Electrical Penetration Graph Investigations of Asian Citrus Psyllid (Diaphorina citri Kuwayama) Feeding Behavior

MISSING IMAGE

Material Information

Title:
Electrical Penetration Graph Investigations of Asian Citrus Psyllid (Diaphorina citri Kuwayama) Feeding Behavior Effects of Insecticides on the Potential Transmission of Candidatus Liberibacter asiaticus
Physical Description:
1 online resource (168 p.)
Language:
english
Creator:
Serikawa,Rosana Harumi
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Entomology and Nematology
Committee Chair:
Rogers, Michael E
Committee Members:
Webb, Susan E
McAuslane, Heather J
Stelinski, Lukasz L.
Brlansky, Ronald H

Subjects

Subjects / Keywords:
asian -- electrical -- feeding -- huanglongbing -- insecticide
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:
An Electrical Penetration Graph (EPG) monitor was used to study the feeding behaviors of Diaphorina citri Kuwayama, vector of Candidatus Liberibacter asiaticus, the presumed causal Huanglongbing (HLB). Effects of insect gender, presence of light, and ability and duration of insecticides to disrupt D. citri feeding behaviors responsible for pathogen transmission were examined. Results showed that duration of phloem ingestion was significantly longer for female D. citri compared to males. Gender based analysis showed that despite previously being considered a phloem feeder, D. citri performs similar amounts of xylem and phloem feeding. Additionally, xylem feeding was more likely to occur during the day and phloem feeding during the night. Application of the soil-applied neonicotinoid insecticide imidacloprid disrupted D. citri phloem feeding behaviors suggesting that application of this product to young trees can reduce the likelihood that trees will succumb to HLB. In contrast, the soil-applied carbamate insecticide aldicarb did not disrupt D. citri phloem feeding behaviors. In fact, phloem ingestion by D. citri was increased on aldicarb-treated plants indicating that use of this product could potentially increase pathogen acquisition rates where aldicarb-treated diseased trees are present. The foliar-applied insecticides chlorpyrifos, fenpropathrin, and imidacloprid disrupted D. citri phloem-feeding behaviors on recently treated plants suggesting that these products will prevent pathogen transmission prior to death of the insect. While application of spinetoram altered D. citri feeding behavior, phloem feeding was not disrupted suggesting that pathogen transmission may still occur prior to insect death. In contrast, spirotetramat did not significantly disrupt D. citri feeding and thus will not prevent pathogen transmission from occurring. In experiments examining duration of feeding disruption provided by foliar insecticide applications, imidacloprid provided the longest duration of feeding disruption (28 d), followed by fenpropathrin (21 d), and chlorpyrifos (1 d). These results demonstrate that in addition to reducing vector populations in the field, certain soil- and foliar-applied insecticides can prevent pathogen transmission from occurring prior to death of pathogen-carrying D. citri. Our findings are of immediate importance for Florida citrus growers who can use this information to improve their current HLB management programs.
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 Rosana Harumi Serikawa.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Rogers, Michael E.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043259:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 ELECTRICAL PENETRATION GRAPH INVESTIGATIONS OF ASIAN CITRUS PSYL LID ( DIAPHORINA CITRI KUWAYAMA) FEEDING BEHAVIOR: EFFECTS OF INSECTICIDES ON THE POTENTIAL TRANSMISSION OF CANDIDATUS LIBERIBACTER ASIATICUS By ROSANA HARUMI SERIKAWA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 R osana H arumi S erikawa

PAGE 3

3 To my parents and my siblings

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank the University of Florida, the Entomology and Nematology Department, the Citrus Research and Educational Center, Lynn Steinmetz Scholarship and Fellowship program, and the Citrus Research and Devel opment Foundation for giving me the opportunity and the resources to finish this study. I would also like to thank my major advisor Dr. Michael E. Rogers; the opportunity, words of encouragement, advice and guidance, his support and enthusiasm made my PhD research a lot easier to be accomplished. I thank all my supervisory committee members, Drs. Lukasz L. Stelinski (co advisor), Ronald H. Brlansky, Heather J. McAuslane, and Susan E. Webb. Without the help of Drs. Stelinski and Brlansky and all the support necessary on this research, nothing would have being accomplished. I thank Drs. McAuslane and Webb for their support and introducing to me to the EPG technique, which was the whole topic of this dissertation. I thank Dr. Elaine S. Backus, for all her enthu siasm and research advice; her expertise in the EPG technique helped me to improve my research techniques, my statistical analysis, and research conclusions. Without her I would not have finished my dissertation. I also thank Dr. Timothy A. Ebert, for his friendship and SAS expertise; Dr. Antonios E. Tsagarakis, for his friendship and unlimited psyllid supply; Daniela M. Okuma for being a friend and great assistant in helping me with the experiments and data analysis; Rhonda Schumann, Perciva Mariner, Dalia M. Shawer, Harry E. Anderson, and Guoping Liu for the friendship and great help in this pr oject; Ian Jackson, Wendy Meyer and Angel Hoyte for helping me with the Faraday cages. Additionally, Tania F. Broisler, Jos Francisco Figueiredo, Jared Ali, Sara He rmann, Sylvia Morais de Souza, Tatiana D. Martinelli, Marina B. Arouca, Luciano Rocha, Fabiano Santinello, Nathalia Vieira, Luiz Henrique Falavigna, Pamela Fvero, Andr Giorgetti, Bruno Zeuli, and many other friends that passed through my life made my lif e a lot easier in the USA with their

PAGE 5

5 love and friendship. Finally I would like to thank the most special people in my life: Gabriel Fachini, became my best friend and with all his love, friendship and patience, helped me format and edit my dissertation; my siblings, Marcel A. Serikawa and Simoni S. Serikawa, for their existence, for without them, my life would not have being so fun; and lastly my parents, Tomoko Nakaema Serikawa and Mitsuo Serikawa, with their love, support, and encouragement which made eve rything possible.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 9 LIST OF FIGURES ................................ ................................ ................................ ....................... 12 ABSTRACT ................................ ................................ ................................ ................................ ... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 15 2 GENDER DIFFERENCE AND EFFECT OF LIGHT A ND DARK ON ASIAN CITRUS PSYLLID ( DIAPHORINA CITRI KUWAYAMA) FEEDING BEHAVIOR ......... 24 Materials and Methods ................................ ................................ ................................ ........... 26 Plants and Insects ................................ ................................ ................................ ............ 26 EPG Equipment ................................ ................................ ................................ ............... 26 EPG Recording ................................ ................................ ................................ ................ 27 Asian Citrus Psyllid Waveform Terminology ................................ ................................ 28 Experimental Design ................................ ................................ ................................ ....... 30 Experiment 1. Gender based differences in D. citri feeding behavior. .................... 30 Experiment 2. Effects of light and dark on D. citri feeding behavior. ..................... 30 Statistical Analysis ................................ ................................ ................................ .......... 31 Results ................................ ................................ ................................ ................................ ..... 32 New Waveform Characterization and Correlations ................................ ........................ 32 Experiment 1. Gender based Differences in D. citri Feeding Behavior. ........................ 32 Feeding behavior of female psyllids (within treatment analysis). ............................ 35 Feeding behavior of male psyllids (within treatment analysis). ............................... 36 Experiment 2. Effects of Light and Dark on D. citri Feeding Behavior. ........................ 36 D. citri feeding behavior in light (within treatment analysis). ................................ 39 D. citri feeding behavior in dark (within treatment analysis). ................................ 40 Discussion ................................ ................................ ................................ ............................... 40 3 EF FECTS OF SOIL APPLIED IMIDACLOPRID ON ASIAN CITRUS PSYLLID ( Diaphorina citri Kuwayama) (HEMIPTERA: PSYLLIDAE) FEEDING BEHAVIOR ...... 54 Material and Methods ................................ ................................ ................................ ............. 56 Plants and Insects ................................ ................................ ................................ ............ 56 EPG Recording and Waveform Analysis ................................ ................................ ........ 56 Experimental Design ................................ ................................ ................................ ....... 57 Experiment 1. D. citri feeding behavior on young leaves. ................................ ....... 57 Experiment 2. D. citri feeding behavior on mature leaves. ................................ ...... 57 Statistical Analysis ................................ ................................ ................................ .......... 57

PAGE 7

7 Results ................................ ................................ ................................ ................................ ..... 58 Experiment 1. Treatment Effects on Feeding, Young Leaves. ................................ ........ 58 Experiment 2. Treatment Effects on Feeding, Mature Leaves. ................................ ....... 62 Discussion ................................ ................................ ................................ ............................... 65 4 EFFECTS OF ALDI CARB ON ASIAN CITRUS PSYLLID (HEMIPTERA: PSYLLIDAE) FEEDING BEHAVIOR AND THEIR POTENTIAL IMPACTS ON TRANSMISSION ................................ ................................ ................................ ................... 78 Materials and Methods ................................ ................................ ................................ ........... 79 Plants and Insects ................................ ................................ ................................ ............ 79 EPG Recording and Waveform Analysis ................................ ................................ ........ 79 Effects of Aldicarb on D. citri Feeding Behavior ................................ ........................... 79 Statistical Analysis ................................ ................................ ................................ .......... 80 Confirmation of Aldicarb in Treated Plants ................................ ................................ .... 80 Results ................................ ................................ ................................ ................................ ..... 81 Confirmation of Aldicarb in Treated Plants ................................ ................................ .... 83 Discussion ................................ ................................ ................................ ............................... 84 5 EFFECTS OF FIVE D IFFERENT FOLIAR APPLIED INSECTICIDES ON ASIAN CITRUS PSYLLID ( DIAPHORINA CITRI ) FEEDING BEHAVIOR AND THEIR POSSIBLE IMPLICATION FOR LAS TRANSMISSION ................................ ................... 96 Materials and Methods ................................ ................................ ................................ ........... 98 Plants and Insects ................................ ................................ ................................ ............ 98 EPG Recording and Waveform Analysis ................................ ................................ ........ 98 Effects of Foliar Insecticides on D. citri Feeding Behavior ................................ ............ 98 Statistical Analysis ................................ ................................ ................................ .......... 99 Results ................................ ................................ ................................ ................................ ..... 99 Chlor pyrifos ................................ ................................ ................................ ..................... 99 Fenpropathrin ................................ ................................ ................................ ................ 101 Imidacloprid ................................ ................................ ................................ .................. 102 Spinetoram ................................ ................................ ................................ ..................... 104 Spirotetramat ................................ ................................ ................................ ................. 105 Discussion ................................ ................................ ................................ ............................. 107 6 RESIDUAL ACTIVITY OF FIVE DIFFERENT FOLIAR APPLIED INSECTICIDES ON ASIAN CITRUS PSYLLID ( DIAPHORINA CITRI ) FEEDING BEHAVIOR AND THEIR POSSIBLE IMPLICATIONS FOR LAS TRANSMISSION ................................ .. 119 Materials and Methods ................................ ................................ ................................ ......... 120 Plants and Insects ................................ ................................ ................................ .......... 120 EPG Recording and Waveform Analysis ................................ ................................ ...... 120 Effects of Residual Foliar Insecticides on Psyll id Feeding Behavior ........................... 120 Residual Analysis ................................ ................................ ................................ .......... 121 Statistical Analysis ................................ ................................ ................................ ........ 121 Re sults ................................ ................................ ................................ ................................ ... 122

PAGE 8

8 Experiment 1. One Day After Treatment ................................ ................................ ...... 122 Experiment 2. Seven Days After Treatment ................................ ................................ .. 124 Experiment 3. Fourteen Days After Treatment ................................ ............................. 125 Experiment 4. Twenty one Days After Treatment ................................ ........................ 128 Experiment 5 Twenty eight Days After Treatment ................................ ...................... 130 Residual Analyses and Temperature Recording ................................ ............................ 134 Discussion ................................ ................................ ................................ ............................. 134 7 SUMMARY ................................ ................................ ................................ .......................... 149 LIST OF REFERENCES ................................ ................................ ................................ ............. 158 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 168

PAGE 9

9 LIST OF TABLES Table page 2 1 Mean ( SE) WDI (s) and PPW for different genders of Diaphorina citri feeding and under different light conditions. ................................ ................................ .................. 44 2 2 Mean ( SE) PDI (s), NPI, and PDE (s) for different genders of Diaphorina citri feeding and under different light conditions. ................................ ................................ ..... 45 2 3 Mean ( S E) NWEP and NPw for different genders of Diaphorina citri feeding and under different light conditions. ................................ ................................ ......................... 4 6 2 4 Mean ( SE) WDP (s) for different genders of Diaphorina citri feeding and under different light conditions. ................................ ................................ ................................ ... 47 2 5 Mean ( SE) NWEI for different genders of Diaphorina citri feeding and under different light conditions. ................................ ................................ ................................ ... 48 2 6 Mean ( SE) WDEI (s) for different genders of Diaphorina citri feeding and under different light conditions. ................................ ................................ ................................ ... 49 3 1 Mean ( SE) WDI (s) and PP W for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants. ................................ .................. 69 3 2 Mean ( SE) PDI (s), NPI, and PDE (s) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants. ................................ ...... 70 3 3 Mean ( SE) WDP (s) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants. ................................ ................................ .... 71 3 4 Mean ( SE) NWEP and NPw for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants. ................................ .................. 72 3 5 Mean ( SE) NWEI for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants. ................................ ................................ .... 73 3 6 Mean ( SE) WDEI (s) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants. ................................ ......................... 74 4 1 Mean ( SE) WDI (s) and PPW for Dia phorina citri feeding on aldicarb treated and untreated citrus plants. ................................ ................................ ................................ ....... 88 4 2 Mean ( SE) PDI (s), NPI and PDE (s) for Diaphorina citri feeding on aldicarb treated and untreated citru s plants. ................................ ................................ ..................... 89 4 3 Mean ( SE) NWEP and NPw for Diaphorina citri feeding on aldicarb treated and untreated plants. ................................ ................................ ................................ ................. 90

PAGE 10

10 4 4 Mean ( SE) waveform duration per probe (WDP) (s) for Diaphorina citri feeding on aldicarb treated and untreated plants. ................................ ................................ ........... 91 4 5 Mean ( SE) number of waveforms events per insec t (NWEI) for Diaphorina citri feeding on aldicarb treated and untreated citrus plants. ................................ ..................... 92 4 6 Mean ( SE) waveform duration event per insect (WDEI) (s) for Diaphorina citri feeding on aldicarb treated and untreated plants. ................................ .............................. 93 4 7 Mean ( SE) waveform duration per event (WDE) (s) for Diaphorina citri feeding on aldicarb treated and untreated plants. ................................ ................................ ........... 94 5 1 Mean ( SE ) WDI (s), W DEI (s), and NWEI for Diaphorina citri feeding on chlorpyrifos treated a nd untreated citrus plants. ................................ .............................. 110 5 2 Mean ( SE) N PI for Diaphorina citri on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, treated and untreated citrus plants. ................................ ............................... 111 5 3 Mean ( SE) PDI (s) for Diaphorina citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, treated and untreated control pla nts. ................................ ...... 112 5 4 Mean ( SE) WDI (s), WDEI (s), and NWEI for Diaphorina citri on feeding fenpropathrin treated and untreated citrus plants. ................................ ........................... 113 5 5 Mean ( SE) WDI (s), WDEI (s), and NWEI for Diaphorina citri feeding on imidacloprid treated and untreated citrus plants. ................................ ............................. 114 5 6 Mean ( SE) WDI (s), WDEI (s), and NWEI for Diaphorina citri feeding on spinetoram treated and untreated citrus plants. ................................ ............................... 115 5 7 Mean ( SE) WDI (s), PPW WDEI (s) and NWEI for Diaphorina citri feeding on spirotetramat treated and untreated citrus plants. ................................ ............................ 116 6 1 Mean ( SE) NPI for D. citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated citr us pla nts up to 28 DAT .......... 138 6 2 Mean ( SE) PDI for D citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetora m, and spirotetramat treated and untreated citrus pla nt s up to 28 DAT .......... 139 6 3 Mean ( SE) WDI for D citri feeding on ch lorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untr eated plants 1 DAT ................................ 140 6 4 Mean ( SE) WDI for D citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated plants 7 DAT ............................... 141 6 5 Mean ( SE) WDI for D citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated plants 14 DA T ............................. 142

PAGE 11

11 6 6 Mean ( SE) WDI for D citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated plants 21 DAT .............................. 143 6 7 Mean ( SE) WDI for D citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated plants 28 DAT). ............................ 144

PAGE 12

12 LIST OF FIGURES Figure page 2 1 A sian citrus psyllid EPG waveforms on sweet orange plants.. ................................ ......... 50 2 2 Asian citrus psyllid EPG non probing waveforms A) Waveforms np. B). Waveform z. ................................ ................................ ................................ ................................ ......... 51 2 3 Percentage of the total waveform duration (TWD) for female and male Diap horina citri feeding. ................................ ................................ ................................ ....................... 52 2 4 Percentage of the total waveform duration (TWD) for Diaphorina citri feeding under different light conditions. ................................ ................................ ................................ ... 53 3 1 P ercentage of the total waveform duration (TWD) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants. .................... 75 3 2 Percentage of th e waveform duration by insect (WDi) for Diaphorina citri feeding on young leaf tissues of imidacloprid treated and untreated plants. ................................ .. 76 3 3 Percentage of the waveforms duration by insect (WDi) f or Diaphorina citri feeding on mature leaf tissues of imidacloprid treated and untreated plants. ................................ 77 4 1 Percentage of the total waveform duration (TWD) for Diaphorina citri feeding on aldicarb t reated and untreated citrus plants. ................................ ................................ ...... 95 5 1 Representative Asian citrus psyllid EPG waveforms on sweet orange pla nts treated ..... 117 5 2 Asian citrus psyllid EPG waveforms C on. A) Control; B) spinetoram treated plants (difficulties in stylet penetration). ................................ ................................ .................... 118 6 1 Mean PDI and NPI for D citri feeding on chlorpyrifos, fenpro pathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated citrus plants through 28 DAT ...... 145 6 2 TWD for D citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram and spirotetramat treated and untreated citrus plants through 28 DAT .......................... 146 6 3 TWD for D citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and u ntreated citrus plants thr ough 28 DAT ........................... 147 6 4 Temperature, relative humidity and insecticide concentrations through 28 d after treatment. ................................ ................................ ................................ ......................... 148

PAGE 13

13 Abstract of Disse rtation Presented to the Graduate School of the University o f Florida in Partial Fulfillment of the Requirements for the Degree o f Doct or o f Philosophy ELECTRICAL PENETRATION GRAPH INVESTIGATIONS OF ASIAN CITRUS PSYL LID ( DIAPHORINA CITRI K UWAYAMA) FEEDING BEHAVIOR: EFFECTS OF INSECTICIDES ON THE POTENTIAL TRANSMISSION OF CANDIDATUS LIBERIBACTER ASIATICUS By R osana H arumi S erikawa August 2011 Chair: Michael E. Rogers Major: Entomology and Nematology An Electrical Penetration Graph (EPG) monitor was used to study the feeding behaviors of D iaphorina citri Kuwayama, vector of Candidatus Liberibacter asiaticus, the presumed causal agent of Huanglongbing (HLB). Effects of insect gender, presence of light, and ability and duration of insecticid es to disrupt D. citri feeding behaviors responsible for pathogen transmission were examined. Results showed that d uration of phloem ingestion was significantly longer for female D. citri compared to males. Gender based analysis showed that despite previou sly being considered a phloem feeder, D. citri performs similar amounts of xylem and phloem feeding. Additionally, xylem feeding was more likely to occur during the day and phloem feeding during the night. Application of the soil applied neonicotinoid inse cticide i midacloprid disrupted D. citri phloem feeding behaviors suggesting that application of this product to young trees can reduce the likelihood that trees will succumb to HLB. In contrast, the soil applied carbamate insecticide aldicarb did not disru pt D. citri phloem feeding behavior s. In fact, phloem ingestion by D. citri was increased on aldicarb treated plants indicating that use of this product could potentially increase pathogen acquisition rates where aldicarb treated diseased trees are present The foliar applied insecticides c hlorpyrifos, fenpropathrin, and imidacloprid

PAGE 14

14 disrupted D. citri phloem feeding behaviors on recently treated plants suggesting that these products will prevent pathogen transmission prior to death of the insect. While app lication of spinetoram altered D. citri feeding behavior, phloem feeding was not disrupted suggesting that pathogen transmission may still occur prior to insect death. In contrast, spirotetramat did not significantly disrupt D. citri feeding and thus will not prevent pathogen transmission from occurring. In experiments examining duration of feeding disruption provided by foliar insecticide applications, imidacloprid provided the longest duration of feeding disruption (28 d), followed by fenpropathrin (21 d) and chlorpyrifos (1 d). These results demonstrate that in addition to reducing vector populations in the field, certain soil and foliar applied insecticides can prevent pathogen transmission from occurring prior to death of pathogen carrying D. citri O ur findings are of immediate importance for Florida citrus growers who can use this information to improve their current HLB management programs.

PAGE 15

15 CHAPTER 1 INTRODUCTION Citrus is one of the most widely produced fruits in the world In 2011 production of 7 9.1 million metric tons (MMT) of citrus are expected to be produced (USDA 2011) In the USA, 10.9 MMT was produced in the 2009/2010 growing season with citrus production in Florida account ing for 65% of that total ; this was a 16% reduction compared to the previous season (USDA 2010a). In Florida citrus grove acreage totaled 517,100 acres in the 2009/2010 growing season. This was a 13,800 acre reduction from the 2008 09 growing season and almost 33% lower than in 2000 (USDA 2010a). Twelve percent of the cit rus losses since 2006 have been attributed to Huanglongbing (HLB) disease bacterial citrus canker, cold temperatures and commercial development (USDA 2010b). In addition the increased need for crop input s to manage citrus diseases and associated arthropo d pests has increased production costs making citrus less profitable C onsequently, some growers have become reluctant to replant citrus acreage lost to disease ( Pollak and Perez 2008). One of the main causes for the increased cost of citrus production is the recent introduction of Huanglongbing (HLB) disease. The causal agent of this disease in Florida is believed to be the bacteri um Candidatus Liberibacter asiaticus. This pathogen was first present in south Florida in 2005 in dooryard trees (Halbert 2005) Since its in itial discovery in Florida, HLB has been found in commercial citrus groves in all citrus growing areas of the state making eradication impossible In many parts of the world, this disease is referred to as Huanglongbing which translates to in Chinese based on one of the primary visual symptoms of the disease. In other parts of the world i t is also known as citrus greening disease, because it causes improper coloration of the fruit (Halbert and Manjunath 2004). Additional sympt oms include discolored

PAGE 16

16 and mottled leaves, yellowing of leaf veins, leaf and fruit drop, as well as small deformed fruit which have a bitter taste (Bove 2006). Thus, overall tree health and fruit production can be severely reduced by this disease. The exa ct origin of HLB is unknown (Yang 2006) ; however it was first described i n southern Asia as early as 1919 (da Graca and Korsten 2004 Bove 2006). The ca usal putative agent of HLB was initially discovered with electron microscopy in 1970 (Bove 2006). The d isease is thought to be caused by gram negative bacteri um and there are three known bacterial species thought to be responsible for this disease: Candidatus Liberibacter africanus (Laf), Candidatus Liberibacter asiaticus (Las) and Candidatus Liberibacter a me ricanus (Lam) (Bove 2006). Laf is found in Africa and is vectored by the African citrus psyllid, T r y oza erytreae Del Guerico (H emi ptera: Psyllidae). Lam is only found in Brazil and is vectored by Asian citrus psyllid, Diaphorina citri Kuwayama (Hemi ptera : Psyllidae) Las, the most severe and widely spread, occurs in Florida, Brazil, throughout Asia, the Indian subcontinent and neighboring islands, and the Saudi Arabian peninsula and i t is also vectored by D. citri (Bove 2006, Polek et al. 2007). The Asian citrus psyllid is a phloem feeding insect originating from the Far East and Asia (Mead 1977, Halbert and Manjunath 2004) and was first described on citrus in Taiwan in 1907. It is widely distributed throughout many of the citrus growing regions of the wor ld including Asia, the Middle East, Central and South America, and the Caribbean (Catling 1970, Halbert and Nez 2004). In the Americas, it was first detected in Brazil in 1942 (Lima 1942) but was not found in Florida until 1998 (Halbert 1998). Since its discovery in Florida, D. citri has been recorded from all citrus producing counties within the s tate. It has now been recorded from Alabama, Arizona, California, Georgia, Hawaii, Louisiana, Mississippi, South Carolina, and

PAGE 17

17 Texas (USDA 2011b). In the absenc e of the HLB pathogen, D. citri was not considered an import ant pest of citrus in Florida. The direct damage caused by D. citri affects mostly new flush and thus p e st control measures were only used to keep psyllid populations low on young trees, o n which new growth needed to be protected to promote tree development and bring trees into production in as short a time as possible. Longevity of adult D. citri varies from 51 117 days depending on the temperature. Following eclosion from the egg stage, D. citri complete five nymphal stages prior to molting to the adult stage. The developmental period from egg to adult is temperature dependent and ranges from 14 49 d. The optimal development temperature is 25 28 o C with mean generation time of 28.6 days (Liu and Tsai 2000). Since the discovery of HLB in Florida, control of HLB has become the primary focus of ci trus pest management programs. Currently, the three main components of HLB management strategies in Florida include: 1) the planting of certified disease free citrus trees, 2) identification and removal of Las infected trees from groves, and 3) effective psyllid control utilizing broad spectrum insecticide applicat ions (Brlansky and Rogers 2007). More recently, as an alternative to the removal of Las infect ed plants, the use of foliar nutritional sprays to maintain the health of diseased trees already in production are being used (Spann et al. 2010). Regardless of whether growers choose to remove Las infected trees or use supplemental nutrient sprays to main tain fruit production on diseased trees use of i nsecticide applications continues to be cons idered necessary. The rationale for insecticide use is to maintain vector populations at low levels so that the rate of disease spread will be reduced. Similar app roaches to managing disease spread via controlling vector populations have been practiced in other cropping systems (Perring et al. 1999 ).

PAGE 18

18 As citrus growers remove diseased trees, either because of HLB or other endemic diseases that render trees unproducti ve, they must replant new trees in order to maintain long term economic viability of their g rove (Morris and Muraro 2008). The ability to plant new trees and bring them into production where HLB is present is a significant concern for citrus growers. Compa red to larger fruit bearing trees, nonbearing trees produce more new leaf growth (flush) throughout the year, which is required by D. citri for oviposition and subsequent nymphal development. Thus, young trees are thought to be more attractive to D. citri than mature trees and at greater risk of becoming infected with HLB compared to mature trees which produce fewer flushes per y ear (Brlansky and Rogers 2007). Because nonbearing trees need to be continually protected due to the frequent production of flush, application of systemic insecticides with extended residual activity is most desirable. Currently, two soil applied systemic insecticides are commonly used in Florida for D. citri control on nonbearing citrus; imidacloprid (Admire Pro 4.6F, Bayer CropSci ences, Research Triangle Park, NC) and thiamethoxam (Platinum 75SG, Syngenta CropProtection, Inc., Greensboro, NC) (Rogers et al. 2011). However due to pesticide labeling restrictions that limit the use rate on a per acre basis, n either of th e se products effectively control D. citri on large (> 3 m height) bearing trees due to the increased canopy volume which dilutes the effectiveness of the systemic pesticide application. Aldicarb (Temik 15 G, Bayer CropSciences, Research Triangle Park, NC) is the only s oil applied systemic insecticide which has been used with some success for controlling D. citri on mature trees (Rogers et al. 2011) however, its usage will be discontinued by the end of 2011 due to food safety concerns (EPA 2010). Consequently, the optio ns for psyllid control on bearing trees are limited to broad spectrum foliar applied insecticides.

PAGE 19

19 The types of broad spectrum insecticides used by citrus growers in Florida to control D. citri includes several classes with different modes of action includ ing acetylcholinesterase inhibitors such as chlorpyrifos (Losban 4E, DowAgroscience, Indianapolis, I N ), the acetylcholine receptor stimulator imidacloprid (Provado 1.6F, Bayer CropSciences, Research Triangle Park, NC), sodium channel modulators including f enpropathrin (Danitol 2.4EC, Valent, Libertyville, IL ), and the lipid biosynthesis inhibitor spirotetremat (Movento 240 SC, Bayer CropSciences, Research Triangle Park, NC). S ome of these insecticide s require ingestion by the insect or absorption through the cuticle following contact in order to induce toxic effects In some cases the effects are not always immediate. Since the HLB putative pathogen is a phloem limited bacterium, acquisition and inoculation by D. citri occurs during the feeding process. Th erefore understanding the feeding behaviors of D. citri as well as the characteristics of pathogen transmission are important for a better understanding of disease epidemiology and to develop effective vector control strategies. There a re several reports in the literature regarding transmission of Las by the D. citri ; however the results are contradictory. Capoor et al. (1974) investigated Las transmission by D. citri and reported that the psyllid requires a minimum of 15 min to 24 h of feeding to acquire the pathogen and a minimum of 15 min to 1 h for successful inoculation implying high transmission efficiency Conversely Xu et al. (1988) reported that successful Las acquisition required a feeding access period of 5 7 h. Vuuren and Merwe (1992) and Bui tenda g and von Broembsen (1993), investigating T erytreae found that this psyllid only acquires Laf after a 24 h feeding access period. More recently, the rate of inoculation of healthy plants by single adult D. citri was found to be only about 4 6% but increased to approximately 88% when groups of 100 individuals were confined on a single plant (Pelz Stelinski et al. 2010).

