribution of crapemyrtle aphids and H. axyridis on 24-September 2007. Circle diam eters are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 127

127 Figure A-9. Distribution of crapemyrtle aphids and H. axyridis on 01-October 2007. Circle diameters are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 128

128 Figure A-10. Distribution of crapemyrtle aphids and H. axyridis on 08-October 2007. Circle diamet ers are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 129

129 Figure A-11. Distribution of crapemyrtle aphids and H. axyridis on 15-October 2007. Circle diamet ers are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 130

130 Figure A-12. Distribution of crapemyrtle aphids and H. axyridis on 22-October 2007. Circle diamet ers are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 131

131 Figure A-13. Distribution of crapemyrtle aphids and H. axyridis on 28-October 2007. Circle diamet ers are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 132

132 Figure A-14. Distribution of crapemyrtle aphids and H. axyridis on 05-November 2007. Circle diam eters are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 133

133 Figure A-15. Distribution of crapemyrtle aphids and H. axyridis on 12-November 2007. Circle diam eters are relative in that 0.3 m equals one insect per sample for aphids per plant (blue), H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). 02550 12.5Meters

PAGE 134

134 Figure A-16. Distribution of crapemyrtle aphids and H. axyridis per plant on 02-July 2008. H. axyridis were trapped in a 24-h period on 01-July 2008. Circle diameters are relative in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yel low). Data for aphids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 135

135 Figure A-17. Distribution of crapemyrtle aphids and H. axyridis per plant on 08-July 2008. H. axyridis were trapped in a 24-h period on 07-July 2008. Circle diameters are relative in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yel low). Data for aphids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 136

136 Figure A-18. Distribution of crapemyrtle aphids and H. axyridis per plant on 15-July 2008. H. axyridis were trapped in a 24-h period on 14-July 2008. Circle diameters are relative in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yel low). Data for aphids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 137

137 Figure A-19. Distribution of crapemyrtle aphids and H. axyridis per plant on 22-July 2008. H. axyridis were trapped in a 24-h period on 21-July 2008. Circle diameters are relative in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yel low). Data for aphids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 138

138 Figure A-20. Distribution of crapemyrtle aphids and H. axyridis on 22-July 2008. Circle diameters are relative in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (ye llow). Data for aphids per plant (blue) are on an exponentia l scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 139

139 Figure A-21. Distribution of crapemyrtle aphids and H. axyridis on 04-August 2008. Circle diameters ar e relative in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (ye llow). Data for aphids per plant (blue) are on an exponentia l scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 140

140 Figure A-22. Distribution of crapemyrtle aphids and H. axyridis on 11-August 2008. Circle diameters ar e relative in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (ye llow). Data for aphids per plant (blue) are on an exponentia l scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 141

141 Figure A-23. Distribution of crapemyrtle aphids and H. axyridis per plant on 19-August 2008. H. axyridis were trapped in a 24-h period on 18-August 2008. Circle diam eters are relative in that 0.3 m eq uals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). Data for aphids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 142

142 Figure A-24. Distribution of crapemyrtle aphids and H. axyridis per plant on 26-August 2008. H. axyridis were trapped in a 24-h period on 25-August 2008. Circle diam eters are relative in that 0.3 m eq uals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). Data for aphids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 143

143 Figure A-25. Distribution of crapemyrtle aphids and H. axyridis per plant on 03-September 2008. H. axyridis were trapped in a 24-h period on 02-September 2008. Circle diameters are relativ e in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). Data for ap hids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 144

144 Figure A-26. Distribution of crapemyrtle aphids and H. axyridis on 10-September 2008. Circle diam eters are relative in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). Data for aphids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of the number of aphids per sample. 02550 12.5Meters

PAGE 145

145 Figure A-27. Distribution of crapemyrtle aphids and H. axyridis per plant on 16-September 2008 and H. axyridis trapped in a 24-h period on 15-September 2008. Circle diameters are relativ e in that 0.3 m equals one insect per sample for H. axyridis per plant (red) and H. axyridis trapped in a 24-h period (yellow). Data for ap hids per plant (blue) are on an exponential scale such that 0.3 m equals the square root of th e number of aphids per sample. 02550 12.5Meters

