The Distribution of, Relationship Between, and Factors Influencing the Abundance of Bemisia Tabaci and the Incidence of ...

MISSING IMAGE

Material Information

Title:
The Distribution of, Relationship Between, and Factors Influencing the Abundance of Bemisia Tabaci and the Incidence of Tomato Yellow Leaf Curl Virus in Southern Florida Tomato
Physical Description:
1 online resource (206 p.)
Language:
english
Creator:
Taylor,James Edwin
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:
Schuster, David J
Committee Members:
Webb, Susan E
Grunwald, Sabine
Polston, Jane E

Subjects

Subjects / Keywords:
geostatistics -- tylcv -- whitefly
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:
Biotype B of the sweetpotato whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae), also known as the silverleaf whitefly, B. argentifolii Bellows and Perring, is a serious pest of many agricultural crops around the world. In Florida, B. tabaci has become a limiting pest species in tomato due to its ability to vector Tomato yellow leaf curl virus (TYLCV) (family Geminiviridae, genus Begomovirus). TYLCV is vectored in a persistent circulative manner and symptoms of infection in tomato include upward curling of leaflet margins, reduction of leaflet area, yellowing of young leaves, abscission of flowers, and stunting of plants. The sampling of adult B. tabaci and TYLCV across commercial Florida tomato farms in four seasons from fall 2007 through spring 2009 combined with mapping of their distribution by Geographical Information Systems (GIS) and analyses by Spatial Analysis by Distance IndicEs (SADIE) and classification and regression tree (CART) analysis have produced a much more detailed explanation of in-field distribution, vector/disease relationship and influencing factors than previously reported. B. tabaci is a mobile pest and has been shown in the present study to have varying aggregation in both space and time. Distributions of B. tabaci were significantly aggregated in every season but spring 2009, when the population of whitefly was very low. Weekly fluctuations throughout the study area suggest that, within the earlier sampling dates of each season, whiteflies were more likely to be aggregated. Inverse distance weighted (IDW) maps created by a GIS program, showed that populations of B. tabaci and incidence of plants with symptoms of TYLCV infection were associated more closely with the edges of tomato fields. Early fall season and late spring season populations of adult B. tabaci had stronger correlations to incidence of symptomatic TYLCV infected plants. SADIE spatial association tests indicated similar conclusions. There were indications that B. tabaci may have migrated from areas in which possible whitefly hosts were destroyed or disturbed. CART analysis confirmed the assumptions that environmental variables such as temperature, wind speed and wind direction influence populations of B. tabaci and TYLCV incidence. Geographical variables such as buffer distance and block size also influence populations of B. tabaci and TYLCV. Shorter buffer distances and smaller block sizes had consistently larger counts of B. tabaci and higher incidence of TYLCV. Rainfall and cropping factor variables such as mulch type and tomato type did not have as much influence as previously thought.
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 James Edwin Taylor.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Schuster, David J.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043169:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 THE DISTRIBUTION OF RELATIONSHIP BETWEEN AND FACTORS INFLUENCING THE ABUNDANCE OF BEMISIA TABACI AND THE INCIDENCE OF TOMATO YELLOW LEAF CURL VIRUS IN SOUTHERN FLORIDA TOMATO By JAMES EDWIN TAYLOR 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 James E. Taylor