PAGE 20

20 The nymphal stage(s) during which D. citri are able to acquire the pathogen have not been thoroughly studied. However, fourth and fi fth instar D. citri have been reported to successfully inoculate healthy plants after a 1 25 d latency period post acquisition ( Capoor et al 1974, Vuuren and Merwe 1992, Roistacher 1991 Xu et al. 1988) In addition, Pelz Stelinski et al. (2010) found th at successful inoculation by adult D. citri is much higher if Las is acquired by the nymphal stage compared to acquisition and subsequent inoculation by adult D. citri The past studies of Las transmission by D. citri have measured acquisition and inocula tion efficiency as a result of the duration of feeding access periods provided for D. citri Indeed, successful pathogen acquisition and inoculation processes depend on the amount of time t hat the vector feeds (Power 1991 ) However, additional experimental variables, such as gender of the vector and environmental factors (light versus dark conditions) can significantly influence the outcome of such studies (Perring et al 1999). Therefore, a better understanding of how these variables affect D. citri feeding behavior will be important in the design of future studies on the transmission of Las in order to provide more consistent and repeatable results. Electrical penetration graph (EPG) monitors have been used for detailed studies of the feeding behaviors of many sap sucking insects (Powell 1991, Collar et al. 1997a, Tjallingii and Prado 2001, Fereres and Collar 2001). Investigating the transmission of barley yellow dwarf virus by Rhopalosiphum padi (L.) Prado and Tjallingii (1994) correlated phloem associate d stylet activities denoted as waveforms E1 and E2 with salivation and ingestion, respectively Jiang et al. (2000) linked salivation into phloem (E(pd)1) with inoculation of tomato yellow leaf curl virus by Bemisia tabaci (Gennadius) In that study, the inoculation efficiency was positively correlated with the total number and duration of E(pd)1 events In addition, a 1.8 min period of salivation was sufficient for successful pathogen inoculation (Jiang et al. 2000). U sing an EPG

PAGE 21

21 monitor to examine the f eeding difference between genders o f Frankliniella occidentalis (Pergande) ( Wetering et al 1998) showed that while female thrips performed longer and more frequent probes than male thrips males transmitted the pathogen at a higher rate due to higher mob ility as a result of shorter durations of probing and ingestion behaviors performed ( Wetering et al 1998). EPG monitors have also been used to study the effects of insecticide applications on insect feeding behavior. Joost and Riley (2005 ) observed that F occidentalis probed more frequently and for longer periods of time on imidacloprid treated plants than on untreated tomato plants. Their results suggest an increase in inoculation of tomato spotted wilt virus a persistent and circulative virus, on imid acloprid treated plants compared with controls. In contrast, when feeding on imidacloprid treated tomato plants, F. fusca exhibited a significant decrease in the number of prob es per insect and decreased probing duration when compared to untreated plants ( Joost and Riley 2005). Since imidacloprid can work either as an agonist and antagonist, Joost and Riley (2005) suggested that the differences in the feeding behavior of those two species are related to the mode of action of the imidacloprid on the differen t thrips. In addition, Collar et al. (1997b) did not observe any significant differences in probing behavior of Myzus persicae (Sulzer) on imidacloprid treated versus untreated pepper plants. The first EPG monitoring study of the feeding behavior of a psyl lid was performed with the pear psylla, Psylla pyricola Foerster, using an AC monitor ( Ullman and Mclean 1988) S alivation ( waveform S) and ingestion ( waveform I) were observed for both nymph s and adult pear psyllids. Waveforms from the AC monitor availab le at the time could not distinguish between phloem versus xylem ingestion. However, while histological examination showed that both nymphs and adults ingest from all types of leaf cells xylem, phloem and bundle sheath cells were found to be the preferred sites for

PAGE 22

22 ingestion. More recently, Bonani et al. (2010) used a Giga 8 DC monitor to identify and histologically correlate (define) D. citri feeding waveforms, based in part on their similarity to the well known aphid waveforms. In addition, Bonani et al. (2010) observed that detectable HLB pathogen acquisition occurred after 1 h of phloem ingestion (waveform E2) by female D. citri However, pathogen inoculation could not be verified but it is likely that it occurs during phloem penetration or salivation ( waveforms D and E1 respectively). Bonani (2009) also observed that when D. citri fed on young leaves phloem ingestion (waveform E2) was longer and more frequent compared to feeding bouts on mature leaves Additionally, 35% more psyllids reached and inges ted phloem sap on young leaves compared to mature leaves Despite the considerable amount of past studies that have been conducted examining the transmission of Las by D. citri more detailed information is needed regarding D. citri feeding behavior in or der to develop management strategies that are effective in reducing the spread of HLB. To date, studies inve stigating the factors affecting D. citri feeding behavior have not been conducted. Furthermore, while studies investigating the effects of insectici des on pathogen transmission have been conducted with other insects, it would be presumptuous to make detailed study. As detailed above, insecticides can have variable effects on feeding behaviors of different groups of insects, even within the same taxonomic order. Therefore, the objectives of this dissertation were: Determine if gender based differences in feeding behavior exist for D. citri (Chapter 2); Exa mine the effects of dark and light conditions on the feeding behavior D. citri (Chapter 2); Determine whether the use of soil applied and foliar applied insecticides can disrupt D. citri feeding behaviors responsible for pathogen transmission (Chapters 3, 4 and 5);

PAGE 23

23 Determine the duration of D. citri feeding disruption that can be expected following the application of foliar applied insecticides (Chapter 6).

PAGE 24

24 CHAPTER 2 GENDER DIFFERENCE AN D EFFECT OF LIGHT AND DARK ON ASIAN CITRUS PSYLLID ( DIAPHORINA CITRI K UWAYAMA) FEEDING BEH AVIOR There have been numerous past studies regarding the transmission of Las by D. citri however the collective results reported are inconclusive and contradictory. Capoor et al. (1974) investigated Las transmission by D. citri and re ported that the psyllid requires a minimum of 15 min to 24 h of feeding to acquire the pathogen and a minimum of 15 min to 1 h for successful inoculation implying a high efficiency of transmission. In contrast, Xu et al. (1988) reported that feeding acces s periods of 5 7 h were sufficient for successful inoculation while 1 3 h feeding access periods were not. Additional studies indicated a D. citri feeding access period of 24h (Inoue et al. 2009) and 1h of phloem ingestion (Bonani et al. 2010) for successf ul pathogen acquisition. However, Pelz Stelinski et al. (2010) recently found that levels of Las are not detectable in adult D. citri during the first 7 days after feeding on an infected plant, suggesting a considerable latency period prior to the ability of D. citri subsequently inoculate a plant following bacterial pathogen acquisition. Successful inoculation was found to require an average feeding access period of 1 d (Pelz Stelinski et al. 2010). Nonetheless, recent investigations of Las transmission by D. citri have only sought to define bacterial pathogen transmission based on duration of feeding access periods. However, feeding behavior, which is fundamental to transmission of insect vectored pathogens, can be influenced by variables other than just d uration of feeding access period. The feeding behaviors associated with pathogen transmission can be influenced by variables including host plant recognition and acceptance, pathogen specificity with the vector, vector gender and age (Perring et al. 1999, Weintraub and Beanland 2006). In general, all acquisition and inoculation processes depend on the sequential steps involved in the feeding process. The sequential behaviors associated with insect feeding are plant surface exploration,

PAGE 25

25 labial dabbing, test probing and the subsequent probing of extended duration (Perring et al. 1999). Several stimuli such visual and chemical, influence those feeding behaviors. The knowledge of those stimuli that interfere with insect feeding behaviors can be manipulated to pr ovide control of pathogen transmission (Perring et al. 1999). One factor that affects the overall success of the pathogen acquisition and inoculation processes is the amount of time t hat the vector feeds (Power 1991 ). Studies on factors that might affect this feeding time, such as gender and presence or absence of light, could help to better define the pathogen transmission process Female whiteflies, Bemisia tabaci Gen., have been shown to be more efficient in the transmission of tomato yellow leaf curl g eminivirus than males (Caciagli et al. 1994). In addition, a higher percentage (55%) of female aster leafhoppers, Macrosteles quadrilineatus Forbes, were able to transmit phytoplasma yellows to lettuce than males (35%), although the disease spread pattern was significantly more clustered for female than males (Beanland et al. 1999) EPG studies of the feeding behavior of the thrips Frankliniella occidentalis (Pergande) showed that female thrips performed longer and more f requent probes than male thrips (We tering et al. 1998 ). However, male thrips were more efficient in pathogen transmission due to their frequent moving behaviors associated with the fact that they fed for shorter durations per probe (Wetering et al. 1998 ) Studies of host plant selection by D. citri have shown that visual cues are also important factor regulating psyllid feeding behavior (Wenninger et al. 2009). More in depth studies of D. citri feeding behavior are still needed to gain a better understanding of the factors affecting feeding Gender differences and presence of light and dark on the feeding behavior of D. citri are not known Therefore, the objective s of our study were to: 1) determine whether male and female

PAGE 26

26 D. citri differ in their feeding behavior, and 2) examine the effect s of the presence of light on D. citri feeding activities. These are potentially important variables which could affect the outcome of experiments investigating the transmission of Las by D. citri Material s and Methods Plants and Insects Plants used in ex periments consisted of s Citrus s inensis (L.) Osbeck) seedlings (15 20 cm tall) grown i n 120 ml tubes containing Fafard Citrus Mix (Fafard, Agawam, MA) Seedlings were maintained in a pathogen free greenhouse at 29 3 C and 60 80% RH. All s eedlings were planted and maintained identically to minimize interplant variation. Four weeks prior to initiation of the experiment s, plants were pruned to force production of new leaf flushes Only plants with young leaves (soft leaves, less lignified) we re used in the experiments. Adult s D. citri (10 15 d) used in experiments were obtained from a greenhouse colony free of Ca Citrus aurantium 29 3C with a photoperiod of 12:12 (L :D) h. Prior to use in EPG recordings, D. citri were transferred to a rearing cage (61cm x 61cm x 91cm, Bioquip, Rancho Domingues, CA) EPG Equipment The EPG system consisted of a Giga 8 amplif ier (Department of Entomology, Wageningen Agricultural University, the Netherlands), analog to digital (AD) converter, and custom software for digitizing, recording, and storing EPG recordings. The Giga 8 is an amplifier whose primary circuit is based on a direct current (DC) system (Tjallingii 1978). The monitor contains eight headstage amplifiers (sometimes termed probes) with a fixed input resistance of 10 9 of 50 100X. Electrical signals are created by a closed circuit formed by the insect a nd plant. A headstage amplifier is connected to the insect and the plant electrode is inserted into the soil

PAGE 27

27 surrounding the plant. When the insect is in contact with the plant the circuit closes, and waveforms are produced in response to the different pr ocesses (biological and physical) inherent to the insect and plant association (Walker 2000, Tjallingii 2000). Because the Giga 8 is very sensitive to electrical noise, the headstage amplifier and the plant insect preparations were housed within a Faraday cage (152 cm x 62 cm x 122 cm). The Giga 8 was connected to an analog to digital (A/D) converter (DATAQ DI 710UHB, DATAQ instruments, Akron, Ohio) connected via a USB port to a personal computer. All EPG recordings were conducted with 100X gain, a conver sion rate of 100 samples per second at 26 2 C and the substrate voltage set to 15 mV DC. EPG Recording After the 48 h acclimation period, psyllids were collected individually using a wet paint brush and immobilized by holding the tips of the wings wit h a pair of soft forceps (Bioquip Products Inc, Rancho Domingues, CA). A gold wire (18.5 m in diameter [sold as 0.0010 in], Sigmund Cohn Corp., Mt. Vernon, NY) was cut to 1.5 cm in length and attached to D. citri using silver glue (white glue: water: silv er powder (8 10m), 1:1:1 [v:v:w], Inframat A dvanced M aterials, Manchester, CT). This length of wire was chosen because it allows psyllids to move freely on the leaf surface when selecting a feeding site, but is less likely to come in contact with the plan t surface to cause interference in the EPG signal. The gold wire was attached to a psyllid by creating a small loop on one end of the wire and then dipping the loop in silver glue. The glue coated loop was then held against the dorsal portion of the insect thorax until dry. The other end of the wire was connected using the silver glue to a copper wire (approx. 0.2 mm in diameter, 3 cm in length) which was soldered to a copper nail electrode inserted into the female BNC connector on the headstage amplifier.

PAGE 28

28 After the psyllid was securely attached to the gold wire, the headstage amplifier was connected to the input of the amplifier control box and the psyllid was placed on the abaxial side of a young leaf on a potted citrus seedling The abaxial side of the le af was chosen according to the observed preference of psyllid feeding. The leaf was held with the abaxial side up (i.e., inverted) by placing it on top of another leaf from the same plant and securing the two leaves together using a loop made of Scotch Ma gicTM tape (3M Center, Saint Paul, MN). Citrus plants were connected individually to each monitor channel. The plant electrode (copper wire, approx. 2 mm in diameter, 10 cm long) was inserted in the soil at the base of each plant. The soil within each po t was moistened with water prior to the start of each recording to facilitate a closed electrical circuit. Pots were then placed on plastic trays to prevent contact with the Faraday cage. Access time for psyllid feeding on citrus seedlings was 12 h for eac h recording, in both experiments. Data were collected, stored, and displayed in real time using WinDAQ Pro software (DATAQ I nstruments, Akron, O H ). The duration of each waveform event was measured post acquisition using DATAQ Windaq Waveform Browser sof tware, version 2.40 (DATAQ I nstruments, Akron, O H ). Asian Citrus Psyllid Waveform Terminology The terminology used for waveforms was based on the conventions of Backus (2000) and the D. citri waveform correlations of Bonani et al. (2010). A probe is the amount of time from s withdrawal from the leaf. A single probe can contain multiple waveform events such as salivation and ingestion (Backus et al. 2007). These behaviors are represented by different waveforms defined by Bonani et al. (2010), summarized below.

PAGE 29

29 Waveform C (Figure 2 1 B ) was histologically associated with stylet pathway activities and salivary sheath secretion through parenchyma cells. Waveform C is complex in appearance and is comprised of all activities that are occurring during stylet penetration through different leaf tissues. Consequently, it may resemble other types of waveforms. The average frequency of waveform C is 11.5 19.0 Hz (Bonani et al. 2010). Waveform D (Figure 2 1 C ) is co rrelated with salivary sheath termini in phloem tissue (though exactly which of the four phloem cell types is unknown); it seems to represent a transition behavior from pathway to phloem phase. There is no analogous waveform produced by aphids. Consequentl y, the precise stylet activities by D. citri resulting in this waveform are completely unknown. However, they occur after waveform C and end prior to initiation of waveform E1. The average frequency of waveform D is 1.0 3.5 Hz (Bonani et al. 2010). Wavefor m E1 (Figure 2 1 D ) represents the beginning of phloem phase and starts after a potential drop (pd) marking the end of waveform D. Consequently, E1 waveforms only occur after a D waveform. Histological correlation of salivary sheath termini showed E1 to occ ur in the phloem, however again, no specific cell type could be identified. E1 is hypothesized to correspond to salivation into phloem sieve elements by D. citri based on its similarity with the well correlated E1 waveforms of aphids. E1 is at a negative voltage level, supporting th e suggestion that the stylets tips may be located intracellularly. Therefore, this waveform represents the point in D. citri feeding at which inoculation of the circulative, phloem limited Las bacteria seems most likely to occur although some inoculation may also occur during waveform D. The average waveform frequency of E1 is 5.0 7.5Hz (Bonani et al. 2010). Waveform E2 (Figure 2 1 E ) is also part of phloem phase, and always occurs after an E1 waveform event and, like E1, is on the negative voltage level. E2 definitely corresponds to D.

PAGE 30

30 citri phloem ingestion, because Bonani et al. (2010) correlated E2 with salivary sheath termini in phloem tissue, as well as with Las acquisition after performance of E2 for approximately 1 h. The average frequency of E2 waveforms is 3.0 8.0 Hz (Bonani et al. 2010). Waveform G (Figure 2 2 F ) is histologically correlated with salivary sheath termini in xylem vessels and has an average frequency of 5.0 7.0 Hz. Again based on an appearance similar to the G waveform of aphids, psyllid waveform G is hypothesized to represent xylem ingestion. G was performed by only 25% of D. citri in 160 h of recordings (Bonani et al. 2010). These for maintaining water balance. Experimental Design Experiment 1. G ender b ased d ifference s in D. citri feeding behavior. A total of 15 plants containing young leaves were used for each treatment. Prior to EPG recordings, psyllids from the acclimatization c age were taken and sexed under a microscope and 15 psyllids of each gender were used in this experi ment. The feeding behavior of single psyllid was individually recorded per plant and each psyllid EPG recording was considered a unique replicate, thus provi ding 15 replicates per treatment for analysis. Experiment 2. E ffect s of light and dark on D. citri feeding behavior. A total of 15 plants were also used in this experiment for each treatment. For the dark treatment one of the Faraday cage w as totally cove red with black card boards causing complete darkness. Individual female psyllids were recorded per plant providing a total of 15 replicates for each treatment. Dark and light treatments were recorded at the same time and unique plants were used for each p syllid recording. Plants and psyllids were acclima ted for 1h under their respective treatment prior recordings.

PAGE 31

31 Statistical Analysis The number of waveform events and duration were analyzed using biologically non sequential parameters, as described by Bac kus et al. (2007). Those parameters were grouped by four different levels: cohort, insect, probe and event. At the cohort level, the parameters analyzed were total probing duration (TPD), total number of probes (TNP), total waveform duration (TWD), and tot al number of waveform events (TNWE). At the insect level, the parameters analyzed were probing duration per insect (PDI), waveform duration per insect (WDI), and number of probes per insect (NPI). At the probe level, the parameters analyzed were waveform d uration per probe (WDP), and number of waveform events per probe (NWEP). At the event level, the parameters analyzed were probing (i.e. penetration) duration per event (PDE), total number of probing (penetration) events (TNPE), number of waveform events pe r insect (NWEI), and waveform duration event per insect (WDEI) (Backus et al. 2007). Another parameter analyzed that was not included in Backus et al. (2007) is the proportion of individuals that produced the specific waveform type (PPW); this parameter wa s defined in Bonani et al. (2009, 2010). For analysis of probe activities (NPI and PDI) waveforms z and np were combined as one waveform type because z and np are both non probing behaviors (see first section under Results, below). square tes t was performed to test the goodness of fit (PROC GLIMMIX, SAS Institute 2001). The waveform duration data were log transformed before statistical analysis, to improve homogeneity and reduce variability. Data were analyzed by protected ANOVA (PROC GLIMMIX, SAS Institute 2001) with the least significant difference (LSD) test (LSMEANS, SAS Institute 2001) used for pairwise comparisons, to determine whether the waveform parameters analyzed were significantly different between psyllids genders, and light

PAGE 32

32 and da rk treatments Data were also compared in a similar manner among waveforms within treatments. Results New W aveform Characterization and C orrelations Waveform np (Figure 2 2 A ) is a baseline waveform that is highly irregular and spikey. It was visually correlated in every recording with the ins ect moving on the plant. Further correlation research has been performed and will be described in detail elsewhere (Youn et al., in press ). Waveform z (Figure 2 2 B ) is similar to baseline with low voltage levels, named z by Backus et al. (2007). This wave form re presents a non probing behavior visually correlated herein with times when D. citri are motionless, being that they are either resting, dead or have moved off the plant. Thus, unlike most EPG studies, D. citri recordings reveal two types of non prob ing behavior. Further correlation of these waveforms will be described elsewhere (Youn et al ., in press ). Experiment 1 G ender based D ifference s in D. citri Feeding B ehavior. One hundred percent (PPW) of female and male D. citri performed pathway/stylet p enetration waveforms (waveform C) and non probing/w alking activities (waveform np) One hundred percent of female D. citri performed non walking/non probing activities (waveform z), while only 73.3 % of the male D. citri performed the same waveform. Forty percent o f both genders penetrated the phloem and salivated (waveforms D and E1, respectively). However, only 20% of the female D. citri performed phloem ingestion (waveform E2), although all the males which penetrated into the phloem ingested phloem sap ( 40%). Approximately 67% of the male D. citri ingested xylem fluid (waveform G) while only 27% the female D. citri performed the same activity (Table 2 1).

PAGE 33

33 Cohort level. D. citri had a total access period of 648,000 s, during which time the 15 female D. cit ri probed a total of 503 times (TNP) and spent 345,611.28 s (TPD) with their stylets inserted in the leaf. For th e remaining duration (53.3 %) of total access period, female D. citri performed non probing activities such as waveform np and z. In contrast, m ale D. citri spent most of their time (67.8 %) performing probing activities (TPD = 439,027.37 s) producing a total of 470 (TNP) probes. The percentages of each waveform duration performed in TPD for both female s and male s is represented by the total wavef orm duration (TWD) shown i n Figure 2 3. Insect Level. The general feeding behavior s for male and female D. citri did not differ significant ly. The number of probes per insect (NPI) ( F = 0.00; df = 1, 28 ; P = 0.97) and the probe duration per insect (PDI) ( F = 3.28; df = 1, 28 ; P = 0.08; Table 2 2) were similar However, when analyzing the waveform duration per insect (WDI), phloem ingestions (waveform E2) were significantly longer for female s than male s ( F = 9.66; df = 1, 7 ; Table 2 1). Probe level. There w as no difference in the number of waveform events per probe (NWEP) between males and females (Table 2 3) Similarly, the re was no difference in the waveform duration per probe (WDP) between males and females. However, male D. citri produced stylet penetrat ions (waveform C) which were longer in duration compared to females ( F = 13.13; df = 2, 971 ; Tabl e 2 4). Event level. In contrast to that observed with PDI and NPI, the re was a significant difference between males and females in probe duration per event (P DE) (Table 2 1) M ale D. citri produced longer PDE compared to female s There were significantly more waveform events per insect (NWEI) for female s than males. Also females produced more phloem ingestion waveform events ( E 2) than did males ( F = 7.58; df = 1, 7 ; Table 2 5). In addition there was no

PAGE 34

34 difference in waveform duration per event per insect (WDEI) between males and females (Table 2 6). Summary of results. There were slight differences in the feeding behaviors of males and females. W hen not probing female s remained motionless (z) for longer durations per insect than males due to longer standing events per insect and the high number of standing still events. In addition, female s performed more walking (np) per insect due to longer walking duration per event and a greater number of event s per insect. When stylet probing, female s exhibited shorter pathway activities (C) per insect because each event was shorter. P hloem contact (D) was not significantly different between males and females for any of t he parameters analyzed. F emale D. citri performed longer durations of phloem contact per insect, however phloem contact duration events were slightly shorter in females than males. T he frequency of the phloem contacts performed per probe was higher in fema les than males For phloem related behaviors phloem salivation (E1) did not differ between females than males. However, phloem salivation for females was half as long as in males. In contra st phloem ingestion (E2) duration per insect was statistically lo nger for female D. citri than males, due to longer phloem ingestion events and the number of times this event was performed per probe This then lead to a longer duration of phloem ingestion per probe with a higher frequency of phloem ingestion events per insect. In contrast, xylem ingestion (G) was shorter in duration per insect because each event was shorter, performed fewer time, in numerically fewer probes, leading to less time per probe and fewer events per insect. Overall female s remained motionless and walked more frequently and for longer Female insects searched less often for vascular tissue, for shorter durations and made fewer probes. Female s insects generally took more time to find the phloem but spent less time salivating. Once

PAGE 35

35 feeding on the phloem females salivated and subsequently ingested for longer duration s compared to male s In contrast, female s found the xylem and ingested from this tissue for shorter durations than males. F eeding b ehavior of f emale p syllids (within treatment analysis) Insect level. Analysis o f the waveform duration per insect (WDI) indicated high variation among the waveforms ( F = 35.33 ; df = 7, 58 ; P < 0.0001). Females perform ed behaviors related to waveform np, pathway (C) and phloem ingestion (E2) for longer durat ion compared to other behaviors performed. W aveforms D and E1 were the shortest in duration as expected (Table 2 1). Probe level. Analysis of the number of waveforms events per probe (NWEP) ( F = 41.78 ; df = 6, 1034 ; P < 0.0001) showed a significantly hig h er frequency of waveform E1 when compared with other behaviors. The behaviors performed the least number of times were znp, C, and G, which were not significantly different from each other (Table 2 3). For waveform duration per probe (WDP) ( F = 14.61, df = 6, 1034 P < 0.0001), waveform E2 and G were the longest in duration and were not significantly different from one another (Table 2 4). Waveforms C, D, E1 had the shortest durations (Table 2 4). Event level. There was a significant effect of waveform du ration per insect WDEI ( F = 20.50, df = 7, 58 P < 0.0001 ). W aveforms E2 and G were the longest in duration and did not differ one another (Table 2 6 ). Waveform D and E1 were the shortest in duration (Table 2 6). There was a significant effect of the numbe r of waveforms events per insect (NWEI) ( F = 17.74, df = 7, 58 P < 0.0001). W aveforms np and C occurred most often and there were no significant difference in the number of timesthe other waveforms were performed (Table 2 5).

PAGE 36

36 F eeding b ehavior of male p sy llids (within treatment analysis). Insect level. Analysis o f the waveform duration per insect (WDI) showed that male D. citri had a slight ly higher variability among the waveforms than female psyllids ( F = 57.28, df = 7, 68 P < 0.0001). The longest WDI w as attributed to waveform C followed by wavefo rm np and shortest durations observed were waveforms D and E1 (Table 2 1). Probe level. The highest number of waveforms events per probe (NWEP) ( F = 19.43, df = 6, 978 P < 0.0001) was waveform E1, and the wave forms which occurred least often were znp and E2 (Table 2 3). Analysis of waveform duration per probe (WDP) ( F = 12.89, df = 6, 978 P < 0.0001) showed significant differences in duration among the waveforms. W aveforms G and E2 were the longest in duration compared to the other three waveforms (Table 2 4). Event level. Waveform duration events per insect (WDEI) ( F = 40.77, df = 7, 68 ; P < 0.0001) were also significantly different between the waveforms produced; waveforms C, E2 and G had the longest duratio ns whereas waveforms D and E1 were shorter in duration (Table 2 6). There were significant differences in the number of waveforms events per insect (NWEI) ( F = 53.21, df = 7, 68 P < 0.0001) with waveforms np and C performed more often than waveforms D, E1 E2 and G (Table 2 5). Experiment 2 Effects of Light and D ark on D. citri Feeding B ehavior. All D. citri tested performed both pathway/stylet penetration waveforms (waveform C) and non probing/walking activities (waveform np) in both light and the dark conditions Approximately 93% of the D. citri performed non walking/non probing activities (waveform z) in light while 86.7 % performed the same waveform in complete darkness Sixty percent of D. citri performed all the phloem activities (phloem penetration salivation and ingestion, respectively ; waveform s D, E1 and E2) in dark conditions while only 40%, 40%, and 33.3 % of the D. citri performed phloem penetration, salivation and ingestion, respectively in the presence

PAGE 37

37 of light Xylem ingestion (waveform G) was perfo rmed by 26 % of the D. citri in light and 53.3 % in dark conditions (Table 2 1). Cohort level. D. citri were given a total access period of 648,000 s, during which D. citri probed 599 times (TNP) and spent 446,698.11 s (TPD) with their stylets inse rted in the leaf in dark conditions During the remaining time, 68.9 % of the total access period, D. citri perfo rmed non probing activities ( waveform s np and z ) In contrast, in light D. citri spent less time performing probing activities (TPD = 295,208.2 6 s) (4 5.6 %), even though the number of probes was higher compared to dark conditions (TNP = 766). The percentage of time spent for each waveform performed in TPD in both light and dark conditions is represented by the total waveform duration (TWD) shown i n Figure 2 4. Insect l evel. Analysis of the probe duration per insect (PDI) indicated significant difference s between feeding behaviors in light and dark conditions ( F = 7.82; df = 1, 28) In complete darkness, D. citri produced probe s of longer duration co mpared to D. citri in light conditions (Table 2 2). With respect to waveform duration per insect (WDI), t here were significant differences between waveforms z ( F = 4.10; df=1, 25), np ( F = 4.13; df=1, 28), E2 ( F = 6.51; df=1, 12), and G ( F = 4.97; df=1, 10) (T able 2 1) W aveforms z, np, and G were longer in duration in light conditions while waveforms C and E2 were longer in duration in dark conditions T he re was a trend for the mean number of probes per insect (NPI) to be higher in the light than in dark condi tions ( F = 2.78; df = 1, 28; Table 2 2) but this trend was not statistically significant. Probe level. The presence of light did not affect the number of waveform events per probe (NWEP) (Table 2 3). In contra st for waveform duration per probe (WDP), wav eform C was significantly in dark compared to light conditions ( F = 13.29, df = 1, 1131 ; Table 2 4).

PAGE 38

38 Event level. For probe duration per event (PDE), like PDI at the insect level, there were significant differences in between light and dark conditions ( F = 18.91; df = 1, 1363 ; Table 2 2). Even though the number of waveform events per insect (NWEI) were not significantly different between light and dark for any of the waveforms (Table 2 5), the waveform duration event per insect (WDEI) was longer for wavefor ms z ( F = 5.46; df=1, 25) in light whereas waveforms C ( F = 5.12; df =1, 28) and E2 ( F = 5 19; df =1, 12) were longer in dark conditions (Table 2 6). Summary of results. D. citri that were recorded in light performed feeding behaviors differently than thos e recorded in dark In light conditions, D. citri stood still (z) for significantly longer durations per insect, and more often than in light In light, D. citri performed significantly more walking events (np) per insect due to longer walking duration per event and a higher number of event s per insect. In light conditions, pathway activity (C) events per insect were shorter than in dark conditions; however, there were more pathway activities performed in light than dark. Phloem contact (D) wa s not signifi cantly different for any of the parameters, but in light, phloem contact was twice as long as in dark conditions because of longer durations on the phloem contact event per insect and the higher num ber of phloem contacts per probe. The re was no difference between phloem salivation activities in light and dark conditions. In contrast, phloem ingestion (E2) duration per insect was statistically longer in dark than light conditions because of longer and less frequent events per probe, resulting in longer phloe m ingestions. In contrast to phloem ingestion, in light conditions, D. citri performed xylem ingestion (G) for significantly longer than in dark with longer duration events per insect and fewer probes probes resulting in less xylem ingestion events per pr obe and a slightly higher number of xylem ingestion events per insect.

PAGE 39

39 In conclusion, under light D. citri stood still and walked more frequently and for longer times, performing search for vascular tissue less frequently and than under light, despite a si milar overall number of probes. G enerally to the phloem was penetrated, they ingested for shorter durations than those under dark. In contrast, under light, D. citri found and ingested from the xylem more briefly than those under dark conditions. D. citri feeding behavior in light (within treatment analysis). Insect level There was a significant effect of light on waveform duration per insect (WDI) ( F = 79.36, df = 7, 63 P < 0.0001). D. citri performed significantly longer d urations of waveforms np, and C for significantly longer durations compared to other behaviors; waveforms D and E1 were the shortest in duration (Table 2 1). Probe level. There was a significant effect of light on the number of waveforms events per probe (NWEP) ( F = 72.62, df = 6, 1365 P < 0.0001). W a veforms E1, followed by waveform D, were performed most often Conversely, waveforms were znp, C, and G, which were not signifi cantly different from one another, were performed the least number of times (Table 2 3). For waveform duration p er probe (WDP) ( F = 37.19, df = 6, 1365 P < 0.0001), w aveform s E2 and G were longest in duration whereas waveforms C, and D were the shortest in duration (Table 2 4). Event level. There was a significant effect of light on w aveform duration events per in sect (WDEI) ( F = 40.66, df = 7, 63 P < 0.0001). W aveforms E2 and G were the longest in duration and not significantly different from one another (Table 2 6). W aveforms D and E1 were shortest in durations (Table 2 6). In contra st to WDEI, the number of wav eforms events per insect (NWEI) showed a lower variation ( F = 31.03, df = 7, 63 P < 0.0001) W aveforms np and C were performed most frequently compared to waveforms D, E1, E2 and G (Table 2 5).

PAGE 40

40 D. citri feeding behavior in dark (within treatment analysis) Insect level. There was a significant effect of dark conditions on waveform duration per insect (WDI) ( F = 78.42, df = 7, 76 P < 0.0001 ) W aveform np, C and E2 were characterized by the longest WDI (Table 2 1) whereas waveforms D and E1 were shortest i n duration (Table 2 1). Probe level. The highest number of waveforms events per probe (NWEP) ( F = 48.83, df = 6, 991 P < 0.0001 ) was waveform E1, followed by waveform D. W aveforms znp, C and G were peformed the least number of times (Table 2 3). There wa s a significant effect of dark conditions on ( F = 21.81, df = 6, 991 P < 0.0001) waveform duration per probe (WDP) with waveforms G and E2 having the longest durations while waveforms C, D and E1 were characterized by the sho rtest durations (Table 2 4). E vent level. There was a significant effect of dark conditions on w aveform duration events per insect (WDEI) ( F = 69.99, df = 7, 76 P < 0.0001). W aveform E2 was performed for the longest duration, followed by waveforms C and G ; w aveforms D and E1 were shor test in duration (Table 2 6). Unlike WDEI the number of waveforms events per insect (NWEI) had a lower variance among the waveforms ( F = 31.00, df = 7, 76 P < 0.0001).W aveforms np and C were performed the highest number of times while waveforms D, E1, E2 and G were performed less often (Table 2 5). Discussion The objective of the cu rrent study was to quantify the effects of gender and light conditions on the feeding behavior of D citri In general, female psyllids performed non probing activities (wavefo rm z and np) more often than males. A lso they performed phloem ingestion (E2) more often and for longer durations than males. Male psyllids performed stylet penetration activities (C) and xylem ingestion (G) more often and for longer durations compared to females. The number and duration of probes was not different between males and females. In light conditions,

PAGE 41

41 stylet penetration activities (C) and phloem ingestion (E2) were less frequent and shorter in duration. Phloem ingestion (E2) and xylem ingestion (G) were longer in light than dark for both sexes. The factors that affect feeding behavior are probably related to both insect physiology (insect age, reproductive stage) and ecology (risk of predation, microenvironment, and temperature) F eeding quantity depends on the nutrient requirements of the insect and potential phagostimulatory effec ts. Therefore it is common for female insect s to feed more in their r eproductive period D. citri is sexually mature 2 3 days after eclosion, and capable of oviposit ion a day after mating (Wenninger and Hall 2007). In our experiments, the insects were 10 15 days old at the time of recording and female and male D. citri were in the same cage during acclimation. Therefore, female s D. citri were likely reproductively mature This could partially explain why females fed on the phloem more than males. Female F. occidentalis (Pergande) also fed on phloem more than males, thus producing more feeding scars than males (Wetering et al. 1998) Male thrips fed occasionally and were m ore mobile; however, unexpectedly their transmission efficiency was greater than that of females (Wetering et al. 1998 ). In some cases, gender does not affect feeding behavior of plant feeding insects. For example, honeydew production by Bucephalogonia xan thophis (Berg) is not different between males and females (Miranda et al. 2008). Plants were acclimatized in the dark so d ifferences in D. citri feeding behavior between dark and light could be attributed to nutritional contents of the plant at different times of day. Since D. citri is a phloem feeder, sugar and amino content may be the main phagostimulants and could be representing an important factor regulating psyllid feeding activities. For example, B. xanthophis fed longer during the light phase proba bly due to xylem content (Miranda et al.