PAGE 146

146 APPENDIX B ACTIVITY OF ADULT VIRGINOPAROUS CRAPEMYRTLE APHIDS Materials and Methods The activity of adult virginopar ous crapemyrtle aphids was monitored between the dates of 23-July to the 27-July in 2007. Ye llow sticky traps were used to monitor the flight activity of aphids and were constructed from corrugated plas tic board that was cut into 3 x 5 cm pieces. The traps were secured to a cloths pin using a wire and hung on a branch using the cloths pin. Traps were placed on eight Apal achee crapemyrtles that line the front entrance of the North Florida Research and Education Center, Quincy Fl. Four traps were placed on each plant at a height of between 1.5 and 2 m. Traps were chec ked every day at 8 am, 12 pm, 4 pm, and 8 pm. Data were analyzed using PROC GLIMMI X SAS 9.1.3 (SAS Institute 2000-2004) using a one way analysis of variance for time of da y and the Poisson distribution link. Results and Discussion Activity of adult virginoparae crapemyrtle aphi ds was diurnal in nature and peak activity occurred between 4 and 8 pm, which was significantly different than the lowe st period of activity between 8 pm and 8 am (Figure B-1). Because crapemyrtle aphid adults are all winged, it may be important to understand their move ment when applying pesticides.

PAGE 147

147 Time of day 8am12pm4pm8pm LS Mean number of aphids trapped 0.0 0.5 1.0 1.5 2.0 2.5 3.0 a ab ab b Figure B-1. LS mean number of crapemyrtle aphi ds ( SE) caught in all sticky traps used to measure activity patterns.

PAGE 148

148 APPENDIX C PERSONAL OBSERVATIONS Field Observations While conducting experiments, I made numerou s observations on aphids in the field and noted how they infested particular plants or cul tivars. Natchez was rarely attacked by aphids under normal conditions, but aphid populations grew rapidly when aphids were confined to sleeve cages. In addition to the growth of aphi d populations confined in sleeves on Natchez, I noticed that individual plants of Natchez (located in a plot not used in experiments) that were located immediately adjacent to infested Tonto plants had high aphid populations. This effect was the most noticeable if the two plants were touching. Colonization of Natchez appeared to be localized such that there were populations only near the interface of the two plants and the infestation spread as a gradient from the interface The previous type of colonization is different from the total plant infestations seen on other cultivars such as A palachee, Sioux, or Tonto. Aphid infestation levels and popula tion spread in crapemyrtles app eared to be associated with plant spacing, which may explain th e colonization of Natchez in the small plot of crapemyrtle not used in this study. Plant spacing in the sma ll plot was 3 m from trunk to trunk compared to 4 m spacing for the plot used in the spatial experime nt. Crapemyrtle aphids are active fliers during daylight hours (see appendix B). If an aphid co lony was greatly disturbed, the adults took flight within the plant canopy and landed on a nearby l eaf. If plants are closely spaced, disturbed adults are likely to disperse to nearby plants as readily as th e plant they are currently occupying, and thus, this would increase rate at which other plants are infested. Laboratory Observations of Crapemyrtle Suitability Before conducting the host suitability expe riment, several cultivars and species of crapemyrtle were evaluated for the ability to support development and reproduction of

PAGE 149

149 crapemyrtle aphids. Tests were carried out on 40mm leaf disks that were prepared as outlined in the procedures of Chapter 2. Four leaf disks pe r cultivar or species were prepared and all four disks were placed in a single 100 x 10 mm agar Petri-plate that was prepared according to the procedures outlined in Chapter 3. The cultivars or species tested were as follows: Fantasy, Townhouse, Woodlanders Chocolate Soldier, L. limii L. speciosa L. chekiangensis and Princess ( L. indica L. speciosa ). Several adult virginoparae aphids were placed in each Petri-plate and observed at 24 h and 7 d, and observations noted if the aphids were on any of the leaf disks and if there were any offspring present. All aphids were alive, when the plates were checked at 24 h, but only two aphids were on a disk in the L. speciosa plate. At the end of seve n days, leaf disks from all cultivars and species, except, L. limii and L. speciosa had adult virginoparae and nymphs present. An interesting outcome of these observations is that while L. speciosa is not suitable for crapemyrtle aphid development and reproduction, the hybridized L. speciosa L. indica cultivar Princess was suitable for crapemyrtle aphid development and reproduction.