PAGE 42

42 2008). Even though xylem is poor in nutrients, there are several studies showing that sharpshooters are able to synchronize their feeding behavior according to the fluctuation in the chemical composition of the xyle m (Andersen et al. 1989, 1992 and Brodbeck et al.1993). However Miranda et al. ( 2008 ) did not measure the sap content from citrus, and no correlation was made between active sharpshooter feeding and xylem nutrition al content. Brodbeck et al (1993) showed that Homalodisca coagulata (Say), Homalodisca insolita (Walker) and Cuerna costalis (F.) feed more on crape myrtle during the photophase than the scotophase The amino acid content of c rape myrtle is 2.5 times higher in photophase compared to scotophase (B rodbeck et al.1993). In contrast the se same species fed more on periwinkle during the scotophase than photop hase which corresponded with 1.9 times higher amino acids concentrations during the scotophase than during the photophase Generally in herbaceous plants, the products of photosynthesis are accumulated in the leaves during the photophase and they leave the leaves during scotophase Although, the diurnal pattern of citrus is different, Goldschimitdt and Koch (1996) showed slight daily fluctuations of soluble sugars and starch levels in citrus leaves in which soluble sugars and starch levels were higher during the day and lower during the evening. Since soluble sugars from the leaves are transported through the phloem, sugar concentrations in the phloe m could be negatively proportional to that of xylem. Consequently, sugar concentrations in the phloem being higher during the night than during the day. Mating, oviposition and movement of D. citri occurs more frequently during the photophase than the sco tophase (Wenninger and Hall 2007). This is likelydue to the use of visual cues for host plant selection (Wenninger et al. 2009) and mating (Wenninger et al. 2008). This

PAGE 43

43 could explain the greater walking durations observed in the study under light than dark conditions. Xylem ingestion was greater during the light than during the dark. Overall, these experiments showed that female D. citri perform more phloem feeding behaviors compared to males and this phloem feeding behavior is more likely to occur at nigh t. Our results suggest that transmission efficiency is likely influenced both by gender and photophase.

PAGE 44

44 Table 2 1. Mean ( SE) waveform duration per insect (WDI) (s) and the proportion of individuals that produced a waveform type (PPW) for different gend ers of Diaphorina citri feeding and under different light conditions. Experiment 1: Gender Waveform Female Male WDI SE PPW WDI SE PPW p value z 5670.19 1198.67 15 /15 b 3925.36 1029.94 11 /15 c 0.5 555 n p 16294.09 2120.02 15 /15 a 12853.21 1890.12 15 /15 b 0.2945 C 20386.14 2119.77 15 /15 a 25552.90 1972.04 15 /15 a 0.0970 D 166.22 60.32 6 /15 c 103.64 30.12 6 /15 d 0.4344 E1 120.46 22.24 6 /15 c 204.16 52.05 6 /15 d 0.8475 E2 7836.0 0 3070.59 3 /15 ab 2090.43 453.29 6 /15 c 0.0171 G 3625.52 931.00 4 /15 b 4129.61 983.59 10 /15 c 0.8662 Experiment 2: Light conditions Light Dark Waveform WDI SE PPW WDI SE PPW p value z 829 7.87 1721.97 14 /15 b 4685.02 1601.17 13 b 0.0538 n p 17574.84 2475.78 15 /15 a 11162.15 1390.45 15 a 0.0517 C 16643.86 3038.89 15 /15 a 20835.76 2764.64 15 a 0.1679 D 214.70 110.62 6 /15 c 124.56 46.31 9 c 0.7048 E1 266.74 133.26 6 /15 c 200.08 64.58 9 c 0.5584 E2 4995.79 2747.51 5 /15 b 12941.43 3072.97 9 a 0.0254 G 4411.60 1105.95 4 /15 b 1836.89 509.39 8 b 0.0498 Different letters indicate significant differences (LSD) ( ) among waveforms within treatments

PAGE 45

45 Tabl e 2 2. Mean ( SE) probe duration per insect (PDI) (s) number of probes per insect (NPI), and probe duration per event (PDE) (s) for different genders of Diaphorina citri feeding and under different light conditions Experiment 1: Gender Female M ale PDI SE PDI SE p value 23040.75 2332.17 29268.49 2192.21 0.0836 NPI SE NPI SE p value 33.8 7.03 31.40 4.17 0.1066 PDE SE PDE SE p value 602.11 74.33 771.57 83.52 0.0005 Experiment 2: Light conditions Light Dark PDI SE PDI SE p value 19680.55 3010.29 29779.00 1996.34 0.0056 NPI SE NPI SE p value 44.66 6.77 31.46 5.69 0.1066 PDE SE PDE SE p value 385.38 47.80 745.73 86.71 <0.0001

PAGE 46

46 Table 2 3. Mean ( SE) number of w aveform event s per probe (NWEP) and number of probes by waveform (NPw) for different genders of Diaphorina citri feeding and under different light conditions Experiment 1: Gender Waveform Female Male NWEP SE NPw NWEP SE NPw p value z np 1.00 0.00 507 c 1.00 0 471 c n/a C 1.04 0.01 503 c 1.08 0.01 470 b 0.0838 D 1.66 0.33 9 b 1.28 0.18 7 b 0.3904 E1 2.11 0.53 9 a 1.85 0.34 7 a 0.8077 E2 1.50 0.50 4 b 1.00 0 6 bc 0.2415 G 1.14 0.1 4 7 c 1.23 0.13 17 b 0.7204 Experiment 2: Light conditions Waveform Light Dark NWEP SE NPw NWEP SE NPw p value z np 1.00 0.00 670 d 1.00 0 472 d n/a C 1.05 0.01 662 d 1.07 0.17 471 d 0.1981 D 2.00 0.42 10 b 1.69 0.30 13 b 0.5410 E1 2.66 0.55 9 a 2.53 0.46 13 a 0.8588 E2 1.40 0.24 5 c 1.25 0.13 12 c 0.5655 G 1.00 0.00 6 d 1.12 0.12 8 cd 0.4082 orms within treatments

PAGE 47

47 Table 2 4. Mean ( SE) waveform duration per probe (WDP) (s) for different genders of Diaphorina citri feeding and under different light conditions. Experiment 1: Gender Waveform Female Ma le WDP SE WDP SE p value z np 649.68 65.51 b 501.01 46.61 b 0.7792 C 608.08 84.97 c 815.51 108.95 b 0.0003 D 110.81 41.80 bc 88.84 23.31 b 0.8745 E1 80.31 15.22 bc 174.99 47.86 b 0.4952 E2 5877 2738.56 a 2090.43 453.2 9 a 0.0772 G 2017.73 610.06 a 2429.18 436.36 a 0.8233 Experiment 2: Light conditions Waveform Light Dark WDP SE WDP SE p value z np 566.85 64.06 b 483.69 70.22 b 0.7956 C 377.12 49.97 c 6 63.63 84.09 c 0.0003 D 128.82 30.34 bc 86.23 21.51 bc 0.2490 E1 177.82 61.63 b 138.51 40.98 bc 0.3506 E2 4995.79 2747.51 a 9706.08 1912.27 a 0.1794 G 2941.07 539.34 a 1836.89 509.39 a 0.0971 Different letters indicate significant treatments

PAGE 48

48 Table 2 5. Mean ( SE) number of waveform event s per insect (NWEI) for different genders of Diaphorina citri feeding and under different light conditions Experiment 1: Gender Wa veform Female Male NWEI SE NWEI SE p value z 6.33 1.27 b 4.90 0.89 b 0.5331 n p 38.60 6.82 a 34.40 4.12 a 0.7065 C 34.93 7.08 a 33.86 4.44 a 0.8985 D 2.50 0.56 b 1.50 0.22 bc 0.1606 E1 3.16 0.83 b 2.16 0.40 bc 0.3927 E2 2.00 0.57 b 1.00 0.00 c 0.0283 G 2.00 0.40 b 2.10 0.45 bc 0.9680 Experiment 2: Light conditions Light Dark Waveform NWEI SE NWEI SE p value z 9.57 1.68 b 9.07 1.92 b 0.8589 n p 51 7.29 a 37.86 5.56 a 0.1575 C 46.46 6.88 a 33.86 5.81 a 0.1277 D 3.33 1.22 bc 2.44 0.62 c 0.5380 E1 4 1 bc 3.66 0.91 bc 0.7672 E2 1.4 0.24 c 1.66 0.33 c 0.6391 G 1.5 0.28 c 1.12 0.12 c 0.1877 Different letters indicat treatments

PAGE 49

49 Table 2 6. Mean ( SE) waveform duration per event per insect (WDEI) (s) for different genders of Diaphorina citri feeding and under different light conditions Experiment 1: Gender Waveform Female Male WDEI SE WDEI SE p value z 964.32 219.42 c 719.90 177.36 b 0.5799 n p 517.79 85.04 c 483.87 155.48 b 0.4161 C 908.95 185.76 bc 931.60 128.17 a 0.4457 D 60.51 1 0.46 d 66.57 14.45 c 0.9157 E1 45.96 7.22 d 107.66 41.10 c 0.4949 E2 3889.76 683.43 a 2090.43 453.29 a 0.0966 G 1789.45 350.48 ab 2034.97 325.73 a 0.9461 Experiment 2: Light conditions Waveform Light Dark WDEI SE WDEI SE p value z 1404.13 523.88 b 542.21 188.11 c 0.0278 n p 411.31 70.06 c 314.53 41.77 c 0.5911 C 545.92 183.3 c 928.81 175.13 b 0.0316 D 51.68 11.9 d 46.24 4.61 d 0.9970 E1 59.07 13.86 d 57.74 20.98 d 0.5600 E2 4683.1 2846.47 a 7942.05 1439.29 a 0.0418 G 2888.07 280.9 a 1820.95 516.66 b 0.1734 treatments

PAGE 50

50 Figure 2 1. Asian citrus psyllid EPG wavef orms on sweet orange plants. A) EPG waveform overview. B) Fragment sect ion of the waveform C. C) Fragment section of the wavefo rm D. D) Fragment section of the waveform E1. E) Fragment section of the waveform E2. F) Fragment section of the waveform G. Waveform G is not represented on the waveform overview (A) A Waveform C C Waveform D D Waveform E1 E Waveform E2 Waveform G F B C B D E

PAGE 51

51 Figu re 2 2 Asian citrus psyllid EPG non probing waveforms A) Waveforms np. B). Waveform z. B Waveform z Waveform np A

PAGE 52

52 Figure 2 3 Percentage of the total waveform duration (TWD) for female and male Diaphorina citri feeding. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Female Male G E2 E1 D C np z

PAGE 53

53 Figure 2 4 Percentage of the total waveform dura tion (TWD) for Diaphorina citri feeding under different light conditions 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Light Dark G E2 E1 D C np z

PAGE 54

54 CHAPTER 3 EFFECTS OF SOIL APPLIED IMIDACLOPRID ON ASIAN CITRUS PSYL LID ( DIAPHORINA CITRI KUWAYAMA) (HEMIPTERA: PSYLLIDA E) FEEDING BEHAVIOR Imidacloprid is a neonicotinoid insecticid e that causes insect mortality by contact and/or ingestion. When applied to the soil surface, imidacloprid is taken up by the plant roots and translocated to citrus leaves as a result of its excellent xylem mobility (Nauen et al. 1999, Sur and Stork 2003). As citrus plants produce new flush, imidacloprid present in the plant xylem moves into the new growth, thus providing an extended period of D. citri control. Under typical Florida growing conditions, this systemic activity can last eight weeks or longer ( Rogers et al. 2011). In contrast, foliar applications do not provide the same systemic activity, due to the poor p hloem mobility of imidacloprid. As a result, imidacloprid does not readily move into flush produced following a foliar application. Thus, the optimal method of imidacloprid application is to nonbearing trees by soil application. Because D. citri ingests primarily from phloem sieve elements and the HLB associated pathogen is a phloem limited bacterium, pathogen acquisition and inoculation probab ly occur during stylet activities that occur in the phloem. Acquisition must occur primarily during phloem ingestion; inoculation probably occurs during phloem salivation, because Las is circulative in the t re enter the plant in saliv a. Imidacloprid is most mobile in the xylem, from which it then migr ates into other plant tissues. It is not known whether acquisition and inoculation of Candidatus Liberibacter asiaticus (Las) by D. citri could occur before imidacloprid exposure, which wo uld subsequently cause cessation of phloem stylet activities. Therefore, a better understanding of the effects of imidacloprid on stylet penetration behaviors of D. citri should help to refine vector control strategies. Electrical penetration graph (EPG) monitors have been used to study the effects of insecticide applications on insect feeding behavior. Joost and Riley (2005), studying

PAGE 55

55 Frankliniella fusca (Hinds) and Frankliniella occidentalis (Pergande), observed that F. occidentalis probed plants more fr equently and for longer periods of time on imidacloprid treated plants compared to untreated plants. Their results suggest an increase in the inoculation of the tomato spotted wilt virus (TSWV) on imidacloprid treated plants, which is a persistent, circula tive virus. In contrast, when feeding on imidacloprid treated tomato plants, F. fusca exhibited a significant decrease in the number of probes per insect and probing duration when compared to untreated plants. In addition, Collar et al. (1997b) did not obs erve any significant differences in probing behavior of Myzus persicae (Sulzer) on imidacloprid treated versus untreated pepper plants. The first use of an EPG monitor to study the feeding behavior of a psyllid was performed with the pear psylla, Psylla p yricola Foerster. Ullman and Mclean (1988) observed salivation (S) and ingestion (I) for both nymphal and adul t pear psyllids. Waveforms from the AC monitor used in that study could not distinguish phloem ingestion versus xylem ingestion. However, histolog ical examination showed that both nymphs and adults ingest from all types of leaf cells; xylem, phloem and bundle sheath cells were found to be the preferred sites for ingestion. More recently, Bonani et al. (2010) used a Giga 8 DC monitor to identify and histologically correlate (define) the D. citri feeding waveforms, based in part on their similarity to the well known aphid waveforms. In the present study, we characterized the feeding behaviors of D. citri on citrus plants treated with a soil applicatio n of imidacloprid compared with untreated plants. The objective of our study was to determine whether the presence of imidacloprid within a plant disrupts feeding behaviors of D. citri particularly, those hypothesized by Bonani et al. (2010) to be respons ible for successful pathoge n acquisition and inoculation. The effects of imidacloprid on psyllid

PAGE 56

56 feeding were examined on both young (approximately 2 3 weeks post bud br eak in age) and mature leaves. While psyllids prefer to feed on young tender leaves, th ey can also be found feeding on mature leaves, especially in the winter time when young leaves are not av ailable (Catling 1970). The results of the current study should provide a better understanding of the effect of soil applied imidacloprid on transmissi on of Las. Material and Methods Plants and Insects Citrus aurantium L.) seedlings (25 30 cm tall) planted in one gallon pots using citrus potting mix (Fafard Citrus Mix, Fafard, Agawam, MA). Seedl ings were grown in a pathogen free greenhouse at 29 3 o C and 60 80% RH. All seedlings were planted and maintained identically, to minimize interplant variation. Adult female D. citri (10 15 d) used in experiments were obtained from a greenhouse colony f ree of Ca. Liberibacter asiaticus, reared on sour ora nges and sweet oranges ( Citrus s inensis (L.) Osbeck) at 29 3 C with a photoperiod of 12:12 (L:D) h. Prior to use in EPG recordings, psyllids (10 15 d old) were transferred to a rearing cage (61cm x 61 cm x 91cm, Bioquip, Rancho Domingues, CA) containing sour orange plants for a 48 h acclimation period. Of these psyllids held for acclimation, only female D. citri were then selected for use in feeding experiments. EPG Recording and Waveform Analysis Recor dings of D. citri feeding for 12 h under constant light conditions on imidacloprid treated and untreated plants were made using a Giga 8 monitor (Department of Entomology, Wageningen Agricultural University, the Netherlands) Setup of EPG recordings and wa veform characterizations were conducted as previously described in Chapter 2.

PAGE 57

57 Experimental Design Twenty days prior to EPG recordings, citrus plants were treated with imidacloprid (0.750 ml Admire Pro 4.6F) diluted in 250 ml of water applied to the soil s urface of each pot. Control plants were treated with tap water only. Experiment 1 D. citri f ee ding b ehavior on y oung l eaves Four weeks prior to initiation of the experiment a total of 20 plants were pruned to force production of young leaves. From thos e 20 plants, 10 plants were randomly chosen and treated with imidacloprid 20 days prior to the EPG recording. These plants were treated with imidacloprid at a rate of 0.75ml product (Admire Pro 4.6F) diluted in 250 ml of water applied as a drench to the so il surface of each pot. Control plants were treated with 250 ml of tap water. Only plants that had young leave s were used in this experiment. A total of seven plants containing young leaves were used for each treatment. The feeding behavior of two psyllids was individually recorded on young leaves located on opposite sides of each plant. Because imidacloprid concentration varies leaf by leaf, even within the same plant (Mendel et al. 2000), each psyllid EPG recording was considered a unique replicate, thus providing 14 replicates per treatment for analysis. Experiment 2 D. citri f e eding b ehavior on m ature l eaves A total of 10 plants were treated with imidacloprid using the same methods and rates described in the young leaf experiment above. Ten additiona l plants treated with tap water only served as controls. Two different psyllids were individually recorded feeding on mature leaves on opposite sides of each plant, providing a total of 20 replicates for each treatment. Statistical Analysis To compare D. c itri feeding behavior on imidacloprid treated and untreated plants, t he number of waveform events and their duration were analyzed between treatments using

PAGE 58

58 biologically non sequential parameters, as described by Backus et al. (2007) and Bonani et al. ( 2010 ). These parameters were grouped by cohort, insect, probe and event level The parameters analyzed at each of these levels were as previously described in Chapter 2. square test was performed to test the goodness of fit (PRO C GLIMMIX, SAS In stitute 2001). The waveform duration data were log transformed and the frequencies square root transformed before statistical analysis, to improve homo geneity and reduce variability. Data were analyzed by protected ANOVA (PROC GLIMMIX, SAS Institute 2001) with the least significant difference (LSD) test (LSMEANS, SAS Institute 2001) used for pairwise comparisons, to determine whether the waveform parameters analyzed were significantly different between the imidacloprid treated and untreated plants. Data wer e also compared in a similar manner among waveforms within treatments. Means were considered Results Experiment 1. Treatment E ffects on F eeding, Y oung L eaves. One hundred percent (PPW) of D. citri performed pathway/styl et penetration waveforms (waveform C) and non probing/walking activities (waveform np) on both imidacloprid treated and untreated plants On the untreated plants, 71.4 % (PPW) of D. citri penetrated the phloem, salivated and ingested (waveforms D, E1and E2, respectively) and only 35.7 % performed non walking/non probing activities (waveform z). On imi dacloprid treated plants, 42.9 % (PPW) of D. citri penetrate d the phloem (waveform D), 35.7 % salivated into the phl oem (waveform E1), and only 7.1 % ingested phloe m sap (waveform E2) (Table 3 1). Cohort level. D. citri had a total access period of 604,800 s, during which psyllids on the imidacloprid treated plants probed 100 times (TNP) and spent 120,556.32 s (TPD) with their stylets inserted in the leaf. For the majority (80. 1 %) of the access period, psyllids performed non

PAGE 59

59 probing activities such as walking, jumping off the leaf (waveform np), or they died from exposure to imidacloprid (waveform z). In contrast, psyllids on untreated plants spent most of their tim e (63.2 %) performing probing activities (TPD = 382,441.72 s) producing 390 (TNP) probes, nearly four times the TNP of D. citri on imidacloprid treated plants. The percentage of each waveform duration performed as part of TPD for both imidacloprid treated a nd untreated plants is represented by the total waveform duration (TWD) (Figure 3 1 ). TWD data clearly show higher percentages of waveform C being produced on untreated plants and higher percentages of waveform z on imidacloprid treated plants. Insect leve l. There was considerable variation among individual insects within the cohort for waveform duration by individual insects (WDi). This variation was observed for both treated and untreated plants, with highest variation on the imida cloprid treated plants ( Figure 3 2 ). The differences among the waveforms for each individual insect, as demonstrated by WDi between treated and untreated plants, was averaged to give the waveform duration per insect (WDI). Significant differences in WDI were found for non probing moving (np) ( F = 14.57; df = 1, 26), non probing motionless (z) ( F = 59.13; df = 1, 17), pathway (C) ( F = 18.47; df = 1, 26), phloem salivation (E1) ( F = 4.49; df = 1, 14), and xylem ingestion (G) ( F = 10.16; df = 1, 9) (Table 3 1). Durations of waveforms np, C, E1 and G were significantly longer on untreated plants, while duration of waveform z was significantly longer on imidacloprid treated plants (Table 3 1). There also was considerable variation among individual psyllids in terms of the number of prob es by insect (NPi) ranging from 3 45 probes on untreated plants, while the NPi on imidacloprid treated plants ranged from 3 17 probes. Thus, the mean number of probes per insect (NPI) was significantly higher for psyllids on untreated plants compared to im idacloprid treated plants ( F = 19.17; df = 1, 26; Table 3 2). Similar results were also observed for the

PAGE 60

60 duration of those probes (PDI), i.e. psyllids on untreated plants produced probes with significantly longer duration ( F = 32.19; df = 1, 26) than psyll ids on treated plants (Table 3 2). Analysis of time to first occurrence of waveform D (T1stD) showed that D. citri reached the phloem in an average of 303.9 min, while psyllids on imidacloprid treated plants took a significantly ( F = 4.60; df = 1, 14; P = 0.0414) shorter time, on average only 130.7 min. Probe level. Analysis of the waveform duration per probe (WDP) revealed a significant difference between treatments for waveforms C ( F = 3.97; df = 1, 388) and znp ( F = 13.85; df = 1, 406) (Table 3 3). In a ddition, the number of waveform events per probe (NWEP) was significantly greater on untreated plants for the phloem salivation phase waveform E1 ( F = 7.90; df = 1, 36) and significantly less for the znp waveform ( F = 26.53, df = 1, 406) (Table 3 4). NWEP for E2 was not significantly different between treatments, although about half the number of events per probe (one vs. two) was performed on treated vs. untreated plants. Event level. The ACP probing duration per event (PDE), unlike PDI at the insect level revealed a non significant difference between treated and untreated plants (Table 3 2). In contrast, significant differences were found in the number of waveform events per insect (NWEI) when imidacloprid treated and untreated plants were compared. Wavef orms np ( F = 19.19; df = 1 26 ), and C ( F = 18.09, df = 1, 26) (Table 3 5) occurred at a significantly higher frequency on untreated plants. NWEI for both E2 and E2 on treated plants were numerically one fourth the frequency of control plants. The waveform duration per event per insect (WDEI) was significantly longer for waveform G ( F = 6.18, df = 1, 9), and C ( F = 6.72, df = 1, 26) (Table 3 6) and significantly shorter for waveform z on untreated plants ( F = 4.09, df = 1, 17) compared with imidacloprid tre ated plants (Table 3 6).

PAGE 61

61 Summary of results. On imidacloprid treated plants, relative to untreated control plants, psyllids performed their typical feeding behaviors differently, in the following ways. When not probing, individual insects on treated plant s were motionless for longer periods of time even though insects performed the same number of standing events. In contrast, treated insects performed less walking on a per insect basis because walking events that occurred were the same length but were perf ormed less often. During stylet probing, the duration of pathway activities (waveform C) was shorter for insects on treated plants because each event performed was shorter. Though performed the same number of times per probe, pathway was performed in nume rically fewer probes, leading to pathway duration per probe that did not differ between probes. Overall, however, this led to significantly fewer events per insect. Durations of phloem contact (waveform D) were the same for all levels of study, therefore D represents a partly stereotypical behavior. However, there were significantly fewer contacts per probe on treated plants, cancelled out by numerically longer durations per probe. The duration per insect for phloem salivation (waveform E1) was shorter for insects on treated plants, because while each event was the same length, they were performed less often per probe, in numerically fewer probes, leading to shorter duration per probe and signific antly fewer events per insect. Phloem ingestion (waveform E2) duration per insect was statistically the same for each treatm ent, at all levels. However, numerically longer events were performed half as often per probe, in only one fourth the number of probes, leading to (numerically) twice the mean duration per probe but only one fourth the number of events per insect. In contrast, xylem ingestion (waveform G) was performed for shorter durations per insect on treated plants because each event was shorter, but performed the same number of times (once

PAGE 62

62 per probe), in nu merically fewer probes, leading to less time per probe and significantly fewer events per insect. Overall, insects on treated plants stood still for longer periods of time and walked less often. Insects on treated plants searched less often for vascular ti ssue, for shorter periods of time in fewer probes. Insects on imidacloprid treated plants also required less time to find phloem and salivate into it, during the fewer occasions they found it. However, once there, they salivated and ultimately ingested for the same duration each of those fewer times. In contrast, insects on treated plants found and ingested from xylem more briefly each time, though the same number of times (once) per probe, but in fewer probes. Experiment 2. Treatment E ffects on F eeding, M a ture L eaves. All D. citri exposed to untreated mature leaves performed stylet penetration of leaf tissues (waveform C), although only 40% (PPW) of the female D. citri reached the phloem and salivated (waveform D and E1), and only 25% ingested phloem sap ( waveform E2). In contrast, while 95% of D. citri on imidacloprid treated mature leaves performed sty let penetration only 25% (PPW) reached the phloem, 20% performed phloem salivation, and only 5% ingested phloem sap (waveform C, D, E1 and E2 respectively) (Table 3 1). Cohort level. From a total access period of 864,000 s, D. citri on the imidacloprid treated mature leaves probed 99 times (TNP) and spent 42,110.38 s (TPD) with their stylets inserted in the leaf tissues. For the remaining 95.1 % of the total access period, psyllids performed non probing activities (waveforms np and z). In contrast, D. citri on untreated plants spent 43. 1% of their total access period performing probing activities (TPD = 372,497 s), producing 268 probes (TNP). The percentage o f each waveform duration performed in TPD for both imidacloprid treated and untreated plants are represented by the total waveform duration (TWD) shown in Fig ure 3 1

PAGE 63

63 Insect level. Similar to the results from experiments with young leaf tissues, there was also considerable variation in waveform duration among individual insects. This was shown by a higher variation in waveform duration by insect (WDi) on imidacloprid treated co mpared to untreated plants (Figure 3 3 ). Analysis of the mean waveform duration p er insect (WDI) showed significantly longer durations for waveforms np ( F = 11.84; df = 1, 38), C ( F = 36.39; df = 1, 37), D ( F = 12.25; df = 1, 11), E1 ( F = 4.65; df = 1, 10), and G ( F = 6.74; df = 1, 13) (Table 3 1) on untreated plants. The number of prob es performed by insect (NPi) also indicated high variation among the insects; D. citri on untreated plants performed 2 28 probes while the ones on imidacloprid treated plants performed 1 15 probes. The significant difference between NPi on treated and untr eated plants was apparent in the number of probes per insect (NPI) ( F = 14.55; df = 1, 38;Table 3 2), in which untreated plants showed a higher number of probes than the imidacloprid treated plants. Additionally, PDI was significantly longer for psyllids o n untreated plants than on imidacloprid treated plants ( F = 40.9; df = 1, 37; Table 3 2). Analysis of time to first D (T1stD) indicated that D. citri reached the phloem of control plants in an average of 262.15 min and an average of 126.78 min on imidaclo prid treated plants; this difference was not significant. Probe level. The waveform duration per probe (WDP) was significantly longer for waveform C ( F = 19.56, df = 1, 366) on untreated plants but znp was significantly shorter ( F = 26.11, df = 1, 396; Tab le 3 3). In contrast, the number of waveform events per probe (NWEP) was not significantly different between treatments for any waveform analyzed (Table 3 4). Event level. Psyllid probing duration per event (PDE) indicated significant differences between t reated and untreated plants ( F = 9.43, df = 1, 616; Table 3 2), with duration per event longer on untreated plants. The number of waveform events per insect (NWEI) was significantly

PAGE 64

64 higher on untreated plants for waveforms np ( F = 12.40; df = 1, 38), and C ( F = 19.91, df = 1, 37) (Table 3 5). The waveform duration per event per insect (WDEI) was longer for waveforms C ( F = 9.24, df = 1, 37), D ( F = 5.90, df =1, 11) and G ( F = 5.11, df =1, 13) (Table 3 6) for D. citri feeding on untreated compared to imidacl oprid treated plants. Summary of results. Feeding behavior of D. citri differed on imidacloprid treated plants and control plants. When not probing, psyllids on treated plants stood still (z) for longer durations per insect. Even though the durations per insect were not significantly different, their event were two times longer than the event durations on the control plants, making this event even longer, because the frequency of this event per insect was the same between treated and untreated plants. Wave form analysis indicated that psyllids stood still on treated plants half as much as on untreated plants In contrast, on treated plants psyllids performed less walking (waveform np), with shorter duration events per insect and those events were also perform ed less often. When stylet probing, psyllids on treated plants performed shorter pathways activities (waveform C) for a shorter duration per insect because each event was shorter. Furthermore, those pathway activities were characterized by shorter event d uration and those events were less frequently performed per insect. In addition, pathway activity was less frequent per probe, and consequently pathway duration per probe was shorter and occurred as fewer events per insect. Durations of phloem contact (D) per insect were significantly shorter on the treated mature leaves than on young leaves because of shorter and less frequent phloem contact per insect. Although, the duration of phloem contact per probe was not significantly different between young and old leaves, there were significantly fewer contacts into the phloem per insect.