PAGE 150

150 LIST OF REFERENCES Alverson, D. R., and R. K. Allen. 1991. Life history of the crapem yrtle aphid. Proc. SNA Res. Conf. 36: 164-167. Alverson, D. R., and R. K. Allen. 1992a. Bionomics of the crapemyrtle aphid (Homoptera: Aphididae). J. Entomol. Sci. 27: 445-457. Alverson, D. R., and R. K. Allen. 1992b. Suitability of 'Natchez' vs. 'Carolina Beauty' crapemyrtle cultivars as hosts for the crap emyrtle aphid. Proc. S NA Res. Conf. 37: 160162. Bianchi, F. J. A., and W. v. d. Werf. 2004. Model evaluation of the function of prey in noncrop habitats for biological control by la dybeetles in agricultural landscapes. Ecol. Model. 171: 177-193. Blackman, R. L. 1967. The effects of different aphid foods on Adalia bipunctata L. and Coccinella 7-punctata L. Ann. Appl. Biol. 59: 207-219. Bommarco, R., S. O. Firle, and B. Ekbom. 2007. Outbreak suppression by predators depends on spatial distribution of pr ey. Ecol. Model. 201: 163-170. Bottrell, D. G., P. Barbosa, and F. Gould. 1998. Manipulating natura l enemies by plant variety selection and modification: a real istic strategy. Annu. Rev. Entomol. 43: 347-367. Brewer, M. J. and N. C. Elliott. 2004. Biological control of cereal aphids in North America and mediating effects of host plant and ha bitat manipulations. Annu. Rev. Entomol. 49:219-242. Clark, T. L., and F. J. Messina. 1998. Foraging behavior of lacew ing larvae (Neuroptera: Chrysopidae) on plants with divergent architectures. J. In sect Behav. 11: 303-317. Costamagna, A. C., D. A. Landis, and C. D. Difonzo. 2007. Suppression of soybean aphid by generalist predators re sults in a trophic cascade in soybeans. Ecol. Appl. 17: 441-451. Dinkins, R. L., W. L. Tedders, and W. Reid. 1994. Predaceous neuropterans in Georgia and Kansas pecan trees. J. Entomol. Sci. 29: 165-175. Dix, R. L. 1999. Cultivars and names of Lagerstroemia http://www.usna.usda.gov/Resear ch/Herbarium/Lagerstroemia/ U. S. National Arboretum. Updated May 25, 2005. Accessed January 28, 2009. Dixon, A. F. G. 1971. The life-cycle and host preferences of the bird cherry-oat aphid, Rhopalosiphum padi L., and their bearing on the theories of host alternation in aphids. Ann. Appl. Biol. 68: 135-147. Dixon, A. F. G. 1973. Biology of aphids. Edwa rd Arnold Publishers, London.