PAGE 65

65 As observed with the psyllids on treated young leaves, D. citri on the mature leaves also performed phloem salivation (E1) for a shorter duration per insect, with events of the s ame length as on control plants. Given that these events were the same length, they were performed less often per probe, in numerically fewer probes, leading to fewer probes that were shorter in duration. Phloem ingestion (E2) duration per insect was stati stically the same for each treatment, at all levels. Both treatments had similar duration of events. However, there was a noticeable trend for reduced phloem ingestion on treated plants which was 62 fold shorter in duration than th at on untreated plants. H owever only half the number of events per insect occurred on untreated than treated plants. In contrast, xylem ingestion (G) was performed for shorter durations per psyllid on treated plants than on untreated plants because each event was shorter and perfo rmed less often per insect. Overall, as observed with young leaves containing imidacloprid, psyllids stood still longer and walked less often on imidacloprid treated mature leaves. Psyllids searched less often for vascular tissue and for shorter durations of time in fewer probes on untreated than treated plants. Then, as insects in the young treated plants, psyllids on mature treated plants stood still for longer times and walked less often. Insects on treated plants searched less often for vascular tissue, for shorter times in fewer probes. While the presence of imidacloprid in plant tissues reduced the overall duration of phloem ingestion and salivation compared to untreated plants, those psyllids on treated plants that were able to locate and salivate int o the phloem did so in a shorter period of time. Psyllids on treated plants found and ingested from xylem less frequently and with fewer probes than on untreated plants. Discussion The objective of the current study was to examine the effects of soil appli ed systemic applications of imidacloprid on the feeding behavior of D. citri to quantify effects of this pesticide on the behaviors that mediate transmission of Las. Overall, the general feeding

PAGE 66

66 behaviors of D. citri were disrupted on imidacloprid treated plants as demonstrated by an increased duration and frequency of standing still, decreased walking and searching for probing sites, and reduced probing behaviors. Similar results were observed for F. fusca on imidacloprid treated pepper, mustard (Groves et al. 2001) and tomato (Joost and Riley 2005). While all D. citri tested on imidacloprid treated plants died during the course of the 12 h EPG recording due to intoxication, D. citri on treated young leaves died sooner on average (4 h) than on old leaves. Consequently, psyllids had an average feeding access period of 4 h before succumbing to the pesticide treatment. In this access period, a considerable percentage of the D. citri were able to reach the phloem (waveforms D, E1, and E2). Because the associat ed agent Candidatus Liberibacter asiaticus, is a phloem limited bacterium (Bove 2006), it is likely that bacterial acquisition and inoculation require phloem ingestion and salivation, respectively. Bonani et al. (2010), examining D. citri feeding on HLB a ffected citrus plants, observed that detectable Las acquisition occurred after 1 h of the D. citri ingestion waveform (E2). In our study, D. citri feeding on young leaves of plants treated with imidacloprid had an average ingestion period of 1 h (0.02 h fo r mature leaves). However, only 14% of psyllids exposed to young treated leaves produced waveform durations longer than 1 h, and from the total of six phloem contact events only two were longer than 1 h. In contrast, psyllids on mature control plants had a total of 42 phloem contacts, in which two waveforms were also longer than 1 h. Consequently, a few psyllids might have a slightly higher chance of acquiring Las during longer, although rare, phloem ingestion (E2) events on an infected tree. On the other h and, the probability of inoculation is probably reduced because both numbers and durations of phloem salivation (E1) events were reduced on treated plants. Although not significantly different due to small sample size, we suggest that these changes in phlo em

PAGE 67

67 salivation and ingestion could be biologically important for Las transmission on imidacloprid treated plants. Thus, if those insects were feeding on a Las positive plant, they would have a reduced, but probably biologically relevant, chance of acquiring Las. However, they would not be able to transmit, due to the pathogen latency period in the psyllid. Studies suggest a latency period of 24 h to 25 d (Xu et al. 1988, Roistacher 1991), thus as our results show, psyllids would be dead before they were able to transmit. In addition, D. citri are poor vectors of Las when they acquire the bacteria as adults Pelz Stelinski et al. ( 2010). Compared to feeding on mature leaves, D. citri on young citrus leaves performed numerically more probes which were also long er in duration on both imidacloprid treated and untreated plants. This result is probably related to leaf tenderness, which may influence the ease of successful stylet penetration. Bonani (2009) showed that during a D. citri access period of 5 h, 50% of th e psyllids on young leaves reached the phloem and ingested while only 15% of the psyllids reached the phloem on the mature leaves reached the phloem. It was postulated that this was due to higher lignifications of the cell walls of mature leaves (Bonani 20 09). The ability of some psyllids to perform extensive bouts of phloem associated behaviors on the young leaves of imidacloprid treated plants may be due to uneven distribution of imidacloprid within a leaf. Mendel et al. (2000) found uneven concentrations of imidacloprid throughout the parenchyma and low concentrations close to the vascular bundles; however, younger shoots had higher imidacloprid concentrations. Similar results were also found by El Hamady et al. (2008) when examining the distribution of r adioactive imidacloprid in cotton. In addition, El Hamady et al. (2008) found higher concentrations of imidacloprid nearest to the edges of young leaves with concentrations at much lower levels near the leaf veins. Because psyllids often feed closer to lea f veins, those insights could explain the ability of some D. citri to successfully ingest from phloem

PAGE 68

68 on imidacloprid treated plants. Likewise, the high variability seen in waveform duration by individual insects could be a result of the location of the ph loem sieve elements chosen by those individual psyllids, which was not accounted for in this study. Implications for managing the spread of HLB. The current results demonstrate that the benefits of soil applied imidacloprid applications extend beyond reduc ing psyllid populations. Prior to causing mortality, imidacloprid application can potentially disrupt psyllid feeding behaviors such that successful pathogen acquisition and inoculation are both less likely to occur. Our findings also support other studies that have shown imidacloprid treated plants to be repellent to D. citri During the course of our EPG recordings, we observed that following a short test probe, D. citri often jumped off imidacloprid treated plants, whereas psyllids on untreated plants ra rely left the plant. This was indicated by the high percentage of time spent by D. citri in non probing behaviors for both young and mature leaves of imidacloprid treated plants. Similarly, Liu and Trumble (2004), examining the feeding behavior of B. cocke relli on tomato, found a significantly higher jumping frequency and a longer off of plant duration for imidacloprid treated compared to untreated tomato plants. Boina et al. (2009) also reported sublethal effects of imidacloprid resulting in repellence of D. citri when encountering citrus with low levels of imidacloprid present. Further studies are needed to determine the duration of this feeding disruption provided by soil applications of imidacloprid to help guide growers in developing effective psyllid c ontrol programs for protection of young trees from HLB disease.

PAGE 69

69 Table 3 1. Mean ( SE) waveform duration per insect (WDI) (s) and the proportion of individuals that produced a waveform type (PPW) for Diaphorina citri feeding on young and mature leaf tis sues of imidacloprid treated and untreated plants. Experiment 1: Young Leaves Waveform Untreated c ontrol Imidacloprid WDI SE PPW WDI SE PPW p value z 6333.82 982.71 5/14 32043.71 2781.96 14/14 <0.0001 n p 15375.28 2807.67 1 4/14 6371.35 1101.15 14/14 0.0008 C 22205.09 2154.62 14/14 5529.63 1505.27 14/14 0.0002 D 747.57 180.88 10/14 544.34 226.05 6/14 0.2337 E1 1378.41 272.65 10/14 308.01 77.70 6/14 0.0525 E2 3269.93 930.40 10/14 3601.42 2061 .69 5/14 0.2295 G 1764.50 290.79 10/14 94.08 N/A 1/14 0.0110 Experiment 2: Mature leaves Waveform Untretead c ontrol Imidacloprid WDI SE PPW WDI SE PPW p value z 13915.58 3769.54 6/20 32390.61 2815.86 20/20 0.2800 n p 22217.0 5 2822.68 20/20 10627.13 2710.06 20/20 0.0014 C 13287.12 2143.33 20/20 2182.41 600.03 19/20 < 0.0001 D 669.77 284.14 8/20 64.59 21.03 5/20 0.0050 E1 250.65 99.78 8/20 46.51 8.05 4/20 0.0565 E2 7598.13 3520.96 5/20 76.18 N/A 1/20 0.0722 G 4381.29 1219.61 14/20 59.54 N/A 1/20 0.0222

PAGE 70

70 Table 3 2. Mean ( SE) probe duration per insect (PDI) (s), number of probes per insect (NPI), and probe duration per event (PDE) (s) for Diaphorina citri feeding on young and ma ture leaf tissues of imidacloprid treated and untreated plants. Experiment 1: Young Leaves Untreated c ontrol Imidacloprid PDI SE PDI SE p value 27317.27 2555.20 7187.86 2083.68 0.0003 NPI SE NPI SE p value 21.14 2.76 8.00 1.1 9 0.0002 PDE SE PDE SE p value 498.09 42.04 488.49 73.26 0.6301 Experiment 2: Mature leaves Untreated c ontrol Imidacloprid PDI SE PDI SE p value 18270.98 2376.14 2216.34 608.17 <0.0001 NPI SE NPI SE p value 1 4.05 1.96 5.80 0.91 0.0005 PDE SE PDE SE p value 750.88 74.07 345.17 44.54 0.0022

PAGE 71

71 Table 3 3. Mean ( SE) waveform duration per probe (WDP) (s) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and un treated plants. Experiment 1: Young Leaves Waveform Untreated c ontrol Imidacloprid WDP SE WDP SE p value znp 834.20 118.26 4726.59 955.99 0.0002 C 1075.96 119.07 973.41 194.26 0.0469 D 219.95 21.94 272.17 96.22 0 .2577 E1 492.29 106.76 184.81 30.92 0.1791 E2 1634.96 530.41 3001.19 1520.88 0.6543 G 1260.35 212.84 94.08 N/A 0.0610 Experiment 2: Mature leaves Waveform Untreated c ontrol Imidacloprid WDP SE WDP SE p value znp 1871 .75 264.61 7038.63 1107.05 <0.0001 C 987.84 125.49 418.75 67.57 <0.0001 D 336.58 140.68 53.82 19.99 0.1538 E1 123.63 41.45 46.51 8.05 0.8918 E2 4748.83 1803.21 76.18 N/A 0.2707 G 2667.44 807.48 59.54 N/A 0.197 7

PAGE 72

72 Table 3 4. Mean ( SE) number of waveform event per probe (NWEP) and number of probes by waveform (NPw) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants Experiment 1: Young Leaves Waveform Untreated c ontrol Imidacloprid NWEP SE NPw NWEP SE NPw p value z np 1.04 0.01 296 1.44 0.12 112 <0.0001 C 1.48 0.09 290 1.36 0.13 100 0.5417 D 4.44 0.42 29 3.25 0.81 12 0.1164 E1 5.50 0.61 28 2.40 0.54 10 0.0080 E2 2.10 0.33 20 1.00 0.00 5 0.0860 G 1.00 0.00 14 1.00 N/A 1 N/A Experiment 2: Mature leaves Waveform Untreated c ontrol Imidacloprid NWEP SE NPw NWEP SE NPw p value z np 1.00 0 282 1.00 0 116 N/A C 1.26 0.07 268 1.07 0.03 99 0.1028 D 3.38 0.87 16 1.33 0.21 6 0.1740 E1 3.75 0.99 16 1.50 0.29 4 0.2809 E2 1.63 0.32 8 1.00 N/A 1 0.5406 G 1.35 0.15 23 1.00 N/A 1 0.6382

PAGE 73

73 Table 3 5. Mean ( SE) number of waveform event s per inse ct (NWEI) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants Experiment 1: Young Leaves Waveform Untretaed c ontrol Imidacloprid NWEI SE NWEI SE p value z 1.40 0.25 3.36 0.77 0.1588 np 21.43 2.75 8.21 1.25 0.0002 C 30.36 4.52 9.71 1.76 0.0002 D 13.30 3.35 6.50 2.25 0.1703 E1 15.40 4.49 4.00 1.03 0.0787 E2 4.20 1.60 1.00 N/A 0.2158 G 1.40 0.22 1.00 N/A 0.5987 Experiment 2: Mat ure leaves Waveform Untreated control Imidacloprid NWEI SE NWEI SE p value z 4.50 1.50 3.65 0.52 0.5018 np 14.65 2.11 6.50 0.99 0.0011 C 16.9 2.29 5.57 0.99 < 0.0001 D 6.75 2.63 1.60 0.24 0.1568 E1 7.5 2.96 1.50 0.29 0.1923 E2 2.60 0.81 1.00 N/A 0.4664 G 2.214 0.43 1.00 N/A 0.4833

PAGE 74

74 Table 3 6. Mean ( SE) waveform duration per event per insect (WDEI) (s) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid treated and untreated plants Experiment 1: Young Leaves Waveform Untreated control Imidacloprid WDEI SE WDEI SE p value Z 6333.82 982.71 32043.71 2781.36 0.0591 N p 1596.30 878.21 803.10 115.45 0.7411 C 985.46 188.93 486.88 87.46 0.0154 D 56.93 4.55 70.01 25.60 0.7770 E1 129.67 38.58 78.01 8.45 0.8140 E2 1145.72 255.27 2585.39 1396.73 0.8712 G 1438.90 250.49 94.08 N/A 0.0346 Experiment 2: Mature leaves Waveform Untreated control Imidacloprid WDEI SE WDEI SE p value z 7358.75 4491.82 14084.04 2953.50 0.2973 np 2345.87 470.63 1909.68 558.24 0.3545 C 838.11 127.95 419.64 67.18 0.0043 D 110.63 22.85 41.15 9.85 0.0335 E1 37.19 7.81 31.56 1.95 0.9808 E2 3330.71 1192.10 76.18 N/A 0.1099 G 2383.16 670.29 59.54 N/A 0.0410

PAGE 75

75 Figure 3 1 P ercentage of the total waveform duration (TWD) for Diaphorina citri feeding on young and mature leaf tissues of imidacloprid t reated and untreated plants 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Control Imidacloprid Young leaves 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Control Imidacloprid Mature leaves z G E2 E1 D C np

PAGE 76

76 Figure 3 2 Percentage of the waveform duration by insect (WDi) for Diaphorina citri feeding on young leaf tissues of imidacloprid treated and untreated plants 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Waveform duration (%) Individual psyllids feeding on untreated young leaves np z G E2 E1 D C 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Waveform duration (%) Individual psyllids feeding on imidacloprid treated young leaves np z G E2 E1 D C

PAGE 77

77 Figure 3 3 Percentage of the waveforms duration by in sect (WDi) for Diaphorina citri feeding on mature leaf tissues of imidacloprid treated and untreated plants 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Waveform Duration (%) Individual psyllids feeding on imidacloprid treated mature leaves np z G E2 E1 D C 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Waveform Duration (%) Individual psyllids feeding on untreated mature leaves np z G E2 E1 D C

PAGE 78

78 CHAPTER 4 EFFECTS OF ALDICARB ON ASIAN CITRUS PSYL LID (HEMIPTERA: PSYL LIDAE) FEEDING BEHAVIOR AND THEIR POTENTIAL IMPACTS ON TRANSMISSION Soil app lied systemic insecticides are commonly used in Florida citrus production for control of plant feeding pests Due to their long r esidual period of activity within treated plants, these systemic insecticides typically provide control of target pests for lon ger durations of time compared to foliar insecticide sprays Currently, three soil applied systemic insecticides are registered for use in Florida citrus : aldicarb (Temik 15G, Bayer CropSciences, Research Triangle Park, N.C.), imidacloprid (Admire Pro 4 .6F, Bayer CropSciences, Research Triangle Park, N.C.) and thiamethoxam (Platinum 75SG, Syngenta CropProtection, Inc., Greensboro, N.C.) (Rogers et al, 201 1). Of these products, aldicarb is the only systemic product for which the product label permits its use at high enough quantities of active ingredient to provide control of target pests on large (> 9 ft height) bearing citrus trees. For this reason, aldicarb has been te pests on both young and mature citrus trees. Aldicarb is a systemic carbamate insecticide that is applied as a granular formulation incorporated into the soil surface surrounding the plant. The maximum use rate of aldicarb in Florida citrus is 36.99 kg/ ha S oon after being applied, aldicarb granules dissolve in the presence of soil moisture and the product is then readily absorbed by the plant root system for translocation throughout the plant (Ware 1994). Because of concerns of groundwater contamination past use of aldicarb has been restricted to the Florida dry season (November 15th through April 30th) More recently however, the worldwide registration of aldicarb is being cancelled with use in Florida citrus no longer permitted after December 31, 2011 Since the discovery of HLB in Florida in 2005, aldicarb has been viewed as an important tool for controlling D. citri and managing the spread of HLB (Qureshi and Stansly 2008).

PAGE 79

79 However, the true value of use of this product over the past six years for c ontrolling the spread of HLB is not well understood. Thus, the objective of the current investigation was to evaluate the eff ects of aldicarb ( Temik 15G) applications on D. citri feeding behavior and determine the po tential impacts on Las transmission. M aterial s and Methods Plants and Insects Citrus s inensis (L.) Osbeck) seedlings (15 20 cm tall) planted in 120ml tubes containing mix Fafard Citrus potting Mix (Fafard, Agawam, MA), grown in a path ogen free greenhouse at 29 3 C and 60 80% RH. All seedlings were planted and maintained identically to minimize interplant variation. F emale psyllids (1 0 15d post emergence) were obtained from a greenhouse colony ( free of Las ) reared on citrus at 29 3 C and 12:12 (L:D) h photoperiod EPG Recording and Waveform Analysis Recordings of D. citri feeding for 12 h under constant light conditions on aldicarb treated and untreated plants were made using a Giga 8 monitor (Department of Entomology, Wageningen A gricultural University, the Netherlands) Setup of EPG recordings and waveform characterizations were conducted as previously described in Chapter 2 Effects of Aldicarb on D. citri Feeding B ehavior Twenty days prior to EPG recordings, a subset of plants was treated with 0.0046g of aldicarb ( Temik 15G ) (field rate) applied to the soil surface of each pot. After insecticide application, the pots were covered with a layer (1cm) of po tting soil to avoid insecticide spill during watering.

PAGE 80

80 A total of 15 plant s were treated with aldicarb, and 15 additional plan ts were used as controls. One young lea f was randomly selected f rom each p lant on which a single psyllid was placed and feeding behavior then recorded. Statistical Analysis To compare D. citri feeding be havior on aldicarb treated and untreated plants, t he number of waveform events and their duration were analyzed between treatments using biologically non sequential parameters, as described by Backus et al. (2007) and Bonani et al. ( 2010). These parameters were grouped by cohort, insect, probe and event level The parameters analyzed at each of these levels were as previously described in Chapter 2. square test was performed to test the goodness of fit (PRO C GLIMMIX, SAS Institute 2001). The waveform duration data were log transformed and the frequencies square root transformed before statistical analysis, to improve homo geneity and reduce variability. Data were analyzed by protected ANOVA (PROC GLIMMIX, SAS Institute 2001) with the least sign ificant difference (LSD) test (LSMEANS, SAS Institute 2001) used for pairwise comparisons, to determine whether the waveform parameters analyzed were significantly different between the aldicarb treated and untreated plants. Means were considered significa Confirmation of Aldicarb in Treated P lants After recordings of D. citri feeding behavior were completed, leaf samples were collected for analysis to confirm the presence of aldicarb in the plant tissue. Since the plants used in th e recordings were small and a minimum of 5 g of leaf tissue was needed for proper analysis, leaves from 5 plants of each treatment w as combined to obtain a sufficient quantity of leaf material for analysis. Leaf samples were sent to the Waters Agricultural Laboratory (Camilla, GA) for analysis of aldicarb content using HPLC/UV chromatography with detection at 205nm.

PAGE 81

81 Results One hundred percent (PPW) of D. citri performed pathway/stylet penetration waveforms (Waveform C) and non probing/walking activities (W aveform np) on both aldicarb treated and untreated plants On the untreated plants, 71.4 % (PPW) of D. citri penetrated the ph loem and salivated (Waveforms D and E1, respectively) and 6 4.3 % ingested phloem (Waveform E2) Approximately seventy nine percent ( 78.6%) of the psyllids performed xylem activities (Waveform G) on untreated pl ants and only 7.1% spend time with non walking/non probing activities (Waveform z). On aldicarb treated plants, 78. 6 % (PPW) of D. citri penetrated the phloem (Waveform D) and sal iva ted into it (Waveform E1), 57.1 % ingested phloem sap (Waveform E2) 85.7% ingested xylem sap (Waveform G) and 64.3 % performed non walking/non probing activities (Table 4 1). Cohort level. D. citri had a total access period of 604,800 s, during which ps yllids on untreated plants performed 217 (TNP) probes and spent 405,759.3 s (TPD) with their mouth parts inserted in to the leaf. P syllids on aldicarb treated plants probed 195 times (TNP) in which the total duration of the probes (TPD) was 337,361.09 s. Th e percentage of each waveform duration performed in TPD are represented by the total waveform duration (TWD) shown i n Figure 4 1. There was o nly a small difference between D. citri feeding on treated and untreated plants. Those differences were investigate d in more detail at the insect level. Insect level. The probe duration per insect (PDI) on untreated plants averaged 28 982.81 s, while on aldicarb treated plants it averaged 24 097.22 s (Table 4 2). The number of probes per insects (NPI) was also observed and no significant difference was found between D. citri feeding on aldicarb treated and untreated plants ( F = 0.16 ; df = 1, 26) (Table 4 2).

PAGE 82

82 ANOVA indicated no significant difference between treatments ( F = 0.3469 ; df = 1, 26). T he wav eform duration per formed per psyllid (WDI) was also not significant ly different (Table 4 1). Analysis of time to first D (T1stD) showed that D. citri on untreated plants required o n average 238.70 min to reached the phloem, while psyllids o n aldicarb treated plants took a n average 252.23 min, this difference was not significantly different ( F = 0.03 ; df = 1, 19 ; p = 0.8721). Probe level. Similar to that observed at the insect level, the number of waveform events per probe (NWEP) did not differ significantly between aldicarb treated and untreated plants for any of the waveforms (Table 4 3). In c ontrast, waveform duration per probe (WDP), differed significant ly for waveforms znp ( F = 11.49; df = 1, 421 ; Table 4 4), but not for any of the other waveforms. Event level. D. citri probing duration per event (PDE), unlike PDI and NPI at the insect level, revealed a significant difference between treated and untreated plants ( F = 7.76; df = 1,1076 ; Table 4 2). In ad d ition, waveform duration per event (WDE) was significant different fo r waveforms np ( F = 7.42; df = 1,435), E1 ( F = 4.92 ; df = 1,204), and E2 ( F = 4.61 ; df = 1, 47), in which non probing activities were longer on control pla nts than aldicarb treated plants (Table 4 7). Also, phloem activities were longer on aldicarb treated plants than untreated plants (Table 4 7). However the number of waveforms events perfo rmed per insect (NWEI) was not significantly different between D. citri on aldicarb treated and untreated plant s (Table 4 5). In addition, waveform duration per event p er insect (WDEI), as NWEI, were not significant ly differen t between treatments (Table 4 6).

PAGE 83

83 Confirmation of Aldicarb in Treated P lants Because there was no apparent difference in the parameters analyzed for D. citri feeding on aldicarb treated compared t o untreated plants analysis of aldicarb content within the leaves was conducted to confirm the presence of aldicarb in treated plants The mean (SE) content of aldicarb per 5 g of plant tissue for aldicarb treated plants was 0.94 0.62 p pm, while aldica rb was not detected in untreated plants. Summary of results. Psyllids on aldicarb treated plants performed their typical feeding behaviors in a manner similar to psyllids on untreated control plants. The differences between the two tre atments were very sm all, with no significant difference for most parameters analyzed. However, those small differences ind icate some toxic effect of aldicarb on the psyllids tested. For example, when not probing, psyllids on treated plants stood still (z) two times longer per insect than those on untreated plants. However, their duration per event per insect was shorter due to a higher number of stan ding events that was performed 2x more frequently In addition, insects on aldicarb treated plants performed less walking (np) pe r ins ect because events durations were smal ler but they were performed alm ost in the same frequency as on untreated plants. However, when looking into the general non probing (walking + standing still) (znp) behavior, there was significantly longer non pro bing behavior per probe on aldicarb treated plants than untreated plants s ince this behavior represents the combinati on of the standing still and walki ng behavior. When stylet probing, insects on treated plants made slight ly shorter pa thway activities (C) per insect. This difference is due to a smaller number of events per insect and the same number of s tylet probing per probe. Durations of phloem contact (D) on aldicarb treated plants were half the duration o f that for the control plants This was probably due to the short duration of phloem contacts performed per insect (10x smaller than control) Even thou gh those parameters were

PAGE 84

84 shorter there was no sig nificant difference between phloem contact duration per probe or per event on treated and untreated pl ants. Sal ivation into the phloem (E1) was slight ly shorter in duration per insect on aldicarb treated compared to untreated plants; this was due to the small difference in the duration of event s per insect However, phloem salivation was longer per event o n treated than untreated plants indicating that even though phloem salivation events were performed less often they were longer on aldicarb treated plants Phloem ingestion (E2) duration per insect was smaller, because each event was shorter in duration per insect. In addition, the number of phloem ingestion events per probe was smaller per insect, in a numerically fewer phloem ingestions. Even though phloem salivation durations were not significant different per probe, the phloem ingestions were slightly higher on aldicarb treated plants, which lead to significantly longer phloem ingestions per event. While phloem salivation events were less frequently per formed per insect on aldicarb treated plants than untreated plants, those events were longer on the t reated plants than on untreated plants. In contrast, xylem ingestion (G) was performed for longer duration s per insect on aldicarb treated plants than untreated due to slight ly longer events per insect. Overall except for longer standing still even ts and walking less frequently on aldicarb treated compared to untreated plants, there was no other differences in the parameters analyzed. Discussion The objective of the current study was to examine the effects of soil applied systemic applications of aldicarb on the feeding behavior of D. citri to determine the potential effects of this insecticide on the behaviors that mediate transmission of Las The overall analysis of D. citri feeding behavior indicated no effect of this insecticide on psyllid probing beha vior. However, differences were found when examining the event level, where waveform duration per event (WDE) was higher during the phloem activities of salivation and ingestion on aldicarb treated

PAGE 85

85 plants than controls. Those results showed that D. citri i ngested phloem sap and salivated into phloem for longer duration s per event on aldicarb treated plants when compared to psyllids on untreated plants. Ragab (1981) suggested that this insecticide might have a direct effect on mineral metabolism of cotton pl ants, mainly the ones that are involve d in nitrogen and phosphor ou s metabolism. Balayannis (1983) showed that aldicarb a pplications increased the leaf content of wa t er soluble sugars in tobacco and the concentration of iron, manganese and zinc in the leave s and roots. A lso aldicarb decrease d lea f nitrate reductase activity and the concentrations of nicotine and crude protein. P lant nutrition has a direct e ffect on insect behavior For example, studies on Psylla pyricola Foerster showed a higher production of honeydew when feeding on pear leaves with very low nitrogen content, indicating compensatory feeding effect due to the low nutrition of the leaves (Pfeiffer and Burts 1984). Consequently, aldicarb may have caused increased phloem ingestion by D. citri f eeding on aldicarb treated plants Previous studies have shown an increase of brix, yield, and peel color i n citrus fruit sampled from trees treated with aldicarb along with an increase i n calcium and potassium content i n the citrus leaves (Wheaton et al. 1985 ). The nitrogen content was not measured however and increases in nit rogen content might also affect D. citri feeding behavior (Tsagkarakis and Rogers, unpublished). After 12 h of feeding on aldicarb treated plants, none of the D. citri w as found dead F or this reason citrus plants we re sent f o r residual ana lysis to confirm the presence of the insecticide in tested plants. R esidual analysis confirmed existence of aldicarb within the plants. T he actual residual concentration necessary to cause psyllid to xicity is unknown. However studies looking into the efficiency of aldicarb treatments for the control of Tryoza erytreae (South African citrus psyllid) showed poor efficiency in egg and nymph al control even when applying 227 g of

PAGE 86

86 aldicarb per tree with tre es averaging 23.2 m 2 canopy (Catling, 1969) However, nymphs were controlled using 907.2 g per tree (de Villiers, 1969 cited by Catling 1969). Such d osages have not been permitted in Florida due to concerns of groundwater contamination. In contrast Quresh i and Stansly (2008) showed that aldicarb successfully reduce d D. citri population s when applied 2 3 months prior to spring flushes at recommended rates. However when they cage d adult psyllids for 25 d on the aldicarb treated plants, mortality was below 50% following 25 days (Qureshi and Stansly, 2008). During the 12 h access period, D. citri probed for 6.7h on aldicarb treated plants, during which time they were able to reach the phloem and salivate (Wavefom D and E1) for an average of 0.26 h, and ingest (Waveform E2) for more than 3 h. Those results differ from the investigations examining soil applied imidacloprid (Chapter 3) D. citri feeding behavior on imidacloprid treated citrus resulted in mortality within a 6 h access period Also, during this acc ess period phloem ingestion averaged 1.0 h for psyllids on young leaves and it did not occur on mature citrus leaves ( Chapter 3 ). Implication for managing the spread of HLB. The current results indicate that aldicarb has negligible effects on the feeding behaviors of D. citri However, there was a slight indication that aldicarb treatment may enhance Las transmission given the longer salivation and phloem ingestion events on aldicarb treated plants. Bonani et al. (2010), observed Las acquisit ion 1 h after initiation of the ingestion waveform (Waveform E2) by D. citri The efficiency of Las acquisition by adult D. citri is low (Pelz Stelinski et al. 2010) and there is a latency period anywhere from 24 h to 25 d (Xu et al. 1988, Roistacher 1991). However, sin ce aldicarb applications can result in less than 50% of adults being controlled under field conditions

PAGE 87

87 (Qureshi and Stansly 2008), this further supports our findings that D. citri can likely inoculate aldicarb treated plants with Las. The current study un derscores the importance of EPG studies for vector transmission dynamics in order to improve management of insect vectored plant diseases. Although aldicarb usage will be banned in citrus groves as of December 31, 2011, investigations such as this one demo nstrate that unlike imidacloprid, not all soil applied systemic insecticides can disrupt transmission of Las by D. citri

PAGE 88

88 Table 4 1. Mean ( SE) waveform duration per insect (WDI) (s) and the proportion of individuals that produced a waveform type (PPW) for Diaphorina citri feeding on aldicarb treated and untreated citrus plants. Waveform Untreated control Aldicarb WDI SE PPW WDI SE PPW p value z 5750.40 N/A 1 /15 10273.44 2507.34 9 /15 0.8771 n p 16040.72 2991.02 14 /15 14201.18 241 2.40 14 /15 0.9049 C 15519.65 2786.05 14 /15 14407.25 2083.64 14 /15 0.6542 D 998.22 526.03 10 /15 442.15 132.11 11 /15 0.3747 E1 622.69 145.02 10 /15 500.85 103.65 11 /15 0.4605 E2 15595.10 5568.77 9 /15 11321.59 4204.89 8 /15 0.56 27 G 2901.74 411.41 11 /15 3228.84 368.71 12 /15 0.4627

PAGE 89

89 Table 4 2. Mean ( SE) probe duration per insect (PDI) (s), mean number of probes per insect (NPI), and probe duration per event (PDE) (s) for Diaphorina citri feeding on aldicarb treated and untreated citrus plants. Untreated control Aldicarb PDI SE PDI SE p value 28982.81 3297.99 24097.22 3473.27 0.3469 NPI SE NPI SE p value 15.86 3.12 14.36 2.00 0.9804 PDE SE PDE SE p value 698.38 119.62 677.43 98.11 0. 0054

PAGE 90

90 Table 4 3. Mean ( SE) number of waveforms event per probe (NWEP) and number of probes by waveform (NPw) for Diaphorina citri feeding on aldicarb treated and untreated plants. Waveform Untreated control Aldicarb NWEP SE NPw NWEP SE NPw p value z np 1.01 0.01 222 1.01 0.01 201 N / A C 1.47 0.11 217 1.48 0.10 195 0.7792 D 3.63 0.67 24 3.33 0.44 24 0.9902 E1 4.92 0.85 24 3.71 0.46 24 0.3488 E2 1.89 0.31 19 1.27 0.19 11 0.1613 G 1.00 0.00 20 1.13 0 .07 23 0.0984

PAGE 91

91 Table 4 4. Mean ( SE) waveform duration per probe (WDP) (s) for Diaphorina citri feeding on aldicarb treated and untreated plants. Waveform Untreated control Aldicarb WDP SE WDP SE p value z np 1037.48 174.12 1469.20 221 .95 0.0008 C 994.74 135.09 1013.69 144.13 0.0864 D 413.94 225.64 202.65 29.56 0.8264 E1 260.11 45.00 229.55 37.70 0.5609 E2 7388.84 3063.89 8233.89 3384.07 0.3469 G 1666.76 208.30 1684.61 219.85 0.7665

PAGE 92

92 Table 4 5. Mean ( SE) number of waveforms events per insect (NWEI) for Diaphorina citri feeding on aldicarb treated and untreated citrus plants. Waveform Untreated control Aldicarb NWEI SE NWEI SE p value z 1.00 N/A 2.33 0.53 0.3787 n p 15.93 3.13 15.29 2.10 0.8495 C 22.93 4.72 20.71 3.33 0.9784 D 8.80 3.42 7.27 2.07 0.8328 E1 11.80 4.96 8.00 2.08 0.6359 E2 3.89 1.81 1.75 0.49 0.2967 G 1.81 0.26 2.17 0.40 0.5224