PAGE 151

151 Dixon, A. F. G. 1998. Aphid ecology: an optimization approach. Chapman & Hall, London, New York. Douglas, A. E. 1993. The nutritional quality of phloem sap utilized by natural aphid populations. Ecol. Entomol. 18: 31-38. Dozier, H. L. 1926. Crepe myrtle plant louse. J. Econ. Entomol. 19: 800-800. Duffey, J. 1980. Sequestration of plant natural products by insects. Annu. Rev. Entomol. 25: 447-477. Dutcher, J. D. 1985. Impact of late season aphid control on pecan tree vigor parameters. J. Entomol. Sci. 20: 55-61. Dutcher, J. D. and T. Htay. 1985. Resurgence and insecticide resistance problems in pecan aphid management. Univ. Ga. Ag. Exp. Stat. Spec. Pub. 38. Egolf, D. R. 1981a. 'Muskogee' and 'Natchez' Lagerstroemia HortScience 16: 576-577. Egolf, D. R. 1981b. 'Tuscarora' Lagerstroemia HortScience 16: 788-789. Egolf, D. R. 1986a. 'Tuskegee' Lagerstroemia HortScience 21: 1078-1080. Egolf, D. R. 1986b. 'Acoma', 'Hopi', 'Pecos', and 'Zuni' Lagerstroemia HortScience 21: 12501252. Egolf, D. R. 1987a. 'Biloxi', 'Miami' and 'Wichita' Lagerstroemia HortScience 22: 336-338. Egolf, D. R. 1987b. 'Apalachee', 'Comanche', 'Lipan', 'Osage', 'Sioux' and 'Yuma' Lagerstroemia. HortScience 22: 674-677. Egolf, D. R. 1990a. Lagerstroemia cultivar 'Choctaw'. HortScience 25: 992-993. Egolf, D. R. 1990b. 'Caddo' and 'Tonto' Lagerstroemia HortScience 25: 585-587. Eisner, T., M. Eisner, C. Rossini, V. K. Iyengar, B. L. Roach, E. Benedikt, and J. Meinwald. 2000. Chemical defense against pr edation in an insect egg. Proc. Natl. Acad. Sci. U.S.A. 97: 1634-1639. Evans, E. W. 2003. Searching and reproductive behaviou r of female aphidophagous ladybirds (Coleoptera: Coccinellidae): a revi ew. Eur. J. Entomol. 100: 1-10. Ferris, J. P., R. C. Briner, and C. B. Boyce. 1971a. Lythraceae alkaloids. VIII. the structure and stereochemistry of the bi phenyl ether alkaloids from Decodon verticillatus J. Am. Chem. Soc. 93: 2953-2957. Ferris, J. P., R. C. Briner, and C. B. Boyce. 1971b. Lythraceae alkaloids. IX the isolation and structure elucidation of the alkaloids of Lagerstroemia indica L. J. Am. Chem. Soc. 93: 2858-2962.

PAGE 152

152 Ferris, J. P., C. B. Boyce, R. C. Briner, U. Weiss, I. H. Qureshi, and N. E. Sharpless. 1971c. Lythraceae alkaloids. X. assignment of ab solute stereochemistries on the basis of chiraloptical effects. J. Am. Chem. Soc. 93: 2963-2968. Fragoyiannis, D. A., R. G. McKin lay, and J. P. F. D'Mello. 1998. Studies of the growth, development and reproductive performance of the aphid Myzus persicae on artificial diets containing potato glycoalkaloids Entomol. Exp. Appl. 88: 59-66. Francis, F., E. Haubruge, P. Hastir, and C. Gaspar. 2001. Effect of aphid host plant on development and reproduction of the third trophic level, the predator Adalia bipunctata (Coleoptera: Coccinellidae). Environ. Entomol. 30: 947-952. Fuentes-Contreras, E., and H. M. Niemeyer. 1998. Dimboa glucoside, a wheat chemical defense, affects host accept ance and suitability of Sitobion avenae to the cereal aphid parasitoid Aphidius rhopalosiphi J. Chem. Ecol. 24: 371-381. Giles, K. L., R. D. Madden, M. E. Payton, and J. W. Dillwith. 2000. Survival and development of Chrysoperla rufilabris (Neuroptera: Chrysopidae) supplied with pea aphids (Homoptera: Aphididae) reared on alfalfa and faba bean. Environ. Entomol. 29: 304-311. Giles, K. L., R. Stockland, and R. D. Madden. 2001. Preimaginal survival and development of Coleomegilla maculata and Hippodamia convergens (Coleoptera: Coccinellidae) reared on Acyrthosiphon pisum : Effects of host plants. Environ. Entomol. 30: 964-971 Giles, K. L., R. C. Berberet, A. A. Zarrabi, and J. W. Dillwith. 2002a. Influence of alfalfa cultivar on suitability of Acyrthosiphon kondoi (Homoptera: Aphidid ae) for survival and development of Hippodamia convergens and Coccinella septempunctata (Coleoptera: Coccinellidae). J. Econ. Entomol. 95: 552-557. Giles, K. L., R. D. Madden, R. Stockland, M. E. Payton, and J. W. Dillwith. 2002b. Host plants affect predator fitness via the nutritional value of herb ivore prey: inve stigation of a plant-aphid-ladybeetle system. BioControl (Dordrecht) 47: 1-21. Gonzales, W. L., E. Fuentes-Contrera s, and H. M. Niemeyer. 2002. Host plant and natural enemy impact on cereal aphid competition in a seasonal environment. Oikos 96: 481-491. Hauge, M. S., F. H. Nielsen, and S. Toft. 1998. The influence of th ree cereal aphid species and mixed diet on larval survival, development and adult weight of Coccinella 7punctata Entomol. Exp. Appl. 89:319-322. Hesler, L. S., and R. W. Kieckhefer. 2008. Status of exotic and previously common native coccinellids (Coleoptera) in South Dakota land scapes. J. Kans. Entomol. Soc. 81: 29-49. Hill, C. B., Y. Li, and G. L. Hartman. 2004. Resistance to the soyb ean aphid in soybean germplasm. Crop Sci. 44: 98-106.

PAGE 153

153 Hodek, I. 1956. The influence of Aphis sambuci L. as prey of the ladybird beetle Coccinella septempunctata L. Acta Soc. Zool. Bohemoslov. 20: 72-74. Hodek, I. 1993. Habitat and food specificity in aphi dophagous predators. Biocontrol Sci. Technol. 3: 91-100. Hydorn, S. B., and W. H. Whitcomb. 1979. Effects of larval diet on Chrysopa rufilabris. Fla. Entomol. 62: 293-298. Kareiva, P., and G. Odell. 1987. Swarms of predators exhibit "preytaxis" if individual predators use area-restricted search. Am. Nat. 130: 233-270. Kareiva, P., and R. Sahakian. 1990. Tritrophic effects of a simple architectural mutation in pea plants. Nature. 345: 433. Kindlmann, P., and F. G. Dixon. 1993. Optimal foraging in ladybird beetles (Coleoptera: Coccinellidae) and its consequences for their us e in biological control. Eur. J. Entomol. 90: 443-450. Kindlmann, P., and A. F. G. Dixon. 1999a. Generation time ratios determinants of prey abundance in insect predator-prey in teractions. Biol. Control 16: 133-138. Kindlmann, P., and A. F. G. Dixon. 1999b. Strategies of aphidophagous predators: lessons for modeling insect predato r-prey dynamics. J. Appl. Entomol. 123: 397-399. Kogan, M., and E. F. Ortman. 1978. Antixenosis-a new term proposed to define Painter's "nonpreference" modality of resistance. Bull. Entomol. Soc. Am. 24: 175-176. Korie, S., J. N. Perry, M. A. Mugglestone, S. J. Clark, C. F. G. Thomas, and M. N. Mohamad Roff. 2000. Spatiotemporal associations in beetle and virus count data. J. Agric. Biol. Envi ron. Stat. 5 214-239. Legaspi, J. C., R. I. Carruthers, and D. A. Nordlund. 1994. Life history of Chrysoperla rufilabris (Neuroptera: Chrysopidae) provided sweet potato whitefly Bemisia tabaci (Homoptera: Aleyrodidae) and ot her food. Biol. Control 4: 178-184. Legaspi, J. C., D. A. Nordlund, and B. C. Legaspi, Jr. 1996. Tri-trophic interactions and predation rates in Chrysoperla spp. attacking the silverleaf whitefly. Southwest. Entomol. 21: 33-42. Legrand, A., and P. Barbosa. 2003. Plant morphological complexity impacts foraging efficiency of adult Coccinella septempunctata L. (Coleoptera: Coccinellidae). Environ. Entomol. 32: 1219-1226. Liao, H. T., M. K. Harris, F. E. Gilstrap, and F. Mansour. 1985. Impact of natural enemies on the blackmargined pecan aphid, Monellia caryella (Homoptera: Aphidae). Environ. Entomol. 14: 122-126.