PAGE 93

93 Table 4 6. Mean ( SE) waveform dur ation event per insect (WDEI) (s) for Diaphorina citri feeding on aldicarb treated and untreated plants. Waveform Untreated control Aldicarb WDEI SE WDEI SE p value z 5750.40 N/A 5474.08 1642.25 0.6771 np 1664.81 509.49 1066.00 20 1.14 0.6760 C 847.72 145.90 784.42 110.45 0.9596 D 602.18 549.32 57.36 7.85 0.2978 E1 120.36 35.88 98.40 30.54 0.6406 E2 8916.69 4030.00 8527.49 3764.27 0.9152 G 1738.52 174.83 1776.21 257.55 0.9834

PAGE 94

94 Table 4 7. M ean ( SE) waveform duration per event (WDE) (s) for Diaphorina citri feeding on aldicarb treated and untreated plants. Waveform Untreated control Aldicarb WDE SE WDE SE p value z 5750.40 N/A 4402.90 1229.04 0.4568 np 1007.04 169.29 929.04 96.40 0.0067 C 676.87 78.92 695.02 69.31 0.2039 D 113.43 62.49 60.80 3.73 0.7364 E1 52.77 6.74 61.90 9.72 0.0277 E2 4010.17 1752.20 6469.48 2787.80 0.0370 G 1595.96 218.89 1440.43 182.44 0.8607

PAGE 95

95 Figur e 4 1. Percentage of the total waveform duration (TWD) for Diaphorina citri feeding on aldicarb treated and untreated citrus plants 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Control Aldicarb TWD np z G E2 E1 D C

PAGE 96

96 CHAPTER 5 EFFECTS OF FIVE DIFF ERENT FOLIAR APPLIED INSECTICIDES ON ASIAN CITRUS PSYLLID ( DIAPHORINA CITRI ) FEEDING BEHAVIO R AND THEIR POSSIBLE IMPLICATION FOR LAS TRANSMISSION One of the components of HLB management programs used by Florida citrus growers is t he use of broad spectrum foliar insecticide applications to reduce psyllid populations. The rationale for this approa ch is that by decreasing psyllid populations the number of psyllids inoculating healthy citrus plants will also decrease. Consequently the number of insecticide sprays in citrus has greatly increased since 2005. I nsecticides commonly used by growers in Fl orida to control psyllids belong to several classes with different modes of action These include acetylcholinesterase inhibitors such as chlorpyrifos, the acetylcholine receptor stimulator imidacloprid, sodium channel modulators including fenpropathrin, a nd the lipid biosyn thesis inhibitor spirotetremat S ome of these insecticide products require ingestion by the insect or absorption onto the cuticle following contact in order to induce toxic effects, which are not always immediate. Because D. citri feeds primarily from phloem sieve elements and the HLB associated agent is a phloem limited bacterium, bacterial acquisition and inoculation probably occur during stylet activities within the phloem. Pathogen a cquisition likely occurs during phloem ingestion wh ile inoculation probably occurs during phloem salivation Las is circulative with and mus t re enter the plant in saliva Given that some insecticides increase vector activities and thus enhanc e pathogen spread (Lowery and Boiteau 1988, Roberts et al. 1993), a better understanding of the effects of insecticides on the stylet penetration behaviors of D. citri should help refine vector control strategies. Electrical penetration graph (EPG) monitors have been used to study the effects of ins ecticide applications on insect feeding behavior. Topically applied and injected pymetrozine

PAGE 97

97 inhibited stylet insertion of Aphis fabae Scopoli, Macrosiphum euphorbia (Thomas), Myzus persicae (Sulzer) and Aphis gossypii Glover on broad beans, potato plants, Chinese cabbage, and cucumber respectively. High doses of pymetrozine caused irreversibly feeding disruption while low doses only affected aphid feeding temporary. Pymetrozine showed no toxic effect on aphids, in contrast, it showed to cause aphids death due to starvation ( Harrewij 1997). In addition, EPG studies on Frankliniella fusca (Hinds) and F occidentalis on tomato plants showed that F. occidentalis probed more frequently and for longer periods of time on imidacloprid treated plants than on untreate d plants suggest ing an increase in the inoculation of the tomato spotted wilt virus (TSWV) on imidacloprid treated plants. In contrast, F. fusca when feeding on imidacloprid treated tomato plants exhibited a significant decrease in the number of prob es per insect and probing duration when compared to untreated plants (Joost and Riley 2005). In the present study, we characterize d the feeding behaviors of D. citri on citrus plants treated with five different foliar applied insecticides : chlorpyrifos (Lorsban 4E TM DowAgroSciences, Indianapolis, IN ), fenpropathrin (Danitol 2.4 EC TM Valent U.S.A. Corporation, Walnut Creek, CA ), imidacloprid (Provado 1.6F TM Bayer CropScience, Research Triangle Park, NC ), spinetoram (Delegate WG TM DowAgroSciences, Indianapolis, IN ), and spirotetramat (Movento 240SC TM Bayer CropScience, Research Triangle Park, NC ) and compared them with untreated plants. The objective wa s to determine whether the presence of th e se insecticides on a plant is able to disrupt feeding behaviors of D citri especially th ose hypothesized to be responsible for successful pathogen acquisition and inoculation ( Bonani et al. 2010).

PAGE 98

98 Material s and Methods Plants and Insects Citrus s inensis (L.) O sbeck) seedlings (20 25 cm tall) planted i n 120ml tubes with Fafard Citrus potting Mix (Fafard, Agawam, MA). Seedlings were grown in a pathogen free greenhouse at 29 3 C and 60 80% RH. All seedlings were planted and maintained identically to minimize in terplant variation. Adult female D. citri (10 15 d) used in experiments obtained from a greenhouse colony free of Ca. Liberibacter asiaticus, reared on citrus at 29 3 C with a photoperiod of 12:12 (L:D) h. Prior to use in EPG recordings, female and mal e psyllids were transferred to a rearing cage 48 h acclimation period. Of these psyllids held for acclimation, only female D. citri were selected for use in feeding experiments. EPG Recording and Waveform Analysis Recordings of D. citri feeding for 6 h under constant light conditions on insecticide treated and untreated plants were made using a Giga 8 monitor (Department of Entomology, Wageningen Agricultural Universi ty, the Netherlands) Setup of EPG recordings and waveform characterizations were conducted as previously described in Chapter 3. Effects of F oliar I nsecticides on D. citri F eeding B ehavior Examination of the effect of foliar applied insecticides on D. ci tri feeding behavior were conducted as five separate experiments in which each experiment evaluated one of the following insecticides: 1) chlorpyrifos (Lorsban 4E), 2) fenpropathrin (Danitol 2.4EC), 3) imidacloprid (Provado 1.6F), 4) spinetoram (Delegate W G), and 5) spirotetramat (Movento 240SC) (Table 5 1).

PAGE 99

99 Twenty four hours prior to EPG recordings, the insecticides were applied to using a 500 ml plastic spray bottle until insecticide runoff from the leaves was achieved. All insectic ides were applied at concentrations which correspond to the recommended field rate listed in the 2011 Florida Citrus Pest Management Guide (Rogers et al 2011). s were used for each treatment, for which a single psyllid was r ecorded on feeding on each plant for 12 h At the end of each recordin gthose insects that had wiring problems resulting in poor waveform recording quality were discarded from the analysis. Statistical A nalysis Feeding behavior of D. citri on plants treated with one of the five selected insecticides was compared to untreated plants as previously described in Chapter 2. However, parameters were only analyzed at the cohort and insect level for each insecticide vs. untreated comparison. square tes t was performed to test the goodness of fit (PR OC GLIMMIX, SAS Institute 2001) and waveform duration parameter data were log transformed before statistical analysis to improve homogeneity of variances Data were analyzed by ANOVA (PROC GLIMMIX, SAS Institu te 2001) with the least significant difference (LSD) test (LSMEANS, SAS Institute 2001) used to determine if the waveform parameters analyzed were significantly different between the insecticide treated and untreated plants. Means were considered significa Results Chlorpyrifos One hundred percent (PPW) of D. citri performed pathway/stylet penetration waveforms (Waveform C) and non probing/walking activities (Waveform np) on both chlorpyrifos treated and untreated plants. On the un treated plants, 64. 3 % (PPW) of D. citri penetrated the phloem,

PAGE 100

100 51.2% salivated and 42. 9 % ingested (Waveforms D, E1 and E2, respectively), 64. 3 % ingested xylem sap (Waveform G), and 78. 6 % performed non walking/non probing activities (Waveform z). On chlorpy rifos treated plants none of the insects penetrated into the phloem (Waveform D), and consequently none of the insects salivated or ingested phloem sap (Waveforms E1and E2), 14. 3 % ingested xylem sap (Waveform G), and 100% performed non walking/non probing activities (Waveform z) (Table 5 1). At the cohort level, D. citri had a total access period of 604,800s, during which psyllids on the chlorpyrifos treated plants probed 221 times (TNP) and spent 44,406.195s (TPD ) with their stylets inserted into the leaf For majority (92. 7 %) of the access period D. citri performed non probing activities such as walking, jumping off the leaf, or they died from chlorpyrifos exposure. In contrast, D. citri on untreated plants spent most of their time (65. 2 %) performing probi ng activities (TPD = 394,031.59s) producing 643 (TNP) probes. Th e differences observed at the cohort level can also be seen when analyzed at the insect level T he number of probes per insect (NPI) was significantly higher for D. citri on untreated plants c ompared to chlorpyrifos treated plants ( F = 8.66; df = 1, 26; P = 0.0067 ; Table 5 2). Similar results were also observed for the duration of those probes (PDI) P syllids on untreated plants produced significantly longer probes ( F = 48.02; df = 1, 26 ; P < 0 .0001) than on treated plants (Table 5 3). For waveform duration per insect (WDI), there were significant differences found between treatments for waveforms C ( F = 39.04; df = 1, 26 ; P < 0.0001 ), np ( F = 17.5; df = 1, 26 ; P = 0.0003), and z ( F = 24.66; df = 1, 23 ; P < 0.0001) ; waveform s np and C were significantly longer in duration on untreated than plants treated whereas waveform z was significantly longer on chlorpyrifos treated plants (Table 5 1). Statistical analyses on D, E1, and E2 are not available because D. citri did not perform those behaviors prior to death on

PAGE 101

101 chlorpyrifos treated plants (Table 5 1). Similar results were also obtained with waveform duration per event per insect (WDEI) ; the duration of waveform C ( F = 12.56; df = 1, 26 ; P = 0.0015 ) was longer on untreated plants whereas the duration of waveform z was longer on chlorpyrifos treated plants ( F = 35.46; df = 1, 23 ; P = <0.0001 ) Analysis o f the number of waveforms events per insect (NWEI), indicated a significantly higher occurrence o f the waveforms np ( F = 7.96; df = 1, 26 ; P = 0.0090) and C ( F = 10.06 ; df = 1, 26 ; P = 0.0039) on untreated compared to chlorpyrifos treated plants. Fenpropathrin All D. citri performed pathway/stylet penetration waveforms (Waveform C) and non probing/walk ing activities (Waveform np) on both fenpropathrin treated and untreated plants. On the untreated plants, 38. 5 % (PPW) of D. citri penetrated and salivated in to the phloem (Waveform D and E1), 23. 1 % ingested phloem sap (Waveform E2), 61.5% ingested xylem sa p (Waveform E2), and 76.9% performed non walking/non probing activities (Waveform z). On fenpropathrin treated plants none of the psyllids penetrated into the phloem and consequently none salivated or ingested phloem sap (Waveform D, E1 and E2 respective ly) Also none of the D. citri performed Waveform G, and 100% performed non walking/non probing activities on fenpropathrin treated plants (Waveform z) (Table 5 4). At the cohort level, D. citri had a total access period of 280,800s, during which psyllids on the fenpropathrin treated plants probed 25 times (TNP) and spent 864.74s (TPD) with their stylets inserted in to the leaf. During the majority (99. 7 %) of the access period psyllids performed non probing activities such as walking, jumping off the leaf, or they died from fenpropathrin exposure. In contrast, psyllids on untreated plants spent most of their time (63. 2 %) performing probing activities (TPD = 177,392 s) producing 125 (TNP) probes.

PAGE 102

102 A t the insect level the NPI was significantly higher for D. c itri on untreated plants compared to fenpropathrin treated plants ( F = 27.4; df = 1, 26; P < 0.0001; Table 5 2). Similar results were also shown for the duration of those probes (PDI) ; psyllids on untreated plants produced longer probes ( F = 224.07; df = 1 20 ; P < 0.0001 ) than psyllids on treated plants (Table 5 3). For the waveform duration per insect (WDI), significant differences between treatments were found for waveform s C ( F = 154.98; df = 1, 20 ; P < 0.0001) np ( F = 55.66; df = 1, 26 ; P < 0.0001) a nd z ( F = 36.35; df = 1, 23 P < 0.0001). W aveform np and C were significantly longer in duration on untreated compared to treated plants whereas waveform z was significantly longer in duration on fenpropathrin treated plants (Table 5 4). D. citri did not p erform waveforms D, E1, E2 and G, since they were intoxicated prior to the performance of those behaviors. Analysis on the waveform duration events per insect (WDEI), indicated a significantly longer duration o f waveform C ( F = 60.11; df = 1, 20 ; P < 0.000 1) on untreated plants compared to fenpropathrin treated plants I n contra st waveform z was significantly longer ( F = 32.82; df =1, 23 ; P < 0.0001) on fenpropathrin treated compared to untreated plants (Table 5 4). In addition, ANOVA showed a significantl y higher frequency of waveform events per insect (NWEI) for waveforms np ( F = 28.33; df = 1, 26 ; P = < 0.0001) and C ( F = 24.27; df =1, 20 ; P < 0.0001) on fenpropathrin treated plants compared to untreated plants (Table 4). Imidacloprid One hundred percent of the D. citri on untreated plants performed pathway/stylet penetration waveforms (Waveform C), non probing/walking activities (Waveform np), 20% (PPW) penetrated and salivated in to the phloem (Waveforms D and E1, respectively), 6. 7 % ingested phloem sap (Waveform E2), 26. 7 % ingested xylem sap, and 66.67% performed non walking/non probing activities (Waveform z). On imidacloprid treated plants 100% of the D. citri performed Waveform np, 86. 7 % performed Waveform C, and none of the psyllids

PAGE 103

103 penetrated into t he phloem, salivated, ingested phloem sap, and ingested xylem sap (Waveform D, E1, E2, G respectively), and 100% performed non walking/non probing activities (Waveform z) on imidacloprid treated plants (Table 5 5). At the cohort level, D. citri had a total access period of 280,800 s, during which psyllids on the imidacloprid treated plants probed 76 times (TNP) and the majority of their time (96. 5 %) was spent in non probing activities D. citri spent 9954.22s (TPD) with their mouth parts inserted in to the l eaf tissues. In contrast, psyllids on untreated plants spent most of their time (64 %) performing probing activities (TPD = 179,828.88 s) producing 253 (TNP) probes. The differences observed at the cohort level can also be seen when the insects we re analyz ed at the insect level. The number of probes per insect (NPI) was significantly higher for D. citri on untreated plants compared to imidacloprid treated plants ( F = 19.37; df = 1, 28; P = 0.0001; Table 5 2). Similar results were also observed for the durat ion of those probes (PDI) P syllids on untreated plants produced probes that were significant ly longer in duration ( F = 43.99; df = 1, 26 ; P < 0.0001) than psyllids on treated plants (Table 5 3). For waveform duration per insect (WDI), significant differen ces were observed for waveform s C ( F = 40.34; df = 1, 26 ; P < 0.0001) np ( F = 12; df = 1, 26 ; P = 0. 0016 ) and z ( F = 32.86; df = 1, 23 ; P < 0.0001). Waveforms np and C were significantly longe r in duration on untreated compared with treated plants and wa veform z was significantly longer on imidacloprid treated than untreated plants (Table 5 5). Waveforms D, E1, E2, and G could not be statistically compared due to the low number of psyllids that performed those waveforms on imidacloprid treated plants. For WDEI, waveforms C ( F = 22.87; df = 1, 26 ; P < 0.0001) and z ( F = 28.37; df =1, 23 ; P < 0.0001) were significantly different; waveform C was longer in duration on untreated plants whereas waveform z was longer on imidacloprid treated plants. T he number of waveforms events per

PAGE 104

104 insect occurred at a significantly higher frequency for waveforms np ( F = 17.53; df = 1, 28 ; P = 0.0003 ) and C ( F = 18.50; df =1, 26 ; P = 0.0002) for D. citri on untreated plants compared to imidacloprid treated plants. Spinetoram One hundred percent (PPW) of D. citri performed pathway/stylet penetration waveforms (Waveform C) and non probing/walking activities (Waveform np) on both spinetoram treated and untreated plant s. On the untreated plants, 33. 3% (PPW) of D. citri penetrated the phloem, salivated, and ingested (Waveforms D E1and E2, respectively), 66.7 % ingested xylem sap (Waveform G), and 93.3 % performed non walking/non probing activities (Waveform z). On spinetoram treated plants 13% of the psyllids penetrated into the phloem, salivated and ingested phloem sap (Waveform D, E1, and E2, respectively), 6. 7% ingested xylem sap (Waveform G), and all (100%) D. citri performed non walking/non probing activities (Waveform z) (Table 5 6). At the cohort level, D. citri had a total access period of 604,800s, during which psyllids on the spinetoram treated plants probed 301 times (TNP) and spent 65,167.41s (TPD) with their stylets inserted in to the leaf. For the majority (89.2%) of the access period D. citri performed non probing activitie s such as walking, jumping off the leaf, or they died from spinetoram exposure. In contrast, D. citri on untreated plants spent most of their time (59.1%) performing probing activities (TPD = 357,680.49s) producing 455 (TNP) probes. A t the insect level s i gnificant difference were found for probe duration per insect (PDI), ( F = 43.99; df = 1, 28; P < 0.0001; Table 5 3). However, the number of probes per insect (NPI) did not differ between spinetoram treated and untreated plants ( F = 1.78; df = 1, 28; P = 0.1 933; Table 5 2). For waveform duration per insect (WDI), significant differences were found for waveform C ( F = 39.67; df = 1, 28 ; P < 0.0001), np ( F = 6.92; df = 1, 28 ; P = 0.0137), G ( F = 14.80 ; df = 1, 9 ; P = 0.0039), and z ( F = 34.84; df = 1, 27 ; P < 0. 0001 ); waveform s C, np and G

PAGE 105

105 were significantly longer in duration on untreated plants and waveform z was significantly longer in duration on spinetoram treated plants (Table 5 6). Waveform duration event per insect (WDEI), showed that waveform C ( F = 27.2 4; df = 1, 28 ; P < 0.0001 ) and G ( F = 33.04; df = 1, 9 ; P = 0.0003) were significantly longer on untreated plants than spinetoram treated plants. WDEI for waveform z ( F = 31.99; df = 1, 9 ; P < 0.0001) was significantly longer in duration on sprinetora m tre ated compared to untreated plants. T he number of waveforms events per insect (NWEI), was not significant ly different between spinoteran treated and untreated plants. Spirotetramat One hundred percent (PPW) of D. citri performed pathway/stylet penetration w aveforms (Waveform C) and non probing/walking activities (Waveform np) on both spirotetramat treated and untreated plants. On the untreated plants, 33. 3% (PPW) of D. citri penetrated the phloem and salivated (Waveforms D and E1, respectively) and 26. 7% in gested phloem (Waveform E2) Sixty percent of the psyllids performed xylem activities (Waveform G) on untreated plants and 86. 7% spend their time performing non walking/non probing activities (Waveform z). On spirotetramat treated plants, 40% (PPW) of D. c itri penetrated the phloem (Waveform D), salivated (Waveform E1), and ingested phloem sap (Waveform E2) Also on spirotetramat treated plants 66.7 % ingested xylem sap and 73.3% performed non walking/non probing activities (Waveform z) (Table 5 7). At the cohort level D. citri had a total access period of 604,800s, during which psyllids on untreated plants performed 588 (TNP) probes and spent 401,039s (TPD) with their mouth parts inserted in to the leaf. O n spirotetramat treated plants psyllids probed 531 times (TNP) and the total duration of probes (TPD) was 391,820s. At the insect level, analyses o f the number of probes per insect (NPI) and the probe duration per insect (PDI) indicated no significant difference between spirotetramat treated and untreated plants. Also, the waveform duration

PAGE 106

106 (WDI) was not significant ly different for waveform C ( F = 0.36 ; df = 1, 28 ): P = 0.5555 ), w aveform D ( F = 0.44 ; df = 1, 9 ); P =0.5248 ), w aveform E1 ( F = 1.00 ; df = 1, 9 ); P = 0.3433 ), w aveform E2 ) ( F = 0.11 ; df = 1, 8 ; P = 0.7524 ), w aveform G ( F = 1.55 ; df = 1, 17 ; P = 0.2300 ), w aveform z ( F = 2.79 ; df = 1, 22 ; P = 0.1088 ), w aveform np ( F = 0.44 ; df = 1, 28 ; P = 0.5138) between treated and untreated plants (Table 5 7). S imilarly no significant differences were found for t he waveform duration event per insect (WDEI) and number of waveforms events per insect (NWEI), with one exception where WDEI for waveform G ( F = 5.74; df = 1, 22 ; P = 0.0255), was significantly s horter in duration on untreated plants than spirotetramat tre ated plants (Table 5 7) Summary of Results Figure 5 1 show s the stereotypical feeding behavior of D. citri on plants treated with one of the five insecticides examined in this study. D. citri on spirotetramat treated plants performed normal feeding when compared to control plants. During the 12 h recording, psyllids on spirotetramat treated plants did no t differ for any of the parameters analyzed and performed phloem penetration, salivation and ingestion normally Chlo r pyrifos, f enpropathrin and imidaclo prid reduced feeding and increased non probing activities (Figure 5 1) These insecticides caused a quick knockdown effect and induced psyllid mortality prior to phloem contact The general feeding behavior of psyllids on spinetoram treated plants was red uced, however, psyllids were still able to reach the phloem, salivate and ingest phloem sap prior to intoxication. However, pathway/stylet activities were abnormal on spinetoram treated plants and fewer numbers of D. citri reached the phloem on this treatm ent compared with the control ( F igure 5 2).

PAGE 107

107 Discussion The goal of this study was to understand what effects foliar insecticide applications have on the transmission of Las by D. citri The results showed that there are differences between insecticides wit h different modes of action in terms of the ability to disrupt the feeding behaviors associated with pathogen transmission. Chlorpyrifos, fenpropathrin and imidacloprid all reduced D. citri phloem feeding behaviors whereas spinetoram and spirotetramat did not prevent phloem feeding behaviors. C hlorpyrifos, fenpropathrin, imidacloprid and spinetoram mainly act as a contact insecticides and act directly on the insect nervous system (acetylcholinesterase inhibitors, sodium channel modulators, acetylcholine rec eptor stimulator, and disruption of nicotinic/gamma amino butyric acid (GABA) gated chloride channels, respectively) S pirotetramat has limited contact activity and is mainly effective following digestion (Nauen et al. 2008). Spirotetramat is a tetra m ic ac id that results in lipid biosynthesis inhibition, acting Our results indicated that D. citri exposed to chlorpyrifos treated plants required on average 4.02h for 100% mo rtality and none of the psyllids were able to reach the phloem. Chlorpyrifos (785 g AI/ha) caused 90% mortality of Cacopsylla melanoneura (Frster), vector of the apple proliferation disease pathogen when overwintering adults a re exposed for 1 day, while 100% mortality occurs 3 days after treatment (Baldessari et al. 20 10 ). Additionally, field experiments in citrus showed that chlorpyrifos treatments were very effective in the control of the D. citri when applied during the Florida winter months (Qureshi and Stansly, 2010). Fenpropathrin treated plants caused a quick knockdown of D. citri Psyllids took in average 0.62 h to die with a total probing time of 0.24 h. This short probing duration was also observed for susceptible Myzus persicae (Sulzer) ; aphids exposed to etofenprox treated plants, had a total probing time of 0.27 h (Jo et al 2009). In addition, the sharpshooter Homalodisca

PAGE 108

108 coagulata (Say), vector of oleander leaf scorch ( Xylella fastidiosa ), was killed within an average of 4 hours. Quantificat ion of honeydew secretions, indicated a significant reduction of probing and settling on fenpropatrin treated than untreated plants. Transmission of the pathogen for these sharpshooters was reduced by 50% (Bethke et al. 2001). D. citri exposed to foliar a pplied imidacloprid treated plants were killed within an average of 1.41 h, and their feeding activities were extremely reduced when compared to the untreated plants. In addition, none of the psyllids were able to reach the phloem. Similar results were obt ained with Frankliniella fusca (Hinds) on tomato plants treated with soil applied imidacloprid (Joost and Riley 2005) The number and duration of probes were reduced when compared with untreated plants. However results obtained with Frankliniella occidenta lis (Pergrande) were different compared with the current study (Joost and Riley 2005) This thrips exhibited longer and more frequent probes when in contact with imidacloprid treated tomato plants than controls. Psyllids feeding behavior on foliar applied imidacloprid w as also different than those obtained with soil applied imidacloprid (Chapter 3) Foliar applied imidacloprid caused faster knockdown, than soil applied imidacloprid ( Chapter 3). Foliar applied imidacloprid completely prevent behaviors associ ated with Las transmission, while similar probing was observed with the systemic soil application (Chap ter 3). This is likely because psyllids must initiate feeding to obtain the active ingredient in the case of the soil applied formulation. It took longer (7.3 h) for mortality to occur on spinetoram treated plants compared to the other insecticide evaluated (except spirotetramat). In additional to the long feeding access period, a small percentage of the D. citri were able to reach the phloem and perform p hloem penetration, salivation and ingestion (Waveform D, E1, and E2 respectively ) on spinetoram treated plants suggesting that Las transmission could occur despite presence of recently applied residues. In addition to those

PAGE 109

109 measurements, it was observed s tylet penetrations (C) were not normal on spinetoram treated plants as compared to the controls. Waveform C did not have a pattern and instead showing long frequencies followed by shorter ones, the waveforms appeared chaotic. Consequently, spinetoram treat ed plants had some anti feeding effects on psyllids. In contrast, feeding behavior of D. citri was not affected on spirotetramat treated plants with similar amounts of phloem salivation and ingestion occurring on treated and untreated plants. Th e se results were similar to our findings for D. citri feeding on aldicarb treated plants (Chapter 4).

PAGE 110

110 Table 5 1. Mean ( SE) waveform duration per insect (WDI) (s) waveform duration event per insect (s) and number of waveform event s per insect (NWEI) for Diapho rina citri feeding on chlo r pyrifos treated and untreated citrus plants. Waveform Treatments Parameters WDI SE PPW WDEI SE NWEI SE np Untreated control 9860.62 1175.79 14/14 315.12 46.45 ns 10.23 1.41 Chlo r pyrifos 4556.51 648.48 14/14 265.24 37.27 2.73 0.43 C Untreated control 20298.24 2637.48 14/14 707.24 123.93 11.92 1.49 Chlo r pyrifos 2971.49 864.15 14/14 403.22 243.07 2.78 0.55 D Untreated control 222.97 63.63 n/a 9/14 58.27 2.86 n/a 4.20 0. 80 n/a Chlo r pyrifos 0/14 E 1 Untreated control 393.68 126.93 n/a 8/14 69.72 16.17 n/a 4.80 0.86 n/a Chlo r pyrifos 0/14 E 2 Untreated control 12850.59 3367.56 n/a 6/14 4769.72 1376.24 n/a 1.75 0.48 n/a Chlo r pyr ifos 0/14 G Untreated control 3066.55 647.42 ns 9/14 2160.55 390.73 ns 1.50 0.27 n/a Chlo r pyrifos 1402.70 122.27 2/14 1402.70 122.27 z Untreated control 8411.21 2485.60 11/14 1127.13 277.65 2.00 0.47 ns Chlo r pyrifos 37014.68 1273.64 14/14 9016.63 1532.42 1.60 0.34 *Significant difference from the untreated control ns Non significant difference from the untreated control P > 0.05

PAGE 111

111 Table 5 2. Mean ( SE) number of probes per insect for Diaphorina citri on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, treated and untreated citrus p lants. Treatment N PI SE p value Chlorpyrifos 16.50 2.43 0.0067 Untreated control 35.64 6.66 Fenpropathrin 2.67 0.49 <0.0001 Untreated control 10.00 1.40 Imidacloprid 6.07 1.59 0.0001 Untreated control 2.42 1.59 Spinetoram 21.00 2.77 0.1933 Untreated control 30.67 6.70 Spirotetramat 27.13 4.63 0.2377 Untreated control 34.20 4.41

PAGE 112

112 Table 5 3. Mean ( SE) probe duration per insect (PDI) (s) for Diaphorina citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, tr eated and untreated control plants. Treatment PDI SE p value Chlorpyrifos 3171.87 861.02 <0.0001 Untreated control 28145.11 2610.06 Fenpropathrin 96.08 38.92 <0.0001 Untreated control 13645.56 1652.54 Imidacloprid 765.71 242.88 <0. 0001 Untreated control 11988.59 1690.97 Spinetoram 4344.49 1182.35 <0.0001 Untreated control 23845.37 2415.16 Spirotetramat 26121.39 3121.59 0.6751 Untreated control 26735.96 2637.55

PAGE 113

113 Table 5 4. Mean ( SE) waveform duration per in sect (WDI) (s) waveform duration event per insect (s) and number of waveform event s per insect (NWEI) for Diaphorina citri on feeding fenpropathrin treated and untreated citrus plants. Waveform Treatments Parameters WDI SE PPW WDEI SE NWEI SE np Untreated control 6050.48 992.77 13/13 677.84 92.97 ns 10.23 1.41 Fenpropathrin 986.51 139.38 15/15 526.67 109.33 2.73 0.43 C Untreated control 9178.22 1318.21 13/13 840.77 105.01 11.92 1.49 Fenpropathrin 96.08 38 .92 9/15 49.43 20.91 2.78 0.55 D Untreated control 300.13 61.88 n/a 5/13 75.38 13.29 n/a 4.20 0.80 n/a Fenpropathrin 0/15 E 1 Untreated control 287.44 62.50 n/a 5/13 59.88 7.91 n/a 4.80 0.86 n/a Fenpropathrin 0/1 5 E 2 Untreated control 6327.93 2464.02 n/a 3/13 4045.22 1964.84 n/a 1.75 0.48 n/a Fenpropathrin 0/15 G Untreated control 3728.24 698.45 n/a 8/13 2837.43 733.54 n/a 1.50 0.27 n/a Fenpropathrin 0/15 z Untreated control 3012.28 1086.52 10/13 2553.64 1136.81 2.00 0.47 ns Fenpropathrin 19401.05 208.44 15/15 16518.67 1544.85 1.60 0.34 *Significant difference from the untreated control ns Non significant difference from the u ntreated control P > 0.05

PAGE 114

114 Table 5 5. Mean ( SE) waveform duration per insect (WDI) (s) waveform duration event per insect (s) and number of waveform event s per insect (NWEI) for Diaphorina citri feeding on imidacloprid treated and untreated citrus pl ants. Waveform Treatments Parameters WDI SE PPW WDEI SE NWEI SE np Untreated control 11377.44 1883.16 15/15 627.38 109.35 ns 18.87 2.57 Imidacloprid 2815.08 492.39 15/15 528.69 119.75 7.33 7.33 C Untreated control 9432.35 1485.20 15/15 638.56 94.03 17.93 2.55 Imidacloprid 765.71 242.88 13/15 150.07 56.33 5.85 1.74 D Untreated control 112.39 12.67 n/a 3/15 93.59 22.50 n/a 1.33 0.33 n/a Imidacloprid 0/15 E 1 Untreated control 95 .52 35.94 n/a 3/15 85.23 41.78 n/a 1.33 0.33 n/a Imidacloprid 0/15 E 2 Untreated control 22761.76 n/a 1/15 22761.76 n/a 1.00 n/a Imidacloprid 0/15 G Untreated control 3730.18 1134.17 n/a 4/15 3078.55 12 77.87 n/a 1.50 0.50 n/a Imidacloprid 0/15 z Untreated control 4311.73 1210.06 10/15 1612.30 465.41 3.60 1.08 ns Imidacloprid 21521.43 481.05 15/15 9743.81 1892.93 3.27 0.44 *Significant difference from the untreate d control ns Non significant difference from the untreated control P > 0.05