PAGE 154

154 Liao, H. T., M. K. Harris, F. E. Gilstrap, D. A. Dean, C. W. Agnew, G. J. Michels, and F. Mansour. 1984. Natural enemies and other factors affecting seasonal abundance of the blackmargined aphid on pecan. S outhwest. Entomol. 9: 404-420. Malcolm, S. B. 1990. Chemical defence in chewing and sucking insect herbivores: plantderived cardenolides in the mona rch butterfly an d oleander aphid. Chemoecology 1: 1221. Mensah, R. K. 1997. Yellow traps can be used to monitor populations of Coccinella transversalis (F.) and Adalia bipunctata (L.) (Coleoptera: Coccinellidae) in cotton crops. Aust. J. Entomol. 36: 377-381. Messina, F. J., and S. M. Sorenson. 2001. Effectiveness of lacewing larvae in reducing Russian wheat aphid populations on susceptibl e and resistant wheat. Biol. Control 21: 1926. Mizell, R. F., III. 2003. Summary and overview of pecan arthropod pest management. Southwest. Entomol. 27: 135-140. Mizell, R. F., III. 2007. Impact of Harmonia axyridis (Coleoptera: Coccinellidae) on native arthropod predators in pecan and crap e myrtle. Fla. Entomol. 90: 524-536. Mizell, R. F., III, and D. E. Schiffhauer. 1987a. Trunk traps and overwintering predators in pecan orchards survey of species and emergence times. Fla. Entomol. 70: 238-244. Mizell, R. F., III, and D. E. Schiffhauer. 1987b. Seasonal abundance of the crapemyrtle aphid, Sarucallis kahawaluokalani in relation to the pecan aphids, Monellia caryella and Monelliopsis pecanis and their common predators. Entomophaga 32: 511-520. Mizell, R. F., III, and D. E. Schiffhauer. 1989. Effects of pesticides on pecan aphid predators Chrysoperla rufilabris (Neuroptera: Chrysopidae), Hippodamia convergens Cycloneda sanguinea (L.), Olla v-nigrum (Coleoptera: Coccinellidae), and Aphelinus perpallidus (Hymenoptera: Encyrtidae). J. Econ. Entomol. 83: 1806-1812. Mizell, R. F., III, and M. C. Sconyers. 1992. Toxicity of imidacloprid to selected arthropod predators in the laborator y. Fla. Entomol. 75: 277-280. Mizell, R. F., III, and G. W. Knox. 1993. Susceptibility of crapemyrtle, Lagerstroemia indica L., to the crapemyrtle aphid (Homoptera: Aphidi dae) in North Florida. J. Entomol. Sci. 28: 1-7. Mizell, R. F., III, F. D. Bennett, and D. K. Reed. 2002. Unsuccessful search for parasites of the crapemyrtle aphid, Tinocallis kahawaluokalani (Homoptera: Aphididae). Fla. Entomol. 85: 521-523. Musser, F. R., J. P. Nyrop, and A. M. Shelton. 2004. Survey of predators and sampling method comparison in sweet cor n. J. Econ. Entomol. 97: 136-144.

PAGE 155

155 Nault, B. A., and G. G. Kennedy. 2003. Establishment of multicolored Asian lady beetle in eastern North Carolina: Seasonal abundance an d crop exploitation within an agricultural landscape. Biol. C ontrol 48: 363-378. New, T. R. 1975. The biology of Chrysopidae and Hemerobi idae (Neuroptera), with reference to their usage as biological control agents: a review. Trans. R. Ent. Soc. Lond. 127: 115140. New, T. R. 1984. Biology of Chrysopidae, pp. x + 294 pp. In K. A. Spencer [ed.], Biology of Chrysopidae. DR W. Junk Publishers, Boston. Obrycki, J. J., and T. J. Kring. 1998. Predaceous Coccinellidae in biological control. Annu. Rev. Entomol. 43: 295-321. Okamoto, H. 1966. Three problems of prey specificit y of aphidophagous coccinellids, In I. Hodek [ed.], Ecology of aphidophagous insects. Academia, Prague. Painter, R. H. 1968. Insect resistance in crop plants. Macmillan, New York. Panda, N., and G. A. Khush. 1995. Host plant resistance to insects. Wallingford, Oxon, UK. Parajulee, M. N., and J. E. Slosser. 2003. Potential of yellow stic ky traps for lady beetle survey in cotton. J. Econ. Entomol. 96: 239-245. Perry, J. N. 1995. Spatial analysis by distance i ndices. J. Anim. Ecol. 64: 303-314. Perry, J. N. 1998. Measures of spatial pattern for counts. Ecology 79: 1008-1017. Perry, J. N., and P. M. Dixon. 2002. A new method to measure spatial association for ecological count data. Ecoscience 9: 133-141. Perry, J. N., E. D. Bell, R. H. Smith, and I. P. Woiwod. 1996. SADIE: Software to measure and model spatial pattern. Aspect. Appl. Biol. 46: 95-102. Perry, J. N., L. Winder, J. M. Holland, and R. D. Alston. 1999. Red-Blue plots for detecting clusters in count data Ecol. Lett. 2: 106-113. Pettis, G. V., D. W. Boyd, Jr., S. K. Braman, and C. Pounders. 2004. Potential resistance of crape myrtle cultivars to fl ea beetle (Coleoptera: Chrysomelidae) and Japanese beetle (Coleoptera: Scarabaeidae) damage. J. Econ. Entomol. 97: 981-992. Pooler, M. R. 2003. Molecular genetic diversit y among twelve clones of Lagerstroemia fauriei revealed by AFLP and RAPD ma rkers. HortScience 38: 256-259. Pooler, M. R., and R. L. Dix. 1999. 'Chickasaw', 'Kiowa', and 'Pocomoke' Lagerstroemia HortScience 34: 361-363. Powell, G., C. R. Tosh, and J. Hardie. 2006. Host plant selection by aphids: behavioral, evolutionary, and applied perspec tives. Annu. Rev. Entomol. 51: 309-330.

PAGE 156

156 Price, P. W., C. E. Bouton, P. Gross, B. A. McPheron, J. N. Thompson, and A. E. Weis. 1980. Interactions among three trophic levels: influence of plants on interactions between insect herbivores and natural enemies. Annu. Rev. Ecol. Syst. 11: 41-65. Quednau, W. 2003. Atlas of the drepanosiphine aphids of the world, II. Mem. American Ent. Inst. 72: 60. Randolph, T. L., M. K. Kroening, J. B. Rudolph, F. B. Peairs, and R. F. Jepson. 2002. Augmentative releases of commercial biolog ical control agents for Russian wheat aphid management in winter wheat. Southwest. Entomol. 27: 37-44. Reilly, C. C., and W. L. Tedders. 1990. A detached-leaf method to study pecan aphid behavior and biology. J. En tomol. Sci. 25: 85-88. Rothschild, M., J. Von Euw, and T. Reichstein. 1970. Cardiac glycosides in the oleander aphid Aphis nerii J. Insect Phys iol. 16: 1141-1145. Ruberson, J. R., M. J. Tauber, and C. A. Tauber. 1986. Plant feeding by Podisus maculiventris (Heteroptera: Pentatomidae): eff ect on survival, development, and preoviposition period. Environ. Entomol. 15: 894-897. Rutledge, C. E., R. J. O'Neil, T. B. Fox, and D. A. Landis. 2004. Soybean aphid predators and their use in integrated pest mana gement. Ann. Entomol. Soc. Am. 97: 240-248. SAS. Institute Inc ., SAS 9.1.3 Help and Documentation, Cary, NC: SAS Institute Inc., 20002004. Schmidt, N. P., M. E. O'Neal, and P. M. Dixon. 2008. Aphidophagous predators in Iowa soybean: a community comparison across mu ltiple years and sampling methods. Ann. Entomol. Soc. Am. 101: 341-350. Singer, M. C. 2000. Reducing ambiguity in describing plantinsect interactions: "preference", "acceptability" and "electivity". Ecol. Lett. 3: 159-162. Skirvin, D. J., J. N. Perry, and R. Harrington. 1997. The effect of climate change on an aphid-coccinellid interaction. Global Change Biol. 3: 1-11. Stephens, E. J., and J. E. Losey. 2004. Comparison of sticky cards, visual and sweep sampling of coccinellid populations in alfa lfa. Environ. Entomol. 33: 535-539. Stewart, C. D., S. K. Braman, and A. F. Pendley. 2002. Functional response of the azalea plant bug (Heteroptera: Miri dae) and a green lacewing Chrysoperla rufilabris (Neuroptera: Chrysopidae), two predators of the azalea lace bug (Heteroptera: Tingidae). Environ. Entomol. 31: 1184-1190. Stoner, A. 1970. Plant feeding by a predaceous insect, Geocoris punctipes J. Econ. Ent. 63: 1911-1915.