PAGE 115

115 Table 5 6. Mean ( SE) waveform duration per insect (WDI) (s) waveform duration event per insect (s) and number of waveform event s per insect (NWEI) for Diaphorina citri feedin g on spinetoram treated and untreated citrus plants. Waveform Treatments Parameters WDI SE PPW WDEI SE NWEI SE np Untreated control 16890.22 2072.11 15 617.14 91.11 ns 33.47 6.67 ns Spinetoram 9398.50 824.47 15 468.81 69.87 23. 67 2.60 C Untreated control 17953.47 2046.91 15 749.75 100.58 31.93 6.56 ns Spinetoram 3625.26 797.77 15 182.43 33.60 20.40 2.76 D Untreated control 224.21 59.85 ns 5 84.71 16.76 ns 3.00 0.94 ns Spinetoram 109.84 68.89 2 65.16 24.21 1.50 0.50 E 1 Untreated control 306.25 95.43 ns 5 94.39 18.13 ns 3.40 0.93 ns Spinetoram 321.36 162.80 2 100.16 20.88 3.00 1.00 E 2 Untreated control 13687.99 5499.34 ns 5 10913.17 4293.62 ns 1.20 0.20 ns Spinet oram 4896.88 4661.68 2 2507.24 2272.04 1.50 0.50 G Untreated control 1728.63 296.40 10 1288.72 137.69 1.30 0.15 ns Spinetoram 161.12 1 161.12 1.00 z Untreated control 4566.52 1246.12 14 1171.13 345.35 4.36 1.05 ns Spinetoram 31254.27 1621.34 15 7118.70 1084.35 6.00 1.01 *Significant difference from the untreated control ns Non significant difference from the untreated control P > 0.05

PAGE 116

116 Table 5 7. Mean ( SE) waveform duration per insect (WDI ) (s) proportion of insects performing a specific waveform (PPW), waveform duration event per insect (s) (WDEI) and number of waveform event s per insect (NWEI) for Diaphorina citri feeding on spirotetramat treated and untreated citrus plants. Waveform Tr eatments Parameters WDI SE PPW WDEI SE NWEI SE np Untreated control 14784.33 2006.10 ns 15/15 464.42 84.64 ns 37.47 5.13 ns Spirotetramat 12906.29 1867.91 15/15 567.14 132.17 29.47 4.68 C Untreated control 20687.67 2847.53 ns 15/15 912.06 376.43 ns 35.87 4.68 ns Spirotetramat 19038.67 2627.90 15/15 818.28 152.28 27.73 4.66 D Untreated control 237.73 71.68 ns 5/15 93.23 11.36 ns 2.40 0.51 ns Spirotetramat 283.71 57.43 6/15 83.48 7.63 3.33 0.61 E 1 Untreated control 275.54 89.11 ns 5/15 99.44 20.57 ns 2.60 0.75 ns Spirotetramat 387.84 77.11 6/15 89.65 17.26 4.50 0.85 E 2 Untreated control 12041.82 5630.75 ns 4/15 8638.85 2973.73 ns 1.25 0.25 ns Spirotetramat 11611.99 5 371.13 6/15 9613.43 5108.85 1.67 0.33 G Untreated control 4441.93 952.70 ns 9/15 2921.30 934.90 1.78 0.22 ns Spirotetramat 3249.88 690.09 10/15 2512.75 542.87 1.40 0.31 z Untreated control 4015.24 1545.90 ns 13/15 883.68 349 .42 ns 6.69 3.42 ns Spirotetramat 8144.86 1775.13 11/15 2149.25 411.73 4.09 0.89 *Significant difference from the untreated control ns Non significant difference from the untreated control P > 0.05

PAGE 117

117 Figure 5 1. Representative Asian citrus psyllid EPG waveforms on sweet orange plants treated with : A) Untreated control ; B) chlo r pyrifos; C) fenpropathrin; D) imi dacloprid; E) spinetoram; F) spirotetramat. F E D C B A

PAGE 118

118 Figure 5 2. Asian citru s psyllid EPG waveforms C on. A) Untreated control ; B) spinetoram treated plants (difficulties in stylet penetration). A B

PAGE 119

11 9 C HAPTER 6 RESIDUAL ACTIVITY OF FIVE DIFFERENT FOLIA R APPL IED INSECTICIDES ON ASIAN CITRUS PSYLLID ( DIAPHORINA CITRI ) FEEDING BEHAVIOR A ND THEIR POSSIBLE IMPLICATION S FOR LAS TRANSMISSION Several investigations of insecticide efficacy have been conducted against D.citri In field experiments, various insecticides and rates were analyzed against D. citri ( Childers and Rogers 2005) Chlorpyrifos (Lorsban 4E), fenpropathrin (Danitol 2.4 EC), and imidacloprid (Provado 1.6F) provided good adult knockdown control up to 5 DAT, and depending on the application date, the r esidual effect of fenpropathrin and imidacloprid were up to 15 DAT ( Childers and Rogers 2005). Similar results were obtained by Qureshi et al. (2009) when looking into foliar applied imidacloprid. In addition, Qureshi et al. (2009) found significantly fewe r adults psyllids 7 DAT on plants treated with spirotetramat (Movento 240SC), but nymphs were reduced even after 24 DAT. Although residu a l activity studies are based on nymphal and adult mortality they do not determine effects on psyllid feeding behavior. Since, Las pathogen is a phloem restricted bacterium, transmission probably occur during feeding activities, which take place in the phloem. Acquisition likely occur s during phloem ingestion and inoculation probably takes place during phloem salivation. C onsequently, insecticides that disrupt psyllid feeding behavior prior to causing mortality are the most likely to prevent spread of disease I nformation concerning the effects of the residual activity on the D. citri feeding behavior is very important in t he improvement of psyllid control and the subsequent success of HLB management programs In Chapter 5, D. citri feeding behavior was examined when exposed to the fresh residues of five different insecticides. In this study, the duration of feeding disrupti on provided by those same five insecticides is examined in detail. More specifically, the objective of this study was to determine residual activity 1, 7, 14, 21 and 28 DAT of chlorpyrifos (Lorsban

PAGE 120

120 4E), fenpropathrin (Danitol 2.4EC), imidacloprid (Provado 1.6F), spinetoram (Delegate WG), a nd spirotetramat (Movento 240SC ) in terms of disrupting D. citri feeding behavior. Material s and Methods Plants and Insects Citrus aurantium L.) seedlings (25 30 cm tall) planted in one gallon pots using citrus potting mix (Fafard Citrus Mix, Fafard, Agawam, MA). Seedlings were grown in a pathogen free greenhouse at 29 3 C and 60 80% RH. All seedlings were planted and maintained identically, to minimize interpl ant variation. Adults D. citri (10 20 d) used in experiments were obtained from a greenhouse colony free of Ca. Liberibacter asiaticus, reared on sour oranges and sweet oranges ( Citrus Sinensis (L.) Osbeck) at 29 3C with a 12:12 (L:D) h photoperiod. Pr ior to use in EPG recordings, psyllids (10 20 d old) were transferred to a rearing cage (61cm x 61cm x 91cm, Bioquip, Rancho Domingues, CA) containing sour orange plants for a 48 h acclimation period. Of the psyllids held for acclimation, both female and m ale D. citri were then selected for use in feeding experiments. EPG Recording and Waveform Analysis Recordings of D. citri feeding for 6 h under constant light conditions on insecticide treated and untreated plants were made using a Giga 8 monitor (Departm ent of Entomology, Wageningen Agricultural University, the Netherlands) Setup of EPG recordings and waveform characterizations were conducted as previously described in Chapter 3. Effects o f Residual Foliar Insecticides on Psyllid Feeding Behavior The ex periments investigating the residual effect of several foliar applied insecticides were divided in to five different experiments Each experiment was conducted 1, 7, 14, 21, and 28 DAT. Fifty plants per treatment were treated individually using one of the f ollowing insecticides:

PAGE 121

121 1) chlorpyrifos (Lorsban 4E), 2) fenpropathrin (Danitol 2.4EC), 3) imidacloprid (Provado 1.6F), 4) spinetoram (Delegate WG), and 5) spirotetramat (Movento 240SC). Twenty four hours prior to the first EPG recordings, insecticides wer e applied runoff from the leaves. Each insecticide was applied at the recommended field rate as described in Chapter 5. Between recording dates (eg 7, 14, 21 or 28 DAT) plants were weathered outdoors inside of a screen cage (180 cm x 360 cm x 180 cm, Bioquip, Rancho Domingues, CA) Te mperature and relative humidity, inside and outside of the screen cage were measured throughout the experiment using a HOBO data logger ( Onset Computer Corporation Bourne, MA). and two psyllid s w ere recorded per plant on separate leaves at the same time At the end of each recording, any insects that had wiring problems resulting in poor quality EPG recordings were discarded from the analysis. Residual Analysis Following EPG recordings, leaves from control, chlopyrifos and imidacloprid treated plants were sampled to obtain 5g of leaf material per treatment Pla nt samples were then sent to the Waters Agricultural Laboratory (Cam illa, GA) for residual analysis. Residual activities were analyzed by HPLC/UV chromatography. Statistical Analysis For each recording date, D. citri feeding behaviors were compared between the treatments. Analyses of feeding parameters were conducted at the cohort and insect levels as previously described in Chapter 2. square test was performed to test the goodness of fit (PROC GLIMMIX, SAS Institute 2001). Data were analyze d by ANOVA (PROC GLIMMIX, SAS Institute 2001)

PAGE 122

122 with the least significant difference (LSD) test (LSMEANS, SAS Institute 2001) used to determine if the waveform parameters analyzed were significantly different between the treatments Means were considered si Results Experiment 1 One Day A fter Treatment All D. citri tested performed non probing/walking activities (waveform np) on both treated and untreated plants. In addition, 100% (PPW) of the psyllids performed non walking/n on probing activities (waveform z ) on fenpropathrin and imidacloprid treated plants while 93.3 % 86.7% 66.7 % and 60% (PPW) of the insects performed the same behavior on spinetoram chlorpyrifos, spirotetramat treated and untreated plants, respectively. One hundred percent (PPW) of the D. citri performed stylet penetration (waveform C) on control, chlorpyrifos, imidacloprid, spinetoram and spirotetramat treated plants ; the same behavior w as observed on 80 % (PPW) of D. citri on fenpropathrin treated pla nts. However n one of the psyllids on chlopyrifos, fenpropathrin and imidacloprid treated plants reached the phloem and salivated However 20% (PPW) of the insects penetrated and salivated in the phloem (D and E1, respectively) on untreated plants while 1 3.3 % (PPW) and 6. 7 % of the D. citri tested performed those same waveforms on spirotetramat and spinetoram treated plants. In addition, only psyllids on spirotetramat treated plants performed phloem ingestions (waveform E2). Xylem ingestion (waveform G) was performed on chlorpyrifos, spinetoram spirotetramat treat ed and untreated plants by 13.3%, 26.7%, 60 % and 46.7 % of D. citri respectively. Cohort level. D. citri had a total access period of 324,000 s during which psyllids on the untreated plants (contr ol) probed 933 times (TNP) and spent 107,532.96 s (TPD) with stylet s inserted in to the citrus leaves. However psy llid probing on spirotetramat treated plants was decreased to 636 times (TNP) and psyllids spent more time probing (TPD=189,584.91 s). In

PAGE 123

123 addi tion, psyllids on spinetoram spin eto ram treated plants probed 601 times (TNP) and spent 90,956.07 s (TPD) probing In contra st probe frequencies and duration s were reduced on fenpropathrin, chlo r pyrifos and imidacloprid treated plants Ps yllids on fenpropa thrin treated plants performed 318 probes (TNP) and spent 10,358.05 s (TPD) with stylets inserted into the leaf tissue O n chlorpyrifos treated plants the total number of probes was 311 and the total probe duration 44,019.06 s O n imidacloprid treated pla nts they probe d 178 times (TNP) and for a duration of 10,383.91 s (TPD). Insect level. The differences found at cohort level, were also found to be significantly different for the number of probes per insect (NPI) P syllids on control, spinetoram and spir otetramat treated plants performed significantly more probes than D. citri feeding on chlorpyrifos, fenpropathrin and imidacloprid treated plants ( F = 9.32; df = 5, 83; P < 0.0001; Table 6 1). In addition, the duration of probes per insect (PDI), was signi ficantly longer in duration on spirotetramat treated plants, than on untreated and spinetoram treated plants ( F = 9.17; df = 5, 81 ; P < 0.0001) T he shortest durations occurred on imidacloprid and fenpropathrin treated plants (Table 6 2). Significant differ ences were also found in the waveform duration per insect (WDI) P syllids on control, chlorpyrifos, spin etoram and spirotetramat treated plants non probed/walked (np) for longer periods of time ( F = 8.68; df= 5, 84 ; P < 0.0001) than psyllids on fenpropathr in and imidacloprid treated plants If the psyllids were not probing and walking, they were performing non probing/standing still behaviors which were significantly longer on the chlorpyrifos, fenpropathrin and imidacloprid treated plants ( F = 19.26; df = 5, 70 ; P < 0.0001) than on control and spirotetramat treated plants. Thus, pathway/stylet penetration ( waveform C) ( F = 18.19; df = 5, 81 ; P < 0.0001) was longer in duration on the control plants, spinetoram and spirotetramat treated plants but shorter on fenpropathrin and imidacloprid treated plants. In

PAGE 124

124 addition, xylem ingestion ( waveform G) ( F = 0.61; df = 3, 18 ; P = 0.61 ; Table 6 3) was not differen t among the treatments. Experiment 2 Seven Days A fter Treatment One hundred percent (PPW) of D. citri t ested performed non probing/walking activities (waveform np) on chlorpyrifos, fenpropathrin, spinetoram spin eto ram, spirotetramat treated, and untreated plants, while 92. 9 % of the psyllids performed the same behavior on imidacloprid treated. When psyllids were not walking or probing, they were non probing/standing still, consequently one hundred percent (PPW) of D. citri performed non walking/non probing activities (z) on spinetoram treated plants, while 92. 9 % (PPW) performed the same behavior on imidaclopr id and fenpropathrin treated plants. In addition, 78. 6 %, 50% and 35.7 % (PPW) of the psyllids non probed/st ood still on chlorpyrifos, spirotetramat treated, and untreated plants respectively. One hundred percent (PPW) of D. citri performed pathway/stylet p enetration (C) on chlorpyrifos, spin e t o ram, spirotetramat treated and untreated plants, while 92. 9 % and 78. 6 % (PPW) of the psyllids performed the same behavior on fenpropathrin, and imidacloprid treated plants. However, only 7.1 % (PPW) of the psyllids on c h lorpyrifos, spirotetramat, and o n control plants were able to penetrate and salivate in to the phloem (D and E1, respec tively). In addition, only 7.1 % (PPW) of D. citri performed phloem ingestion (E2) on control plants. Xylem ingestion (G) was performed by 50 % (PPW) of D. citri exposed to chlorpyrifos, spirotetramat treated a nd untreated plants, while 35.7 % (PPW) of D. citri performed this same behavior on spin e t o ram treated plants. Cohort level D. citri had a total access period of 302,000 s on control plants during which psyllids inserted and withdr e w their stylets 582 times (TNP) which last ed for 152,652.98 s (TPD). In addition, psyllids on the chlopyrifos and spirotetramat treated plants had a total probe

PAGE 125

125 duration (TPD) of 117,595.37 s and 117,090.00 s respectively with a total number of probes (TNP) performed at 593 and 313 times respectively P syllids exposed to spin e t o ram treated plants, probed 359 times (TNP) and with a total duration of 69,471.09 s (TPD). Ps yllids on imidacloprid treated plants probed 135 times (TNP) and spent 9,331.80 s ( TPD) probing. O n fenpropathrin treated plants psyllids probed 117 times (TNP) and spent 7,125.33 s (TPD) with their stylets inserted in to the leaf tissue. Insect level. P syllids on spirotetramat treated and un treated plants exhibited significantly higher NPI, than those on fenpropathrin and imidacloprid treated plants ( F = 8.38; df = 5, 77; P < 0.0001; Table 6 4) In addition, the duration of those probes per insect (PDI) w as significantly longer for psyllids o n untreated plants than on spirotetramat treated plants ( F = 18.28; df = 5, 74 ; P < 0.0001) and was the shortest on fenpropathrin and imidacloprid treated plants (Table 6 4). Significant differences were also found for the parameter WDI Non probing/ standi ng still (z) behavior was longer in duration for psyllids on imidacloprid and fenpropathrin treated plants ( F = 6.63; df =5, 57 ; P < 0.0001 ) with shorter duration found for psyllids on control, chlorpyrifos, spin e t o ram and spirotetramat treated plants. N on probing/walking behaviors (np) were significantly longer in duration on chlorpyrifos spin e t o ram spirotetramat and untreated plants compared to fenpropathrin and imidacloprid treated plants ( F = 10.50; df = 5, 77 ; P < 0.0001). S tylet penetration ( wavefo rm C) ( F =17.33; df = 5, 74 ; P < 0.0001), was longer on control plants than spirotetramat and chlorpyrifos treated plants. Stylet penetrations were shortest on fenfopathrin and imidacloprid treated plants. X ylem ingestion (G), was not significant ly differe n t among the treatments tested ( F = 0.34; df = 3, 22; P = 0.7990; Table 6 4). Experiment 3 Fourteen Day s A fter Treatment One hundred percent (PPW) of D. citri performed non probing/walking activities (np) on control, chlorpyrifos, fenpropathrin and spirot etramat treated plants. In addition, the same

PAGE 126

126 behavior s w ere performed by 93.3% (PPW) of psyllids exposed to imidacloprid and spin e t o ram treated plants. If the insects were not walking or probing, they were performing non probing/standing still behaviors ( z) T his behavior was observed for all (PPW) of the D. citri on control, chlorpyrifos and fenpropathrin treated plants; on 93.3 % (PPW) of the psyllids on imidacloprid and spin e t o ram treated plants; and on 80 % (PPW) of the insects on spirotetramat treated plants. All of the (PPW) of D. citri performed pathway/stylet penetration (C) on control, chlorpyrifos and fenpropathrin treated plants and the same behavior was performed by 93.3 % (PPW) of the psyllids on imidacloprid, spinetoram and spirotetramat trea ted plants. For psyllids probing the phloem, only 13.3 % (PPW) of th ose exposed to chlorpyrifos, spirotetramat treated and untreated plants were penetrated and salivated into the phloem (D and E1, respectively). However only 13.3 % and 6. 7 % (PPW) of D. ci tri on control and spinetoram treated plants were ingested phloem sap (E2). Xylem ingestion (G) was performed by 53.3 % (PPW) of D. citri on spirotetramat treated plants, 40 % (PPW) on control plants, 33.3 % (PPW), chlorpyrifos treated plants and 20 % (PP W) on spinetoram treated plants. In addition, none of the psyllids performed xylem ingestions on imidacloprid and fe n propathrin treated plants. Cohort level. D. citri had a total access period of 324,000 s during which psyllids on the control plants psylli ds probed 965 times (TNP) and spent 111,842.24 s (TPD) with their probing activities. P syllids on spinetoram treated plants probed 844 times (TNP) and spent 89,068.19 s (TPD) with their mouth parts inserted in to the leaves. In addition, insects on chlorpyr ifos treated plants psyllids probed 694 times (TNP) with a total duration of 104,961.59 s (TPD) O n the spirotetramat treated plants psyllids probed 577 times (TNP) with a total duration of 155,952.30

PAGE 127

127 s (TPD). However total frequency and duration of prob ing was greatly r e duced on fenpropathrin and imidacloprid treated plants P syllids on fenpropathrin treated plants probed 196 times (TNP) and spent 26,223.16 s (TPD) feeding O n imidacloprid treated plants, psyllids probed 177 times (TNP) and spent 9,199.9 1 s (TPD) performing this behavior. Insect level. The differences found between TPD and TNP, were also observed with the number of probes per insect (NPI) and probe duration per insect (PDI). NPI was significantly higher ( F = 5.74 df = 5, 82; P = 0.0001; T able 6 5) for D. citri on control, chlorpyrifos, spinetoram and spirotetramat treated plants, whereas it was shortest on fenpropathrin and imidacloprid treated plants. PDI was also longest on control, chlorpyrifos, spinetoram and spirotetramat treated plan ts ( F = 3.81; df = 5, 81 ; P < 0.0001) and shortest on fenpropathrin and imidacloprid treated plants (Table 6 5). In addition, insects stood still (z) for the longest durations ( F = 7.65; df = 5, 79 ; P < 0.0001) on the fenpropathrin and imidacloprid treated plants and the shortest duration on control, chlorpyrifos, spinetoram and spirotetramat treated plants. The opposite results were observed for the non probing/walking behavior (np) ( F = 9.58; df = 5, 82 ; P < 0.0001) These behaviors were performed for the longest duration on control, chlorpyrifos, spinetoram and spirotetramat treated plants and shortest duration on fenpropathrin and imidacloprid treated plants. For WDI there were significant differences between treatments for pathway/stylet penetration ( w aveforms C) ( F = 11.67; df = 5, 81 ; P < 0.0001) The longest duration of waveform C occurred for control, chlorpyrifos, spinetoram and spirotetramat treated plants with shorter durations on fenpropathrin and imidacloprid treated plants. However there was a high percentage of psyllids performing stylet penetration W aveform D did not occur in all treatments. For p syllids which performed waveform D there was no significant difference between treatments ( F = 0.42; df = 1, 2 ; P = 0.58) which was also the case for D. citri

PAGE 128

128 performing phloem salivation (E1) ( F = 0.00; df = 1, 2; P = 0.99) and phloem ingestions (E2) ( F = 0.87; df = 1, 1 ; P = 0.5214 ; Table 6 5). For WDI x ylem ingestion (G) was significantly differen t between treatments ( F = 5 27; df = 3, 18 ; P = 0.0087 ). The longest duration occurred on the control plants and the shortest duration s were on chlorpyrifos, spinetoram and spirotetramat treated plants. Experiment 4 Twenty one Day s A fter Treatment All psyllids (PPW) on treatments tested performed non probing/walking activities (np) W hen they were not probing or walking, they were standing still. Consequently, 100%, 93.3%, 73.3 % and 53.3 % (PPW) of the psyllids on imidacloprid, chlorpyrifos, spinetoram treated and untreated plants, respectively, perf ormed non probing/non walking (z) activities. In addition, 80 % (PPW) of D. citri performed waveform z on fenpropathrin and spirotetramat treated plants. All psyllids on all of the treatments performed pathway/stylet penetration (C). However only 26.7 % and 20% of D. citri penetrated and salivated in the phloem (D and E1, respectively) on control and chlorpyrifos treated plants respectively. Thus, only 6.7 % (PPW) of D. citri performed the same behavior on fenpropathrin and spirotetramat treated plants. Of the psyllids which reached the phloem, only 20 % (PPW) performed phloem ingestion (E2) on control and chlorpyrifos treated plants Also, 6 % (PPW) performed the same behavior on fenpropathrin treated plants. In addition to phloem ingestion, a large perc entage of the psyllids performed xylem ingestion (G) Specifically, 60%, 46.7 %, 40 %, 33.3 % and 6. 7 % (PPW) of the tested psyllids performed waveform G on control, spirotetramat, chlorpyrifos, spinetoram and fenpropathrin treated plants respectively. Coh ort level. D. citri had a total access period of 324,000 s during which psyllids on the control plants psyllids probed 493 times (TNP) and spent 153,742.06 s (TPD) feeding behavior P syllids on the spirotetramat treated plants probed 402 times (TNP) and ha d a total probing

PAGE 129

129 duration of 103,795.11 s (TPD). Similarly psyllids on chlorpyrifos treated plants probed 301 times (TNP) and spent 110,890.77 s (TPD) with their stylet s inserted into the leaves. However psyllids on spinetoram treated plants probed 708 times (TNP), and for (TPD= 82,140.57 s). In addition, psyllids on fenpropathrin and imidacloprid treated plants performed a total number of probes (TNP) of 339 and 209 respectively, with total probing duration (TPD) of 39,600.71 s and 19,495.92 s respecti vely. Insect level. The differences found in the cohort level were also analyzed at the insect level. The number of probes per insect (NPI) was higher for psyllids on control and spinetoram treated plants than on chlorpyrifos, fenpropathrin spirotetramat and imidacloprid treated plants ( F = 3.87; df = 5, 84; P = 0.0033; Table 6 6) ; D. citri on imidacloprid tre ated plants had the lowest NPI. In addition, for the probe duration per insect (PDI) there was a significantly longer duration ( F = 8.12; df = 5, 84 ; P < 0.0001 ) for psyllids on control, chlorpyrifos, spinetoram and spirotetramat treated plants compared to fenpropathrin and imidacloprid treated plants (Table 6 6). Significant differences were also found in the waveform duration per insect (WDI) No n p robing/non walking activities (z) were longe r on imidacloprid treated plants ( F = 3.05; df = 5, 66 ; P = 0.015) th an control, chlorpyrifos, spinetoram and spirotetramat treated plants. The non probing/walking waveforms (np) ( F = 2.40; df = 5, 84 ; P = 0.044) were significantly longer in duration on control, chlorpyrifos, fenpropathrin, spinetoram and spirotetramat treated plants compared to imidacloprid treated plants (Table 6 6) Thus pathway/stylet penetration (C) was longest on chlorpyrifos, spinetoram an d spirotetramat treated plants ( F = 6.31; df = 5, 84 ; P < 0.0001,) than on fenpropathrin and imidacloprid treated plants. There was no difference among the treatments for phloem penetration (D) ( F = 2.07; df = 3, 5 ; P = 0.2200), phloem salivation

PAGE 130

130 (E1) ( F = 0.88; df = 3, 5 ; P = 0.5090), phloem ingestion (E2) ( F = 0.1; df = 2, 4 ; P = 0.8500) and xylem ingestion (G) ( F = 0.55; df = 4, 23 ; P = 0.703 ; Table 6 6). Experiment 5 Twenty eight Day s A fter Treatment All (PPW) of the D. citri performed non probing/walk ing activities (np) on control, chlorpyrifos, fenpropathrin, imidacloprid and spinetoram treated plants. In addition, 92. 7 % (PPW) of D. citri performed the same behavior on spirotetramat treated plants. While psyllids were not walking and probing, they we re performing non probing/standing still (z) events C onsequently 92. 7 %, 85.7 %, and 57.1% (PPW) of D. citri performed waveform z on imidacloprid, spinetoram and chlorpyrifos treated plants respectively Also 64. 3 % (PPW) of D. citri performed standing still events on fenpropathrin and spirotetramat treated plants. For WDI all D. citri performed pathway/stylet penetration (C) on control, fenpropathrin, and spinetoram treated plants Also, 92. 9 % of psyllids (PPW) on chlorpyrifos treated plants and 85.7 % (PPW) on imidacloprid, and spirotetramat treated plants performed stylet penetration Only a few D. citri reached and penetrated the phloem (waveform D) on control, chlorpyrifos and imidacloprid treated plants (21.4 %) and 7.1 % (PPW) on spirotetramat t reated plants. Of th ose only 21.4 % (PPW) of D. citri salivated i nto the phloem (E1) on chlorpyrifos and imidacloprid treated plants; 14. 3 % on control plants and 7.1 % (PPW) performed the same behavior on spirotetramat treated plants. However, only 14. 3 % (PPW) of D. citri were able to perform phloem ingestion (E2) on chlorpyrifos and imidacloprid treated plants, while 7.1 % performed the same behavior (PPW) on control plants and spirotetramat treated plants. In addition, a high percentage of psyllids wer e able to ingest xylem Specifically 71.4 %, 64. 3 %, 50 %, 42. 7 %, 21.4 %, and 7.1 % of D. citri ingested xylem (G) on chlorpyrifos, control, spirotetramat, spinetoram fenpropathrin, and imidacloprid treated plants.

PAGE 131

131 Cohort level. There was a considerable amount of variation at the cohort level data 28 DAT. D. citri had a total access period of 302,000 s P syllids on control, spirotetramat, fenpropathrin, spinetoram chlorpyrifos and imidacloprid treated plants performed a TNP of 371, 323, 317, 313, 263, an d 141 respectively. Those probes lasted (TPD) 109,293.61 s; 143,469.34 s; 80,848.41 s; 109,116.97 s; 119,905.33 s and 53,097.00 s respectively Insect level. There were significant differences in the number of probes per insect (NPI) ( F = 2.57; df = 5, 7 7; P = 0.0334; Table 6 7) between treatments. T otal probe frequency was higher for psyllids on control, fenpropathrin, spinetoram and spirotetramat treated plants compared to chlorpyrifos and imidacloprid treated plants PDI was significantly longer on spi rotetramat treated plants ( F = 4.08; df = 5, 73 ; P = 0.0025) than on imidacloprid treated plants (Table 6 7). In addition, for waveform duration per insect WDI non probing moving (np) ( F = 3.97; df = 5, 77 ; P = 0.0030) was longer on control, chlorpyrifos, fenpropathrin, imidacloprid and spirotetramat treated plants than spinetoram treated plants Also, non probing and standing still (z) ( F = 3.82; df = 5, 54 ; P = 0.0049) was longer on imidacloprid treated plants than on control, chlorpyrifos, spinetoram an d spirotetramat treated plants P athway/stylet penetration waveforms (C) ( F = 3.98; df = 5, 73 ; P = 0.0030) was longer on spirotetramat treated plants than on imidacloprid treated plants. However, phloem penetration (D) ( F = 4.33; df = 3, 6 ; P = 0.0601 ), p hloem salivation (E1) ( F = 0.63; df = 3, 5 ; P = 0.6273), phloem ingestion (E2) ( F = 1 32; df = 3, 3 ; P = 0.4587) and xylem ingestion (G) ( F = 1.00; df = 5, 30; P = 0.4328) were not significantly different between treatments (Table 6 7). Summary of results Overall, 1 DAT, the duration of the probing activities w as reduced on fenpropathrin and imidacloprid treated plants compared with untreated plants Feeding activities ceased fastest on imidacloprid and fenpropathrin treated plants as a result of decrease d

PAGE 132

132 probe, a reduc tion in the number and duration of stylet penetrations (C), phloem activities (waveforms D, E1 and E2), xylem ingestion (G), and non probing/walking (np) Consequently, there was an increase in the non probing/standing still activities (z) by D. citri on fenpropathrin and imidacloprid treated plants. Chlopyrifos, spinetoram and spirotetramat did not affect probing activities compared to untreated plants. However psyllids on chlorpyrifos treated plants probe d half as long as on untreated pla nts P syllids on spinetoram and spirotetramat treated plants probed for the same duration or even longer than psyllids on untreated plants. In addition, psyllids on spinetoram and spirotetramat treated plants penetrat ed the phloem and salivat ed normally; h owever, psyllids were able ingest phloem only on spirotetramat treated plants At 7 DAT, psyllids on chlorpyrifos treated plants doubled the time spent probing. In contra st all the other insecticides had slightly smaller number and duration of probes, bu t only fenpropathrin, imidacloprid and spinetoram treated plants were significantly diffe rent than the control. Fenpropathrin, imidacloprid and spinetoram showed smaller duration in sty let penetration behaviors (C), and no performance of phloem activities (D, E1 and E2) C onsequently insects had longer durations on the non probing/standing still (z) events, although, they had a smaller duration of non probing/walking (np) when compared to chlorpyrifos, control and spirotetramat treated plants At 14 DAT, feeding was disrupt ed on fenpropathrin and im i dacloprid treated plants. Psyllids on fenpropathrin and imidacloprid treated plants had their stylet penetration (C), phloem activities (waveform D, E1, and E2), and xylem ingestion (G) reduced when compared t o chlorpyrifos, control, spinetoram, spirotetramat treated plants. C onsequently they also had their non probing/standing (z) duration increased S ince those insecticides were still causing insect mortality, the non probing/walking (np) events were also red uced. Although D. citri on untreated

PAGE 133

133 plants did not successfully reach the phloem, psyllids on chlo r pyrifos and spinetoram treated plants were successful in performing phloem penetration (D), salivation (E1), and ingestion (E2) behaviors Compared to the o ther insecticides, D. citri feeding on spirotetramat treated plants did not show any significant difference in feeding behaviors when compared to untreated plants. There were still significant reductions 21 DAT in probe duration for D. citri feeding on fe npropathrin and imidacloprid treated plants However the frequency of probes started to homogenize and fenpropathrin treated plants showed not to be significantly different from control, chlorpyrifos, and spirotetramat treated plants. In addition, wavefor m duration per insect were still different for the duration of the stylet penetrations (C) on both fenpropathrin and imidacloprid treated plants N on probing/walking events were significantly smaller for imidacloprid treated plants and the non probing/stan ding still events were significantly higher. Phloem activities (D, E1 and E2), were performed by psyllids on fenpropathrin treated plants and were not significantly difference from the control, chlorpyrifos, and spirotetramat treated plants. In addition, t here was no significant difference in xylem feeding behavior between any of the treatments. At 28 DAT, the imidacloprid treatment was the only case where D. citri were still performing probes of shorter duration compared to control plants. This was a direc t result of shorter durations of the stylet penetrations (C) and xylem ingestion (G) yet longer non probing/standing still (z) events. Although, a small percentage of those psyllids were able to find the phloem, salivate and ingest into the phloem. However the number of D. citri that were able to perform these phloem behaviors was too small to allow statistical comparison. There was no significant difference in D. citri feeding behavior on c hlorpyrifos, fenpropathrin, spinetoram, and spirotetramat treated plants when compared to control plants.