PAGE 157

157 Stoner, A., A. M. Metcalfe, and R. E. Weeks. 1974. Plant feeding by a predaceous insect, Podisus acutissimus Environ. Entomol. 3: 187-189. Systat Software Inc. SigmaPlot version 11 2008. Tedders, W. L. 1983. Insect management in deciduous orchard ecosystems: habitat manipulation. Environ. Manage. 7: 29-34. Tedders, W. L., B. W. Wood, and J. W. Snow. 1982. Effects of feeding by Monelliopsis nigropunctata, Monellia caryella, and Monelliopsis caryaefoliae on growth of pecan seedlings in the green house. J. Econ. Entomol. 75:287-291. Tedders, W. L., and P. W. Schaefer. 1994. Release and establishment of Harmonia axyridis (Coleoptera: Coccinellidae) in the southeastern United States. Entomol. News 105: 228243. Tosh, C. R., G. Powell, N. D. Holmes, and J. Hardie. 2003. Reproductive response of generalist and specialis t aphid morphs with the same genotype to plant secondary compounds and amino acids. J. Insect Physiol. 49: 1173-1182. Van Emden, H. F. 1995. Host plant-Aphidophaga interactions. Agric. Ecosys. Environ 52: 311. Vidal, S., and T. Tscharntke. 2001. Multitrophic plant-insect inte ractions. Basic Appl. Ecol. 1: 1-2. Wang, R., and D. A. Nordlund. 1994. Use of Chrysoperla spp. (Neuroptera: Chrysopidae) in augmentative release programs for control of arthropod pests. Biocontrol News and Information. 15: 51N-57N. Weibull, J. H. W. 1988. Free amino acids in the phloem sap from oats and barley resistant to Rhopalosiphum padi Phytochemistry. 27: 2069-2072. Wilkinson, T. L., and A. E. Douglas. 2003. Phloem amino acids and the host plant range of the polyphagous aphid, Aphis fabae Entomol. Exp. Appl. 106: 103-113. Wood, B. W., W. L. Tedders, and J. M. Thompson. 1982. Effects of an infestation of blackmargined aphid on carbohydrates in matu re 'Stuart' pecan. HortScience 17:236-238.

PAGE 158

158 BIOGRAPHICAL SKETCH John Herbert was born in Kettering, Ohio to his parents David and Marg aret Herbert. At the age of 13, John joined the Civil Air Patrol and learned i nvaluable leadership skills and experienced the thrill of flying small airplanes. John graduated from Centerville High School in 1993 and attended Wright State University for a fe w years before transferring to The Ohio State University. While working on his un dergraduate degree, John worked at an auto repair shop and worked on vintage race cars. Johns summers were spent traveling with the race team, and his adventures included trips to famous race cour ses like Watkins Glen in New York and Laguna Seca in California. John received his BS in biol ogy from The Ohio State University in the fall of 1999 and returned a year later to pursue his masters degree in entomology. His thesis investigated the interac tions that occur among aphids, ants, and aphid natural enemies. During his time as a graduate student at OSU, he partic ipated in a program to reintroduce the federally endangered American burying beetle, Nicrophorus americanus After completing MS in entomology in 2004, John spent six months in Ha waii, where he lived on the beautiful garden island of Kauai and the crowded city of Honol ulu on Oahu. In January of 2005, John packed his car in Centerville, OH and drove to Gainesville, FL to pursu e his Doctor of Philosophy in entomology. John conducted research in the Flor ida Panhandle in a small town called Quincy and lived on the research station for five months every summ er. John finished his last field season in 2008 and received his PhD in entomology on May 2, 2009.