PAGE 134

134 The average of the probe duration per insect (PDI), average of number of probes per insect (NPI) and total waveform duration (TWD) through the whole experiment are shown on figures 6 1, 6 2 and 6 3, respectively TW D of waveforms np, C, G, D, E1 and E2 shown to be negatively proportional to the residual concentration of chlorpyrifos and imidacloprid (figure 6 3), while NPI, and TWD of waveform z, were positively proportional. Residual A nalyses and T emperature R ecord ing The average residual activity of chlorpyrifos and imidacloprid in leaf tissues are represented in figure 6 4. Temperature and relative humidity variations were measured through the insecticide residual break down; they were also recorded through the ex periment and are represented in the figure 6 4. T emperatures did not differ inside and outside of the cage, but relative humidity did differ. Humidity inside of the cage was significantly higher th an outside of the cage ( F = 28.57 ; df = 1, 1486 ; P < 0.0001 ). Discussion The current study investigated the effect of residual activity of five different insecticides on D. citri feeding behavior. The data indicated that feeding by D. citri was reduced on fenpropathrin and imidacloprid treated plants for up to 21 and 28 days, respectively Chlorpyrifos affected psyllid phloem contact for only 1 DAT. D. citri on spinetoram treated plants were able to reach the phloem 1 DAT; however, they did perform probes of shorter duration 1 and 7 DAT. Treatment of plants with sp irotetramat did not affect psyllid feeding. In fact D. citri probed for longer durations on spirotetramat treated plants compared to untreated plants. Breakdown of insecticides depends on environmental factors (temperature, relative humidity, and wind s peed), and plant factors (species and plant growth) (Edwards 1975).

PAGE 135

135 Temperature was not affected by our cages; however, relative humidity was higher inside of cages. In order to trace insecticide break down, chlorpyrifos and imidacloprid leaf samples were analyzed. Ideally, all the insecticides should have been analyzed, but not all the residual activity anal y s e s were available. Feeding behaviors indirectly indicate d breakdown of chlorpyrifos, fenpropathrin, imidacloprid and spinetoram treated plants. P syl lids on those treated plants showed a negative correlation on probing, stylet penetration, phloem penetration, salivation and ingestion, and xylem ingestion when compared to the chlorpyrifos and imidacloprid residual concentrations. Thus we are able to see a positive relationship of the non walking/standing still event when comparing to chlorpyrifos and imidacloprid residual concentrations. Chlorpyrifos, fenpropathrin, and imidacloprid are mainly contact insecticides and act directly on the insect nervous s ystem (acetylcholinesterase inhibitors, sodium channel modulators, and acetylcholine receptor stimulator s respectively) These insecticides can c ause over stimulation of the nervous system quickly killing the insects. Consequently, chlorpyrifos fenpropat hrin, and imidacloprid cause a rapid effect on the insect nervous system which could explain the immediate reduction in D. citri feeding behaviors in this study. Spinetoram is also a contact insecticide. However, this insecticide acts via disruption of th e nicotinic/GABA gated chloride channels trigger ing either an up regulation or a down regulation of neurotransmitters which is slower acting but still results in insect death (Casida and Quistad 2004) and could explain the fact that D. citri were still ab le to perform phloem feeding behaviors prior to the occurrence of mortality, even 1 DAT. S pirotetramat has limited contact activity and is mainly effective after insect digestion and prevents formation of lipids required for reproduction or ecdysis (Nauen et al. 2008) which

PAGE 136

136 would explain the lack of noticeable effects on psyllid feeding behavior which may take a much longer period of time for subleathal effects on feeding behavior to occur. In addition to different modes of action, insect i ci des formulati on characteristics could have some affect on the feeding behavior of D. citri Insecticide break down also depend s on the insecticide characteristics, such as active ingredient stability, volatility, and formulation (Edwards 1975). Therefore, it is possible that the five different insecticides tested c ould have degrade d at different rates affect ing p syllid behavior differentially Chlorpyrifos (Lorsban 4E) and fenpropathrin (Danitol 2.4 EC) formulations were applied as emulsifiable concentrate s. I midacloprid (Provado 1.6F) as a flowable, spinetoram (Delegate WG) as a wettable granulate and spirotetramat (Movento 240SC) i s a suspension concentrate. Emulsifiable concentrated insecticides applied to plants have their active ingredients immediately available for insect control However, those types of insecticides are not as persistent as granulates and they are easily washed away (Edwards 1975, Montemurro et al. 2002). Flowables and suspension concentrate s are a liquid suspension A fter application they form sma ll crystals on the treated surfaced, and since those crystals are insoluble, they have a better residual activity W ettable granulate s are insecticides that are not immediately active and the active ingredient releases within 1 to 30 h after applications which depend s on light exposure (Montermurro et al. 2002). These facts could explain the longe r residual activity of products such as imidacloprid compared to chlorpiryfos and fenpropathrin and perhaps why spinetoram (a wettable granule) was more effective 7 DAT compared to 1 DAT. R esults from this study demonstrate that while some insecticides may cause relatively rapid mortality of adult psyllids, there is considerable variability that exists among these products in terms of the duration of feeding disru ption p rovided. While some insecticides provided

PAGE 137

137 feeding disruption lasting 3 4 weeks (fenpropathrin and imidacloprid), protection provided by ot her products was much shorter. Overall, the results of this study can be used to help guide citrus growers in p roduct selection and also determine when additional applications will be necessary.

PAGE 138

138 Table 6 1. Mean ( SE) number of probe per insect (NPI) for Diaphorina citri feeding on chlo r pyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated an d untreated citrus plants up to 28 d after treatment (DAT) DAT a Treatments 1 7 14 21 28 NPI SEM NPI SEM NPI SEM NPI SEM NPI SEM Untreated control 62.60 10.58 a 41.85 8.98 ab 64.93 14.53 a 33.33 6.17 ab 26.85 3.88 a Chlo rpyrifos 21.53 3.22 b 22.78 3.82 c 46.73 8.68 a 20.73 2.51 bc 19.21 4.64 ab Fenpropathrin 23.50 8.26 b 9.35 1.70 d 14.00 2.52 b 23.46 5.70 bc 23.50 4.62 a Imidacloprid 12.86 1.89 b 11.30 2.42 d 13.50 2.07 b 14.86 2.76 c 11.00 3.04 ab Spinetora m 41.06 5.97 a 26.64 5.96 bc 61.07 16.11 a 48.06 9.53 a 23.07 4.86 a Spirotetramat 42.66 4.99 a 43.14 7.30 a 38.93 8.45 a 27.33 4.06 bc 25.23 4.58 a a Means with different lower case letters are not significantly d ifferent (LSMEANS, P < 0.05)

PAGE 139

139 Table 6 2. Mean ( SE) probe duration per insect (PDI) (s) for Diaphorina citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated citrus plants up to 28 d after treatme nt (DAT). DAT a 1 7 14 21 Treatment PDI SE PDI SE PDI SE PDI SE Untreated control 119.78 22.43 ba 181.73 24.55 a 124.27 19.14 a 170.82 28.21 a Chlorpyrifos 48.91 12.76 b 139.41 26.87 ba 116.62 23.04 a 123.21 24. 72 a Fenpropathrin 14.39 5.75 c 9.14 2.42 c 29.14 19.15 b 44.00 20.82 b Imidacloprid 11.54 3.82 c 14.14 3.37 c 10.95 3.01 b 21.66 6.41 b Spinetoran 101.06 15.28 ba 82.70 17.21 b 106.03 28.39 a 91.27 19.65 a Spirotetramat 210.65 25.30 a 139.99 21.72 ba 182.09 28.54 a 115.33 22.64 a a Means with different lower case letters are not significantly different (LSMEANS, P < 0.05) Table 6 2. Continued DAT 28 Treatment PDI SE Untreated control 130.13 25.88 ba Chlorpyrifos 153.72 24.41 ba Fenpropathrin 96.25 28.71 b Imidacloprid 73.75 26.97 c Spinetoran 129.90 26.37 ba Spirotetramat 199.26 26.06 a a Means with different lower case letters are not significantly different (LSMEANS P < 0.05)

PAGE 140

140 Table 6 3. Mean ( SE) waveform duration per insect (WDI) (s) for Diaphorina citri feeding on chlo r pyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetrama t treated and untreated plants 1 d after treatment (DAT) 1st Day Waveforms z n p C G Treatment WDI SE WDI SE WDI SE WDI SE Untreated control 45.82 19.07 c 166.02 1768.48 a 89.43 15.22 ab 59.79 10.39 a Chlorpyrifos 208.70 24.03 ab 109.91 1327.96 a 36.34 7.13 b 94.28 79.05 a Fenpropathrin 296.40 22.12 ab 35.41 495.54 b 14.39 5.76 c n/a n/a n/a Imidacloprid 305.89 9.68 a 41.05 322.58 b 11.50 3.80 c n/a n/a n/a Spinetoram 163.33 24.58 b 93.13 826.14 a 92.14 13.66 a 33.03 6.10 a Spirotetramat 76.41 3 8.00 c 96.80 968.17 a 165.11 21.28 a 57.23 10.50 a 1 Means with different lower case letters are not significantly different (LSMEANS, P < 0.05) Table 6 3. Continued Waveforms D E1 E2 Treatment WDI SE WDI SE WDI SE Untreated c ontrol 2.77 2.30 a 7.77 7.44 a n/a n/a n/a Chlorpyrifos n/a n/a n/a n/a n/a n/a n/a n/a n/a Fenpropathrin n/a n/a n/a n/a n/a n/a n/a n/a n/a Imidacloprid n/a n/a n/a n/a n/a n/a n/a n/a n/a Spinetoram 0.95 n/a a 0.64 n/a a n/a n/a n/a Spirotetramat 1.47 0.90 a 0.68 5.50 a 81.80 11.55 n/a 1 Means with different lower case letters are not significantly different (LSMEANS, P < 0.05)

PAGE 141

141 Table 6 4. Mean ( SE) waveform duration per insect (WDI) (s) for Diaphorina citri f eeding on chlo r pyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated plants 7 d after treatment (DAT) 1 Means with different lower case letters are not significantly different (LSMEANS, P < 0.05) Tab le 6 4. Continued Waveforms D E1 E2 Treatment WDI SE WDI SE WDI SE Untreated control 11.7725 n/a n/a 18.70 n/a n/a 45.95 n/a n/a Chlorpyrifos 0.4 n/a n/a 1.56 n/a n /a n/a n/a n/a Fenpropathrin n/a n/a n/a n/a n/a n/a n/a n/a n/a Imidacloprid n/a n/a n/a n/a n/a n/a n/a n/a n/a Spinetoram n/a n/a n/a n/a n/a n/a n/a n/a n/a Spirotetramat 2.58 n/a n/a 2 n/a n/a n/a n/a n/a 1 Means wi th different lower case letters are not significantly different (LSMEANS, P < 0.05) 1st Week Waveforms z np C G Treatment WDI SE WDI SE WDI SE WDI SE Untreated control 31.96 15.27 d 151.05 22.02 a 153.19 21.16 a 47.29 16.34 a Chlorpyrifos 94.07 34.77 cd 150.69 23.69 a 106.44 21.48 ab 65.63 11.96 a Fenpropathrin 286.24 26.33 ab 86.49 32.22 b 9.14 2.42 c n/a n/a n/a Imidacloprid 303.83 9.08 a 45.38 7.08 b 14.14 3.37 c n/a n/a n/a Spinetoram 124.49 22.79 bc 157.12 18.79 a 60.31 11.24 b 62.69 23.14 a Spirotetramat 97.91 33.66 cd 175.16 16.75 a 107.09 20.23 ab 65.15 20.00 a

PAGE 142

142 Table 6 5. Mean ( SE) waveform duration per insect (WDI) (s) for Diaphorina citri feeding on chlo r pyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat tre ated and untreated plants 14 d after treatment (DAT) 2nd Week Waveforms z np C G Treatment WDI SE WDI SE WDI SE WDI SE Untreated control 75.90 21.88 b 152.25 17.58 a 76.26 13.34 a 119.96 20.03 a Chlorpyri fos 97.89 28.72 b 142.81 17.79 a 92.98 17.37 a 56.13 13.36 b Fenpropathrin 288.77 23.73 a 41.62 10.62 b 29.14 19.15 b n/a n/a n/a Imidacloprid 301.63 10.83 a 44.05 5.77 b 10.95 3.01 b n/a n/a n/a Spinetoram 98.59 26.93 b 152.39 24.97 a 77.75 19.24 a 31.97 4.28 b Spirotetramat 79.02 35.25 b 126.62 20.50 a 145.30 27.74 a 63.84 9.63 b Means with different lower case letters are not significantly different (LSMEANS, P < 0.05) Table 6 5. Continued Waveforms D E1 E2 Treatment WDI SE WDI SE WDI SE Untreated control n/a n/a n/a n/a n/a n/a n/a n/a n/a Chlorpyrifos 1.50 1.05 a 2.44 0.05 a 32.73 30.49 a Fenpropathrin n/a n/a n/a n/a n/a n/a n/a n/ a n/a Imidacloprid n/a n/a n/a n/a n/a n/a n/a n/a n/a Spinetoram 0.61 0.00 a 4.31 3.58 a 178.13 n/a a Spirotetramat n/a n/a n/a n/a n/a n/a n/a n/a n/a Means with different lower case letters are not significantly di fferent (LSMEANS, P < 0.05)

PAGE 143

143 Table 6 6. Mean ( SE) waveform duration per insect (WDI) (s) for Diaphorina citri feeding on chlorpyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated plants 21 d after treatment (DAT). 3rd Week Waveforms z np C G Treatment WDI SE WDI SE WDI SE WDI SE Untreated control 74.40 33.48 b 152.85 24.20 ab 105.80 23.32 a 58.89 8.61 a Chlorpyrifos 72.16 24.73 b 172.79 23.14 a 81.76 17.36 a 5 4.13 11.79 a Fenpropathrin 151.83 30.78 ab 197.89 27.30 a 30.48 10.21 b 26.56 n/a a Imidacloprid 223.08 22.58 a 115.23 22.82 b 21.65 6.41 b n/a n/a n/a Spinetoram 100.90 27.85 b 198.10 20.56 a 73.15 15.82 a 54.31 31.06 a Spiro tetramat 101.52 28.55 b 166.82 21.92 a 89.45 18.98 a 55.18 13.96 a Means with different lower case letters are not significantly different (LSMEANS, P < 0.05) Table 6 6. Continued Waveforms D E1 E2 Treatment WDI SE WDI SE WDI SE Untreated control 1.92 0.85 a 4.26 2.45 a 140.11 67.21 a Chlorpyrifos 4.06 0.81 a 5.53 1.70 a 89.40 29.21 a Fenpropathrin 1.68 n/a a 2.42 n/a a 172.15 n/a a Imidacloprid n/a n/a n/a n/a n/a n/a n/a n/a n/ a Spinetoram n/a n/a n/a n/a n/a n/a n/a n/a n/a Spirotetramat 0.95 n/a a 0.86 n/a a n/a n/a n/a Means with different lower case letters are not significantly different (LSMEANS, P < 0.05)

PAGE 144

144 Table 6 7. Mean ( SE) waveform durat ion per insect (WDI) (s) for Diaphorina citri feeding on chlo r pyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetrama t treated and untreated plants 28 d after treatment (DAT) 4th Week Waveforms z np C G Treatment WDI SE WDI SE WDI SE WDI SE Untreated control 76.99 29.76 b 186.01 20.64 a 95.43 21.00 a b 43.03 5.87 ab Chlorpyrifos 82.47 28.48 b 154.25 17.71 a 81.50 14.90 a b 72.50 23.55 ab Fenpropathrin 110.25 45.07 b 200.34 36.55 a 81. 61 21.07 b 68.31 34.75 a Imidacloprid 243.31 29.22 a 78.32 17.48 a 54.26 19.94 c 9.72 n/a b Spinetoram 106.87 35.37 b 176.71 22.95 b 103.44 22.78 a b 61.75 17.25 ab Spirotetramat 53.89 35.65 b 146.20 22.32 a 161.23 23.85 a 59.48 11.21 a a Means with different lower case letters are not significantly different (LSMEANS, P < 0.05) Table 6 7. Continued Waveforms D E1 E2 Treatment WDI SE WDI SE WDI SE Untreated control 1.68 0.61 a 1.06 0.66 a 84.91 n/a Chl orpyrifos 3.19 1.20 a 3.75 1.83 a 96.54 6.55 Fenpropathrin n/a n/a n/a n/a n/a n/a n/a n/a Imidacloprid 1.81 0.06 a 2.20 0.51 a 105.94 44.98 Spinetoram n/a n/a n/a n/a n/a n/a n/a n/a Spirotetramat 0.19 n/a b 305.70 n/a a 20 88.00 n/a a Means with different lower case letters are not significantly different (LSMEANS, P < 0.05)

PAGE 145

145 Figure 6 1. Mean probe duration per insect (PDI) and the mean number of probes per insect (NPI) for Diaphorina citri feeding on chlo r pyrifos, f enpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated citrus plants through 28 d after treatment ( DAT ) 0 50 100 150 200 250 1 7 14 21 28 Duration (min) DAT PDI Control Chlorpyrifos Fenpropathrin Imidacloprid Spinetoram Spirotetramat 0 10 20 30 40 50 60 70 1 7 14 21 28 Frequency DAT NPI Control Chlorpyrifos Fenpropathrin Imidacloprid Spinetoram Spirotetramat

PAGE 146

146 Figure 6 2. Total waveform duration (TWD) for Diaphorina citri feeding on chlo r pyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated citrus plants through 28 d after treatment ( DAT ) 0 500 1000 1500 2000 2500 3000 3500 1 7 14 21 28 Duration (s) DAT Waveform np 0 1000 2000 3000 4000 5000 1 7 14 21 28 Duration (s) DAT Waveform z Control Chorpyrifos Fenpropathrin Imidacloprid Spinetoram Spirotetramat 0 500 1000 1500 2000 2500 3000 1 7 14 21 28 Duration (s) DAT Waveform C 0 100 200 300 400 500 600 700 800 1 7 14 21 28 Duration (s) DAT Waveform G Control Chorpyrifos Fenpropathrin Imidacloprid Spinetoram Spirotetramat

PAGE 147

147 Figure 6 3. Total waveform duration (TWD) for Diaphorina citri feeding on chlo r pyrifos, fenpropathrin, imidacloprid, spinetoram, and spirotetramat treated and untreated citrus plants through 28 d after treatment ( DAT ) 0 2 4 6 8 10 12 14 1 7 14 21 28 Duration (s) DAT Waveform D Control Chorpyrifos Fenpropathrin Imidacloprid Spinetoram Spirotetramat 0 5 10 15 20 25 1 7 14 21 28 Duration (s) DAT Waveform E1 Control Chorpyrifos Fenpropathrin Imidacloprid Spinetoram Spirotetramat 0 100 200 300 400 500 1 7 14 21 28 Duration (s) DAT Waveform E2 Control Chorpyrifos Fenpropathrin Imidacloprid Spinetoram Spirotetramat

PAGE 148

148 Figure 6 4. Temperature relative humidity and insecticide concentrations through 28 d after treatment A) Temperature (max, average an d min) and relative humi dity (RH) inside of the cage. B) Temperature (max, average an d min) and relative humid ity (RH) outside of the cage. C) Insecticide residues 10 20 30 40 50 1 6 11 16 21 26 31 Temperature ( C) DAT Inside of the cage Min Temp Mean Temp Max Temp RH 10 20 30 40 50 1 6 11 16 21 26 31 Temperature ( C) DAT Outside of the cage Min Temp Mean Temp Max Temp RH 0 50 100 150 200 250 1 7 14 21 28 Insecticide concentration (ppm) DAT Chlorpyrifos Imidacloprid A B C

PAGE 149

149 CHAPTER 7 SUMMARY HLB) was first found in Florida the estimated per acre pest management costs for mature citrus groves have increased nearly 41% (Morris and Muraro 2008). These additional costs are due to scouting for HLB infected trees and an increased use of pesticides for controlling the disease vector, the Asian citrus psyllid ( Diaphorina citri Kuwayama). The HLB associated pathogen ( Candidatus Liberibacter asiaticus) is a phloem limited bacterium and is transmitted during the feeding activities of D. citri Consequently, studies designed to gain a better understanding of psyllid feeding behavior are justi fied In th e research reported here an electrical penetration graph (EPG) monitor was used to record psyllid feeding activities in order to de termine the effects of gender, presence of light and insecticides on D. citri feeding behaviors. EPG was first de veloped by McLean (1964) and later modified by Tjallingii and Prado (2001) to allow quantification of the probing and feeding behaviors of certain sap sucking insects, thus making it possible to unravel the mechanisms underlying pathogen acquisition and tr ansmission (McLean 1964 and Tjallingii and Prado 2001). The first EPG studies with D. citri were cond ucted by Bonani et al. (2010). In those studies, the basic Asian citrus psyllid feedin g behaviors were characterized. Bonani et al. (2010) also was able to correlate phloem ingestion (Waveform E2) by D. citri with acquisition of Candidatus Liberibacter asiaticus However, bacterial inoculation could not be correlated with phloem penetration or salivation (waveforms D and E 1 respectively) but there was a str ong indication that this might be happening during either of these feeding behaviors. In C hapter 2 of the current research, the effects of gender and photoperiod on D. citri feeding behavior were investigated. When comparing the feeding behaviors of D. ci tri between

PAGE 150

150 genders, males reached the phloem (waveform D) 20% more often than females. However, when females were successful in penetrating the phloem, the duration of phloem ingestion (waveform E2) was significantly longer for females when compared to ma les. One possible explanation for female D. citri performing more phloem ingestion than males is the physiological needs of females that are related to reproduction. In other insects, it is common for females to feed more frequently during their reproduct ive period (Nation 2002). D. citri is sexually mature 2 3 days after emergence, therefore, it is likely that most (if not all) female psyllids used in this st udy were reproductively viable. Whether longer durations of phloem feeding behaviors will result i n increased pathogen acquisition and/or inoculation remain uncertain. In studies done with Frankliniella occidentalis male thrips were responsible for higher rates of pathogen transmission despite the fact that female thrips fed for longer durations of ti me (Wetering et al. 1998). Consequently, conclusions regarding the transmission efficiency of male versus female D. citri could not be made since the definitive studies of transmission using psyllids containing bacteria were not conducted. The e ffect of li ght presence on D. citri feeding behavior was also presented in C hapter 2. Results showed that non probing activities (waveform z and np), phloem penetration and salivation (waveform D and E1, respectively), and xylem ingestion (waveform G) were generally longer in duration during the light Conversely, stylet penetration (waveform C) and phloem ingestion (waveform E2) were longer in duration during the dark Within treatment analysis indicated some effects of light on xylem ingestion (waveform G) and phloe m ingestion (waveform E2) with phloem ingestion waveforms being longer in duration during the dark. However, there was no significant difference in the presence of light Since D. citri is considered to be a phloem feeder, sugar and amino acid content are likely the main phagostimulants

PAGE 151

151 regulating psyllid feeding activities. Differences found in D. citri feeding between light and dark periods could be related to changes in nutrient availability in the plant. Studies with Bucephalogonia xanthophis (Miranda e t al. 2008), and other sharpshooters (Andersen et al. 1989, 1992 and Brodbeck et al.1993) have shown that those insects are able to synchronize their feeding behavior according to the fluctu ation in xylem fluid chemistry. Goldschimitdt and Koch (1996) show ed slight daily fluctuations of soluble sugars and starch levels in citrus leaves, such that soluble sugars and starch levels were higher during the day and lower during the evening. Since soluble sugars from the leaves are transported through the phloem, sugar concentrations in the phloem are negatively inverted, and thus are higher during the night and lower during the day. Reduction in feeding behaviors during light could also be a result of increases in other behaviors that must occur during the photop hase thus leaving less time f or feeding behaviors to occur. For example, D. citri perform higher mating, oviposition and dispersal behaviors during the photophase (Wenninger and Hall 2007). These biological attributes of D. citri could explain the longer d urations and increased frequency of walking and standing still during the light and longer phloem ingestion during the dark Chapters 3 6 examine the effects of soil applied systemic and broad spectrum foliar applied insecticides on the D. citri feeding be havior. Since some insecticides have been shown to increase pathogen transmission rates in some pathosystems (Joost and Riley 2005) it is therefore important to know what effects insecticide applications will have on D. citri and subsequent potential tran smission of Las. More specifically, can insecticide applications be used to disrupt psyllid feeding behaviors associated with pathogen acquisition and inoculation prior to

PAGE 152

152 insecticide induced mortality ? Also, how long will such effects last, and are there any unintended consequences such as an increased potential for bacterial transmission? The effect of the soil applied imidacloprid on D. citri feeding behavior is presented in C hapter 3. Overall, the general feeding behavior of D. citri was disrupted on i midacloprid treated plants as demonstrated by an increased duration and number of times in which the insect st ood still or attempt ed to jump off plant s a decrease in walking and searching behaviors, a reduction in the number of probes, and a reduction on mean durations per event at the probe and insect level. While all D. citri tested on imidacloprid treated plants died during the course of the 12 h EPG recording due to imidacloprid exposure, D. citri feeding on young leaves of imidacloprid treated plants took on average 4 h (6 h for mature leaves) to die. Consequently, psyllids had an average feeding access period of 4 h before succumbing to the pesticide treatment. During this feeding access period, some of the psyllids tested were able to perform phloem related behaviors (waveforms D, E1, and E2). In addition, compared to feeding on mature leaves, D. citri on young citrus leaves performed a numerically higher number of probes which were also longer in duration on both imidacloprid treated and untreated pl ants. T his result is probably related to leaf tenderness, which may influence the ease of successful stylet penetration. Results from this study demonstrate that the benefits of soil applied imidacloprid applications extend beyond reducing the overall psyl lid population. Prior to causing mortality, imidacloprid application can potentially disrupt the psyllid feeding process such that successful pathogen acquisition and inoculation are both less likely to occur. It was also found that, on rare occasions, the probability of acquisition of Las could actually increase as a result of slight increases in phloem ingestion behaviors (waveform E2). However, this may not matter epidemiologically due to the pathogen latency

PAGE 153

153 period in the insect coupled with the fact th at psyllids would be dead before being capable of successfully inoculating a healthy tree with the pathogen. In C hapter 4, the effects of another systemic insecticide, aldicarb (Temik 15 G), on psyllid feeding behavior was investigated. Aldicarb is a soil applied systemic carbamate insecticide that has been widely used in Florida commercial citrus production for psyllid control. However, one noticeable difference in terms of control provided compared to soil applied imidacloprid is that the mortality of adu lt psyllids resulting from exposure to aldicarb is not as rapid. The overall results of this study showed no significant reduction in adult D. citri feeding behavior between aldicarb treated and untreated plants when data were analyzed at both the insect a nd probe levels. Analysis of walking (np), stylet penetration (C), phloem penetration (D) and phloem salivation (E1) behaviors were slightly reduced and incidence of insects standing still (z), phloem ingestion (E2) and xylem ingestion (G) were slightly in creased on aldicarb treated plants. However, significant differences were found at the event level, in which waveform duration per event (WDE) was higher for the phloem related activities of salivation and ingestion on aldicarb treated plants when compared to controls. Thus, D. citri on aldicarb treated plants ingested from and salivated into phloem for longer durations per event when compared to psyllids on untreated plants indicating a potentially negative effect of aldicarb application. In other words, t hese results suggest that aldicarb applications could lead to increased pathogen transmission rates. While aldicarb does provide long term reduction in psyllid populations through control of developing nymphal populations, these results point to the need f or additional use of foliar insecticide applications targeting adult psyllids to offset these negative effects. The effect of broad spectrum foliar applied insecticides and their residual activity on the feeding behavior of D. citri was studied in C hap te rs 5 and 6, respectively. In C hapter 5,

PAGE 154

154 differences in feeding disruption provided by foliar applied insecticide with differing modes of action was found. Overall, D. citri on chlorpyrifos, fenpropathrin, and imidacloprid treated plants had a reduction in the probe duration, number of probes, walking activities, stylet penetration, phloem activities and xylem ingestion, while psyllids on spinetoram treated plants only had reductions in the probe duration, walking activities, stylet penetration, and xylem in gestion behaviors. However, psyllids on spinetoram treated plants were still able to reach the phloem. In contrast, there was no effect of spirotetramat application on any D. citri feeding behaviors when compared with untreated plants. For each of the ins ecticides evaluated, the more important differences noted were in the level of feeding disruption provided as follows: D. citri exposed to chlorpyrifos treated plants took an average 4.02 h to achieve 100% mortality with probing behaviors performed an aver age of 0.88 h per insect. Psyllids feeding on fenpropathrin treated plants took an average of 0.62 h to die with an average probing duration of only 0.03 h per insect. Similar results were obtained for psyllids exposed to foliar applied imidacloprid treate d plants, in which average time to mortality was 1.41 h and the average probing duration was reduced to 0.21 h per insect. Psyllids on spinoteram treated plants probed an average of 1.20 h per insect and required 7.31 h to achieve 100% mortality. For spiro tetramat treated plants, there was no resulting psyllid mortality during the 12 h of recording and there was also no discernable effects on probing duration which averaged 7.26 h per insect. The differences found in psyllid feeding behavior on plants trea ted with these various insecticides is likely due to the manner in which these different modes of action affect the insect to cause mortality. Chlorpyrifos, fenpropathrin, imidacloprid and spinetoram act primarily as a contact insecticides inhibiting the i

PAGE 155

155 sodium channel modulators, acetylcholine receptor stimulator, and disruption of nicotinic/gamma amino butyric acid (GABA) gated chloride channels, respectively). Spirotetramat, on the other hand, has limited contact activity and is mainly effective following ingestion by the insect (Nauen et al. 2008). Spirotetramat is a tetramic acid that acts as a lipid (N auen et al. 2008). The results from this study demonstrate that certain foliar applied insecticides may have important benefits in terms of reducing the spread of HLB that extended beyond just reducing the overall population levels of the insect vector. P rior to causing mortality, chlorpyrifos, fenpropathrin, imidacloprid and spinetoram application can potentially disrupt the psyllid feeding process such that successful pathogen acquisition and inoculation are both less likely to occur. It was also found t hat psyllids on spinetoram treated plants reach the phloem of the citrus plants prior to mortality, albeit, stylet penetrations were abnormal indicating that some feeding disruption might still be occurring. Spirotetramat was the only treatment that did no t affect psyllid feeding behaviors. Even though this insecticide has been shown to reduce psyllid populations under field conditions via effects on immature stages, infected adult psyllids that land on a healthy plant to feed may still successfully inocula te that plant with the pathogen. Thus, similar to aldicarb as discussed above, citrus growers using spirotetramat should consider combining this insecticide with another that will provide the desired level of adult psyllid control. In C hapter 6, the resid ual effects, or the duration of feeding disruption provided by the insecticides evaluated in C hapter 5, were examined. For chlorpyrifos, at 1DAT, while there was no significant difference in probing duration, psyllids on chlorpyrifos performed half of the

PAGE 156

156 probing durations when compared to untreated plants. Phloem related feeding behaviors were prevented up to 7 DAT on chlorpyrifos treated plants when chlorpyrifos residues on the leaf surface averaged 27.13 ppm. For both fenpropathrin and imidacloprid treat ed plants, there was an immediate reduction 1 DAT in D. citri probe duration, number of probes, non probing/walking activities, stylet penetration, phloem activities and xylem ingestion when compared to untreated plants. Over time, the duration of those be haviors increased as insecticide residue levels decreased. Fenpropathrin residues provided significant disruption of phloem related feeding behaviors by D. citri up to 21 DAT. Disruption of phloem feeding behaviors for psyllids on imidacloprid treated plan ts lasted up to 28 DAT. At this time imidacloprid residue levels were 34.66 ppm, which was 83% lower than the concentration found at 1DAT. Psyllids on spinetoram treated plants performed probes which were shorter in duration, but disruption of phloem feedi ng behaviors was not prevented, even at 1 DAT. The decrease in probing duration for psyllids on spinetoram treated plants was only significant 7 DAT. Psyllids on spirotetramat treated plants performed normal phloem activities throughout the entire experime nt. The results from this study demonstrate that while some insecticides may cause relatively rapid mortality of adult psyllids, there is considerable variability that exists among these products in terms of the duration of feeding disruption provided. Whi le some insecticides provided feeding disruption lasting 3 4 weeks (fenpropathrin and imidacloprid), protection provided by other products was much shorter. Overall, the results of this study can be used to help guide citrus growers in product selection an d also determine when additional applications will be necessary. In conclusion, utilizing an electrical penetration graph monitor to study psyllid feeding behavior, we have a better understanding of how certain aspects of psyllid biology (gender) and env ironment (photoperiod) influence D. citri feeding behavior. These two factors have been

PAGE 157

157 shown to be important variables which could affect the outcome of experiments investigating the transmission of Candidatus Liberibacter asiaticus by D. citri Future st udies by researchers dealing with pathogen transmission should be aware of and account for these factors which could confound the results of planned experiments. Unaccounted for, these factors may have played a role in reported inconsistencies in past stud ies of psyllid transmission of the HLB associated bacterium With regards to managing the spread of HLB, this work sheds new light on the benefits provided by insecticides commonly used for psyllid control. We have shown that such benefits go beyond just r educing overall psyllid populations, but certain products may actually prevent bacterial transmission (acquisition and or inoculation) for an extended period of time via disruption of psyllid phloem feeding behaviors. Additionally, it was shown that certai n insecticides that have been commonly relied upon as stand alone psyllid control products may not have provided as much reduction in pathogen spread as desired or perhaps in some cases, enhanced pathogen spread by increasing certain psyllid feeding behavi ors. This new information should be useful in helping citrus growers refine their current HLB/psyllid management programs based on a better understanding of how these insecticides affect psyllid feeding behavior.

PAGE 158

158 LIST OF REFERENCES Almeida, R. P. P. a nd E. A. Backus. 2004. Stylet penetration behaviors of Graphocephala atropunctata (Signoret) (Hemiptera, Cicadellidae): EPG waveform characterization and quantification. Ann. of the Entomol. Soc. of Am. 97 : 838 851. Andersen, P.C., B. V. Brodbeck, and R. F Mizell III. 1989. Metabolism of amino acids, organic acids and sugars extracted from the xylem fluid of four host plants by Homalodisca coagulat a Entomol. Exp. Appl. 50: 149 159 Andersen, P. C., R. F. Mizell III, B. V. Brodbeck, T. G. Beckman, and G. K rewer. 2008. Abundance and consumption rate of glassy winged sharpshooter (Hem iptera: Cicadellidae) on peache s and plums. J. Entomol. Sci. 43:394 407. Aubert, B. 1987. Trioza Erytreae (Del Guercio) and Diaphorina Citri Kuwayama (Homoptera, Psylloidea), the 2 Vectors of Citrus Greening Disease Biological Aspects and Possible Control Strategies. Fruits 42 : 149 162. Backus, E. A. 1988. Sensory systems and behaviours which mediate hemipteran plant feeding: a taxonomic overview. J. Insect Physiol. 34: 151 165. Backus, E. A. 2000. Our own jabberwocky: clarifying the terminology of certain piercing sucking behaviors of homopterans, pp. 1 13. In G. P. Walker and E. A. Backus [eds.], Principles and applications of electronic monitoring and other techniques in the s tudy of homopteran feeding behavior. Thomas Say Publications in Entomology, Entomological Society of America, Lanham, MD. B ackus, E. A., J. Habibi, F. Yan, M. Ellersieck. 2005. Stylet p enetration by a dult Homalodisca coagulata on g rape: e lectrical p enetrat ion g raph w aveform c haracterization, t issue c orrelation, and p ossible i mplications for t ransmission of Xylella fastidiosa Ann. Entomol. Soc. Am. 98 (6): 787 813. Backus, E. A., A. R. Cline, M. R. Ellerseick, and M. S. Serrano. 2007. Lygus hesperus (Hemipte ra: Miridae) feeding on cotton: New methods and parameters for analysis of nonsequential electrical penetration graph data. Ann. Entomol. Soc. Am. 100 : 296 310. Backus, E. A., W. J. Holmes, F. Schreiber, B.J. Reardon, and G. P. Walker. 2009. Sharpshooter X w ave: c orrelation of an Electrical Penetration Graph Waveform with Xylem Penetration Supports a Hypothesized Mechanism for Xylella fastidiosa Inoculation. Ann. Entomol. Soc. Am. 102 : 847 867. Balayannis P.G. 1983. Effect of soil applications of aldicarb on the growth, yield, and chemical composition of tobacco plants J. Agric. Food Chem. 31: 1351 1355. Baldessari, M., F. Trona, G. Angeli, and C. Ioriatti. 2010. Effectiveness of five insecticides for the control of adults and young stages of Cacopsylla melan oneura (Frster) (Hemiptera: Psyllidae) in a semi field trial. Pest. Manag. Sci. 66: 308 312.

PAGE 159

159 Beanland, L., C.W. Hoy, S.A. Miller, and L.R. Nault. 2000. Influence of aster yellows phytoplasma on the fitness of a ster leafhopper (Homoptera: Cicadellidae) Ann Entomol Soc. Am 93: 271 276. Bethke J.A., M. J. Blua, and R. A. Redak. 2001. Effect of selected insecticides on Homalodisca coagulata (Homoptera: Cicadellidae) and transmission of oleander leaf scorch in a greenhouse study. J. Econ. Entomol. 2001. 94:1 031 1036. Boina, D. R., E. O. Onagbola, M. Salyani, and L. L. Stelinski. 2009. Antifeedant and sublethal effects of imidacloprid on Asian citrus psyllid. Pest Manag. Sci. 65 :870 877. Bonani, J. P. 2009. Caracterizao do aparelho bucal e comportamento ali mentar de Diaphorina citri Kuwayama (Hemiptera: Psyllidae) em Citrus sinensis (L.) Osbeck. PhD dissertation. Escola Superior de Agricultura Luiz de Queiroz (ESALQ), Piracicaba SP, Brazil. Bonani, J. P., A. Fereres, E. Garzo, M. P. Miranda, B. Appezzato Da Gloria, and J. R. S. Lopes. 2010. Characterization of electrical penetration graphs of the Asian citrus psyllid, Diaphorina citri in sweet orange seedlings. Entomol. Exp. Appl. 134 : 35 49. Bove, J. M. 2006. Huanglongbing: A destructive, newly emerging, ce ntury old disease of citrus. J. Plant Pathol. 88 : 7 37. Brlansky, R.H. and Rogers, M.E. 2007. Citrus huanglongbing: understanding the vector pathogen interaction for disease management. APSnet. http://www.apsnet.org/publications/apsnetfeatures/Pages/Huanglongbing.aspx Brodbeck, B.V., R.F., Mizell III, and P.C.Andersen, 1993. Physiological and Behavioral Adaptations of Three Species of Leafhoppers in Response to the Dil ute Nutrient Content of Xyl em Fluid. J. of Insect Physiol. 39: 73 81 Brck, E., A. Elbert, R. Fischer, S. Krueger, J. Khnhold, A. M. Klueken, R. Nauen, J.F. Niebes, U. Reckmann, H.J. Schnorbach, R. Steffens, and X. van Waetermeulen. 2009. Movento, an inn ovative ambimobile insecticide for sucking insect pest control in agriculture: Biological profile a nd filed performance. Crop Prot 28:838 844. Buitendag, C. H., and L. A. Von Broembsen. 1993. Living with citrus greening in South Africa, pp. 269 273 In P. Moreno, J. V. da Graa, and L. W. Timmer [eds.], Proc. 12th Conference of the International Organization of Citrus Virologists. University of California, Riverside. Capoor, S. P., D. G. Rao, and S.M. Viswanath. 1974. Greening disease of citrus in the Decc an Trap Country and its relationship with the vector, Diaphorina citri Kuwayama." Proceedings of the Sixth Conference of the International Orga nization of Citrus Virologists. 43 49. Catling, H. D. 1969. The control of citrus psylla, Trioza erytreae del Gue rcio (Homoptera:Psyl lidae). S. Afr. Citrus J 426:9 16.

PAGE 160

160 Catling, H. D. 1970. Distribution of the p syllid vectors of c itrus greening disease, with notes on the biology and bionomics of Diaphorina citri F.A.O. Plant Protection Bulletin. 18 : 8 15. Caciagli, P., D. Bosco, and L. Al Bit ar. 1995. Relationships of the s ardinian isolate of tomato yellow leaf curl geminivirus with its whitefly vector Bemisia tabaci Gen. Eur. J. Plant Pathol. 101: 163 170. Casida, J.E., and G.B. Quistad. 2004. Why insecticides are mo re toxic to insects than people: the unique toxicology of insects. J Pestic. Sci. 29: 81 86. Childers, C.C, and M. E. Rodgers. 2005. Chemical control and management approaches of the asian citrus psyllid, Diaphorina citri Kuwayama (homoptera: psyllidae) i n Florida citrus. Proc. Fla. State Hort. Soc. 118:49 53. Chisholm, I. F., and T. Lewis, 1984. A new look at thrips (Thysanoptera) mouthparts, their action and effects of feeding on plant tissue. B Entomol Res 74: 663 675. Chung, K. and X. C. Fan. 1990. Successful integrated management of huanlungbin disease in several farms of guangodng and Fujian, by combining early eradication with target insecticide sprayings. pp. 145 148. In Proceedings, 4 th International Asia Pacific Conference on Citrus Rehabilitat ion, Chian Mai, Thailand. Collar, J.L., C. Avillar, and A. Fereres. 1997a. New correlations between aphid stylet paths and nonpersisten virus transmission. Environ. Entomol. 26: 537 544. Collar, J.L., C. Avillar, M. Duque, and A. Fereres. 1997b. Behavioral response and virus vector a bility of Myzus persicae (Homoptera: Aphididae) probing on pepper plants treated with a phicides. J. of Econ. Entomol. 6: 1628 1634. Da Graca, J., and L. Korsten. 2004. Citrus huanglongbing: review, present status and future stra tegies. Dis Fruits veg. 1: 229 245. Edwards, C.A. 1975. Factors that affect the persistence of pesticides in plants and soils. Pure Appl. Chem. 42:39 56. El Hamady, S. E., R. Kubiak, and A.S. Derbalah. 2008. Fate of imidacloprid in soil and plant after ap plication to cotton seeds. Chemosphere 71 : 2173 2179. EPA. 2010. Pesticides registration: agreement to terminate all uses of aldicarb. http://www.epa.gov/oppsrrd1/REDs/factsheets/ald icarb_fs.html Fereres, A. and J. L. Collar. 2001. Analysis of noncirculative transmission by electrical penetration graphs. pp. 87 109. In: F. K. Harris, O. P. Smith and J. E. Duffus Virus insect plant interactions, Academic press San Diego, CA Goldsch midt E.E. & Koch E.K. 1996. Citrus pp. 797 823. In E. Zamski & A. A.Scaffer. Photoassimilate Di stribution in Plants and Crops. Marcel Dekker, New York, NY.

PAGE 161

161 Gottwald, T. R., da Graa, J. V., and R. B. Bassanezi. 2007. Citrus Huanglongbing: The pathogen and its impact. Online. Plant Health Progress doi:10.1094/PHP 2007 0906 01 RV. http://www.apsnet.org/publications/apsnetfeatures/Pages/HuanglongbingImpact.aspx Graf ton Cardwell1, E.E., Godfrey, K.E., Rogers, M.E., Childers, C.C. and Stansly, P.A., 2005. Asian citrus psyllid. http://ccpp.ucr.edu/news/PsyllidbrochureAug05.pdf Groves, R.L., C.E. Sorenson, J .F., Walgenbach, and G.G. Kennedy. 2001. Effects of imidacloprid on transmission of tomato spotted wilt tospovirus to pepper, tomato, and tobacco by Frankliniella fusca Hinds (Thysanoptera: Thripidae). Crop Prot. 20:439 445. Halbert, S. E. 1998. Entomology Section. Tri ology. 37: 6 7. Halbert, S.E. 2005. Pest alert: citrus greening/Huanglongbing. Florida Department of Agriculture and Consumer Services. Division of Plant Industry. ( http://www.doacs.state.fl.us/pi/chrp/greening/citrusgreeninalert.htlm ) Halbert, S. E. and K. L. Manjunath. 2004. Asian citrus psyllids (Sternorrhyncha : Psyllidae) and greening disease of citrus: A literature review and assessment of risk in Florida Fla. Entomol. 87 : 330 353 Harrewijn, P. and H Kayser. 1997. Pymetrozine, a fast acting and selective i nhibitor o f aphid feeding i n situ studies with electronic monitoring of feeding b ehavior Pestic. Sci. 49:130 140. Hunter, W. A., and D. E. Ullman, 198 9. Analysis of mouthpart movement during feeding of Frankliniella occidentalis (Pergande) and F. schultzei Trybom (Thysanoptera: Thripidae). Int J of Insect Morphol 18: 161 171. Inoue, H., and J. Hirao. 1978. Transmission efficiency of rice waika virus by the green rice leafhoppers, Nephotettix spp (Hemipter a: Cicadellidae). Appl. Entomol Zool 13:264 2733. Jiang, Y. X., C. de Blas, L. Barrios, and A. Fereres. 2000. Correlation between whitefly (Homoptera: Aleyrodidae) feeding behavior and transmission of tomato yellow leaf curl virus. Ann. Entomol. Soc. Am. 93:573 579. Jo, C. W. C. R. Park, K. S. Yoon, M. A. Kang, H. R. Kwon, H.B. Seok, E. J. Kang, M J. Seo, Y. Y. Yu, and Y. N. Yo un. 2009. Feeding behavior of etofenprox resistant gree n peach aphid ( Myzus persicae ) against thiamethoxam and fenpropathrin. Korean J. Appl. Entomol. 48: 493 501. Joost, P.H. and D. G. Riley. 2005. Imidacloprid effects on probing and settling behavior of Frankliniella fusca and Frankliniella occidentalis (Thysanoptera: Thripidae) in tomato. Hortic. E ntomol.. 98:1622 1629. Laszlo, P. 2007. Citrus: A History, University of Chicago Press, Chicago, IL.

PAGE 162

162 Li, W. B., J. S. Hartung, L Levy. 2006. Quantitative re al time PCR for detection and identification of Candidatus Liberibacter species associated with citrus huanglongbing. J.of Microbiol. Meth. 66 : 104 115. Lima. A. M. C. da. 1942. Insetos do Brasil: Hompteros Rio de Janeiro: Escola Nacional de Agronomia, v .3, p.1 327. Liu, D. G. and J. T. Trumble. 2004. Tomato psyllid behavioral responses to tomato plant lines and interactions of plant lines with insecticides. J. Econ. Entomol. 97 : 1078 1085. Liu, Y.H. and J.H. Tsai. 2000. Effects of temperature on biology and life table parameters of the Asian citrus psyllid, Diaphorina citri Kuwayama (Homoptera: Psyllidae). Ann. Appl. Biol. 137:201 206. Lowery, D.T. and G. Boiteau. 1988. Effects of five insecticides on the probing, walking, and settling behavior of the gre en peach aphid and the buckthorn aphid (Homoptera: Aphididae) on potato. J. Econ. Entomol. 181:208 214. Mahmoud, M. F., G. A. EI Kady, M.A.M. Osman, and I.M. Bahgat. 2009. Efficiency of spinetoram as a biopesticide to onion thrips ( Thrips tabaci Lindeman) and green peach aphid ( Myzus persicae Sulzer) under laboratory and field contidions. J. Biopest. 2: 223 227. McLean, D. L. and M. G. Kinsey. 1964. A technique for electronically recording aphid feeding and salivation. Nature. 202 : 2 Mead, F. W. 1977. The As iatic citrus psyllid, Diaphorina citri Kuwayama (Homoptera: Psyllidae). Entomology Circular, Division of Plant Industry, Florida Department of Agriculture and Consumer Services. 180: 4. Mendel, R. M., U. Reckmann, and F. Fhr. 2000. Xylem transport of the pesticide imidacloprid in citrus. Acta Hort. (ISHS) 531:129 134 http://www.actahort.org/books/531/531_18.htm Miranda, M. P., D.N.Viola, R.N. Marques, J.P. Bonani and J.R.S. Lopes 2008. Feeding si tes and food intake of Bucephalogonia xanthophis (BERG) (Hemiptera: Cicadellidae), a sharpshooter vector of Xylella fastidioas on citrus plants. Rev. Bras. Frutic. 30: 913 918. Montemurro, N., F. Grieco, G.Lacertosa, and A. Visconti. 2002. Chlorpyrifos de cline curves and residue levels from different commercial formulations applied to oranges. J. Agicul and Food Chem. 50: 5975 5980. Montesino, L.H., J.H.C. Coelho M.R. Felippe, and P.T. Yamamoto. 2006. Xylem sap ingestion sweet Orange ( Citrus sinensis (L.) Osbeck) and infected ones by Xylella fastidiosa Oncometopia facialis and Dilobopterus costalimai (Hemiptera: Cicadellidae). Rev. Bras. Frutic. 28: 199 204.

PAGE 163

163 Morris, A. and R. Muraro. 2008. Economic Evaluation of Citrus Gr eening Man agement and Control Strategies. Extension Digital Information Source (EDIS) FE 712, Food and Resource Economics Department, University of Florida, June, 2008. http://edis.ifas.ufl.edu/FE712 Nation, J .L., 2002. Insect Physiology and Biochemistry. CRC Press, Boca Raton. Nauen, R., U. Reckmann, S. Armborst, H.P. Stupp, and A. Elbert. 1999. Whitefly active metabolites of imidacloprid: biological efficacy and translocation in cotton plants. Pestic. Sci. 55 :265 271. Nauen, R. U., Reckmann, J. Thomzik, and W. Thielert. 2008. Biological profile of spirotetramat (Movento) a new two way systemic (ambimobile) insecticide against sucking pest species. Pflanzenschutz Nachrichten Bayer 61: 403 436. Olson, E. R., G.P Dively, and J. O. Nelson. 2004. Bioassay determination of the distribution of imidacloprid in potato plants: implications to resistance development. J. Econ. Entomol. 97:614 620. Pelz Stelinski, K. S., R. H. Brlansky, T. E. Ebert and M. E. Rogers. 2010. Transmission parametersfor Candidatus Liberibac t er asiaticus by Asian citrus psyllid (Hemiptera: Psyllidae). J. Econ. Entomol. 103:1531 1541. Pfeiffer, D. G., and E. C. Burts. 1984. Effect of tree fertilization on protein and free amino acid content and fe eding rate of pear psylla (Hom o ptera: Psyllidae). Environ. Entomol. 6:1487 1490. Perring, T. M., N. M. Gruenhagen, and C. A. Farrar. 1999. Management of plant viral diseases through chemical control of insect vectors. Annu. Rev. Entomol. 44:457 481. Polek M., G. Vidalakis, and K. Godfrey. 2007. Citrus bacterial canker disease and huanglongbing (Citrus greening).Publication 8218. University of California Agriculture and Natural Resources. Oakland, California ( http://citrusent.uckac.edu/8218CitrusCankerandHuanglongbing.pdf ) Pollak S.L., and Perez, A. 2008. A report from the economic research service: Fruit and tree nut outlook. USDA. ( http://www.ers.usda.gov/publications/fts/2008/03MAR/FTS331.pdf ). Accessed on October 16 th 2008. Powell, G. 1991. Cell membrane punctures during epidermal penetrations by aphids: consequences for the transmission of two potyv iruses. Ann. Appl. Biol. 119:313 321. Power, A. G. 1991. Virus spread and vector dynamics in genetically diverse plant populations. Ecology 72:232 241. Prado, E. and W. F. Tjallingii. 1994. Aphid activities during sieve element punctures. Entomol. Exp. App l. 72: 157:165.

PAGE 164

164 Purcell, A. H. 1982. Insect vector relationships with prokaryotic plant pathogens. Ann u. Rev. Phytopathol. 20:397 417. Qureshi, J. A. and P. A. Stansly. 2008. Rate, placement and timing of aldicarb applications to control Asian citrus psyll id, Diaphorina citri Kuwayama (Hemiptera: Psyllidae), in oranges. Pest Manag. Sci. 64 : 1159 1169. Qureshi, J.A., M E. Rodgers, D. G. Hall, and P. A. Stansly. 2009. Incidence of Invasive Diaphorina citri (Hemiptera: Psyllidae) and Its Introduced Parasitoid Tamarixia J. Econ. Entomol. 102:247 256. Qureshi, J.A., and P. A. Stansly. 2010. Dormant season foliar sprays of broad spectrum insecticides: An effective component of integrated management for Diaphorina citri (Hemiptera:Psyllidae) in citrus orchards. Cr op. Prot. 29, 860 866. Ragab S.M. 1981 Cotton growth and nutrient uptake following Temik (aldicarb) application. J. Agr. Sci. 97: 731 737. Reese, J. C., W. F. Tjallingii, M. van Helden, and E. Prado 2000. Waveform comparisons among AC and DC eletronic m onitoring systems for aphid (Homopetra: Aphididae) feeding behavior. 70 101. In G. P. Walker and E. A. Backus Principles and aplications of electronic monitoring and other techniques in the study of homopteran feeding behavior. Thomas Say Pubs. in Entomol ogy. Entomological society of A merica. Roberts, J. M. F., C. J. Hodgson, L. E. N. Jackai, G. Thottappilly, and S.R. Singh. 1993. Interactions between two synthetic pyrethroids and the spread of two non persistent viruses in cowpea. Ann. Appl. Biol. 122:57 67. Rogers, M. E., P. A. Stansly, and L. L. Stelinski. 2010. 2010 Florida citrus pest management guide: Asian citrus psyllid and citrus leafminer. ENY 734. IFAS extension. Rogers, M. E., P. A. Stansly, and L. L. Stelinski. 2011. 2011 Florida citrus pest m anagement guide: Asian citrus psyllid and citrus leafminer. ENY 734. IFAS extension. Roistacher, C. N. 1991. Techniques for biological detection of specific citrus graft transmissible diseases. pp. 35 45. Greening. Food Agric. Organ., Rome. SAS Institute. 2001. PROC user's manual, version 6th ed. SAS Institute, Cary, NC. Serikawa, R. H., M. E. Rogers, and E.A. Backus, in press. Effects of soil applied imidacloprid on Asian citrus psyllid (Hemiptera: Psyllidae) feeding behavior. J Econ Entomol. Serrano, M S., E. A Backus, and C. Cardona 2000. Comparison o f AC electronic monitoring and h eld data for estimating tolerance to Empoasca kraemeri (Homoptera: Cicadellidae) in common bean genotypes. J. Econ. Entomol. 93: 1796 1809.

PAGE 165

165 Spann, T.M., R.A. Atwood, M.M. Dewdney, R.C. Ebel, R. Ehsani G. England, S. Futch, T. Gaver, T. Hurner, C. Oswalt, M.E. Rogers, F. M. Roka, M. A. Ritenour and M. Zekri. 2010. IFAS guidance for huanglongbing. Citrus Industry. http://www.crec.ifas.ufl.edu/academics/faculty/rogers/PDF/Spann%20et%20al%202010. pdf Steer, B. T. 1973. Diurnal variations in photosynthetic products and nitrogen metabolism in expanding leaves. Plant Physiol. 51: 74 4 748. Sur. R., and A. Stork. 2003. Uptake, translocation and metabolism of imidacloprid in plants. B. Insectol. 56(1): 35 40. Tiwari, S., K. Pelz Stelinski, and L. Stelinski. 2011 Effect of Candidatus Liberibacter asiaticus i nfection on susceptibility of Asian citrus psyllid, Diaphorina citri, to selected insecticides. Pest. Manag. Sci. 67: 94 99 Tjallingii, W.F. 1978. Electronic recording of penetration behaviour by aphids. Entomol. Exp. Appl 24 : 721 730. Tjallingii, W. F. 2000. Comparison of AC and DC s ystems for electronic monitoring of stylet penetration activities by homopterans, pp. 41 69. In G. P. Walker and E. A. Backus [eds.], Principles and applications of electronic monitoring and other techniques in the study of homopteran feeding behavior. Ent omological Society of America, Lanham, MD. Tjallingii, W. F. and E. Prado. 2001. Analysis of circulative transmission by electrical penetration graphs. Pp. 69 85. Virus insect plant interactions F. K. Harris, O. P. Smith and J. E. Duffus. San Diego, Academic press UF/IFAS. 2008. Eradication. Impact of eradication. http://www.crec.ifas.ufl.edu/extension/canker/eradication .htm Ullman, D. E. and D. L. Mclean. 1988. The p robing b ehavior of the s ummer f orm p ear p sylla. Entomol. Exp. Appl. 47 : 115 125. USDA. 2010a. Citrus Fruits: 2010 summary. http://usda.mannlib.cornell.edu/usda/current/CitrFrui/CitrFrui 09 23 2010.pdf USDA. 2010b. Florida Citrus statistics 2009 2010. http://www. nass.usda.gov/Statistics_by_State/Florida/Publications/Citrus/fcs/2009 10/fcs0910.pdf USDA. 2011 a. Citrus: World Markets and Trade. Citrus: 2010/11 Forecast. http:/ /usda.mannlib.cornell.edu/usda/current/citruswm/citruswm 01 27 2011.pdf USDA. 2011 b. National quarantine citrus greening and Asian citrus psyllid. http://www.aphis.usda.gov/plant_health/plant_pest_info/citrus_greening/downloads/pdf_ files/nationalquarantinemap.pdf

PAGE 166

166 Van Den Berg, M.A., Dippenaar Schoeman, A.S., Deacon, V.E., and S.H., Anderson, 1992. Interactions between the citru s psylla, Trioza erytreae (Hemiptera: Triozidae), and spiders in an unsprayed citrus orchard in the Transvaal Lowveld. Entomophaga 37 : 599 608. Van Rensburg, G. D. J., and J.H. Giliomee. 1990. Comparison of females and males of Cicadulina anestae and C. mb ila (Homoptera: Cicadellidae) as vectors of maize streak virus. Phytophylactica 22 : 241 243. Van Vuuren, S. P. v. and M. J. v. d. Merwe. 1992. Efficacy of citrus psylla, Tr ioza erytreae as vector of citrus greening disease. Phytophylactica 24 : 285 288. Walker, G. P. 2000. A beginner guide to electronic monitoring of homopteran feeding behavior, pp. 14 20. In G. P. Walker and E. A. Backus [eds.], Principles and application s of electronic monitoring and other techniques in the study of homopteran feeding behavior. Thomas Say Publications in Entomology, Entomological Society of America, Lanham, MD. Ware, G.W. 1994. Chapter 4: Insecticides. pp. 41 74 In G .W.Ware The Pesticid e Book, Fourth Edition Thomson Publications, Fresno C A Weintraub, P. G., and L. Beanland. 2006. Insect vectors of phytoplasmas. An nu Rev.of Entomol. 51: 91 111. Wenninger, E. J. and D. G. Hall. 2007. Daily timing of mating and age at reproductive matur ity in Diaphorina citri (Hemiptera: Psyllidae). Fl a Entomol. 90 : 715 722. Wenninger, E. J., L. L. Stelinski, and D. G. Hall. 2008. Behavioral evidence for a female produced sex attractant in Diaphorina citri Kuwayama (Hemiptera: Psyllidae). Entomol. Exp. Appl. 128: 450 459. Wenninger, E.J., L.L. Stelinski, and D.G. Hall. 2009. Roles of olfactory cues, visual cues, and mating status in o rientation of Diaphorina citri Kuwa yama (Hemiptera: Psyllidae) to four d ifferent host p lants. Environ Entomol.38: 225 234 Wetering, F. van, Hulshof, J., Posthuma, K., Harrewijn, P., Goldbach, R., and D. Peters, 1998. Distinct feeding behavior between sexes of Frankliniella occidentalis results in higher scar production and lower tospovirus transmission by females. Entomol. Exp. Appl. 88: 9 15. Wheaton, T. A., C. C. Childers, L. W. Timmer, L. W. Duncan, an S. Nikdel. 1985. Effects of aldicarb on yield, fruit quality, and tree condition of Florida citrus. Proc. Fla. State Hort. Soc 98:6 10. Xu, C. F., Y. Xia, K.B. Li, and C. Ke. 1988. Further study of the transmission of citrus Huanglungbin by a psyllid, Diaphorina citri Kuwayama. pp. 243 248. In Proceedings, 10th Conference of the International Organization of Citrus Virologists. IOCV. Riverside, CA.

PAGE 167

167 Ya ng, Y., M. Huang, G. An drew, G.A.C. Beattie, Y, Xia, G. Ouyang, and J. Xiong. 2006. Distribution, biology, ecology and control of the psyllid Diaphorina citri Kuwayama, a major pest of cit rus: a status report for China. Int J Pest Manag e. 52 (4): 343 352. Youn Y., E.A. Backus, R H. Serikawa, and L.L. Stelinski. in press. Correlation of an electrical penetration graph waveform with walking by asian citrus psyllid, Diaphorina citri kuwayama. Fla Entomol.

PAGE 168

168 BIOGRAPHICAL SKETCH Rosana Harumi Serikawa was born in So Carlos, SP Bra zil. Rosana was educated at Colgio So Carlos for elementary and middle school and Anglo for high school. Since she was always helping her mother in the chemistry lab, she got involved with several science projects in middle and high school. After a natio nal general exam she was accepted to the Universidade Estadual Paulista Faculdade de Cincias Agrrias e Veterinrias Jaboticabal, where she majored in agronomic engineering as an undergraduate. Her first trip to the United States was in 2003 when she was invited to do an internship at University of Nebraska Lincoln, which she continued on to complete her M.Sc. in Entomology under the advisement of John E. Foster, studying population genetics of sugar beet root aphids. She moved to Florida to pursue a Ph.D. under Michael E. Rogers, where her research was based on the feeding behavior of the Asian citrus psyllid. Upon the completion of her Ph.D. at the University of Florida, Rosana plans to accept an opportunity to work as Research Entomologist at DuPo nt Brazil.