PAGE 3

3 ACKNOWLEDGMENTS I thank my parents and my sister who provide d support and guidance throughout my life and my scholastic career. I give a special thanks to my beautiful wife, Kate who I am blessed to have as my best friend and personal editor. Mo re importantly she was always there with support and encouragement t hroughout the research and writing process. I thank my major professor, Dr. David Schuster for all the guidance and life a helpful hand and allowed me the freedom to explo re. I thank my committee members Dr. Sabine Grunwald, Dr. Jane Polston and Dr. Susan Webb for their continued support and direction they have given me throughout my research. I would like to extend a huge thanks to the scouting team because without their eagerness to ride around tomato fields and the patience to look for a little white insect this project would have never been completed. This includes Don MacIntyre, Harvey Wade, Tad Connine, Bill MacIssac, Lloyd Stetler, Jim Watson, Lynne Bullerman Deborah Kiser and Jim Blackstone. Thanks to the Vegetable Entomology Laboratory at the Gulf Coast Research and Education Center and the entire GCREC for their support.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 3 LIST OF TABLES ................................ ................................ ................................ ............ 6 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 10 1 INTRODUCTION ................................ ................................ ................................ .... 12 Overview ................................ ................................ ................................ ................. 12 Purpose of the Study ................................ ................................ .............................. 14 2 REVIEW OF LITERATURE ................................ ................................ .................... 16 Bemisia tabaci (Gennadius) ................................ ................................ .................... 16 Taxonomy ................................ ................................ ................................ ......... 16 Biology ................................ ................................ ................................ .............. 17 History ................................ ................................ ................................ .............. 18 Hosts ................................ ................................ ................................ ................ 19 Management ................................ ................................ ................................ .... 23 Cultural control ................................ ................................ ........................... 24 Chemical control ................................ ................................ ........................ 26 Biological control ................................ ................................ ........................ 29 Spatial Distribution ................................ ................................ ........................... 31 Migration and Dispersal ................................ ................................ .................... 32 Tomato yellow lea f curl virus ................................ ................................ ................... 36 Biology and Distribution ................................ ................................ .................... 36 Hosts ................................ ................................ ................................ ................ 37 TYLCV Plant Relationship ................................ ................................ .............. 39 TYLCV B. tabaci Relationship ................................ ................................ ........ 39 Management ................................ ................................ ................................ .... 43 Geo graphic Information Systems and Spatial Statistics ................................ .......... 47 GIS ................................ ................................ ................................ ................... 47 Geostatistics ................................ ................................ ................................ ..... 49 GIS Uses in Entomology ................................ ................................ .................. 52 GIS Uses in Plant Disease Management ................................ ......................... 57 Spatial Analysis by Distance IndicEs (SADIE) ................................ .................. 59 Classification and Regression Tree Analysis ................................ .................... 62 3 SPATIAL AND TEMPORAL DISTRIBUTION OF BEMISIA TABACI AND TYLCV IN TOMATO ................................ ................................ ................................ ............ 65 Purpose ................................ ................................ ................................ .................. 65

PAGE 5

5 Methods and Materials ................................ ................................ ............................ 67 Study Sites ................................ ................................ ................................ ....... 67 Data Analyses ................................ ................................ ................................ .. 68 Results ................................ ................................ ................................ .................... 70 Fall 2007 ................................ ................................ ................................ ........... 70 Spring 2008 ................................ ................................ ................................ ...... 71 Fall 2008 ................................ ................................ ................................ ........... 71 Spring 2009 ................................ ................................ ................................ ...... 71 IDW Interpolat ion ................................ ................................ .............................. 72 SADIE Analysis ................................ ................................ ................................ 72 Discussion ................................ ................................ ................................ .............. 73 4 RELATIONSHIP OF ABUNDANCE OF BEMISIA TABACI TO INCIDENCE OF TYLCV IN THE FIELD AND ITS IMPLICATIONS TO MANAGEMENT AND EPIDEMIOLOGY ................................ ................................ ................................ .. 104 Purpose ................................ ................................ ................................ ................ 104 Methods a nd Materials ................................ ................................ .......................... 105 Study Sites ................................ ................................ ................................ ..... 105 Data Analyses ................................ ................................ ................................ 106 Results ................................ ................................ ................................ .................. 109 Fall 2007 ................................ ................................ ................................ ......... 109 Spring 2008 ................................ ................................ ................................ .... 110 Fall 2008 ................................ ................................ ................................ ......... 110 Spring 2009 ................................ ................................ ................................ .... 111 IDW Interpolation ................................ ................................ ............................ 111 Discussion ................................ ................................ ................................ ............ 111 5 FACTORS INFLUENCING ABUNDANCE AND SEVERITY OF BEMISIA TABACI AND TYLCV IN TOMATO ................................ ................................ ....... 140 Purpose ................................ ................................ ................................ ................ 140 Methods a nd Materials ................................ ................................ .......................... 142 Study Sites ................................ ................................ ................................ ..... 142 Explanatory Variables ................................ ................................ .................... 143 CART Ana lysis ................................ ................................ ............................... 144 Results ................................ ................................ ................................ .................. 145 B. tabaci ................................ ................................ ................................ ......... 145 TYLCV ................................ ................................ ................................ ............ 147 Discussion ................................ ................................ ................................ ............ 148 6 CONCLUSIONS ................................ ................................ ................................ ... 160 LIST OF REFERENCES ................................ ................................ ............................. 166 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 206

PAGE 6

6 LIST OF TABLES Table page 3 1 Total scouting area and sample sites, 2007 2009 ................................ .............. 79 3 2 Summary data for distribution of B. tabaci adults and TYLCV incidence including SADIE analysis, Fall 2007 ................................ ................................ ... 80 3 3 Summary data for distributio n of B. tabaci adults and TYLCV incidence including SADIE analysis, Spring 2008 ................................ .............................. 81 3 4 Summary data for distribution of B. tabaci adults and TYLCV incidence including SADIE analysis, Fall 2008 ................................ ................................ ... 82 3 5 Summary data for distribution of B. tabaci adults TYLCV incidence including SADIE analysis, Spring 2009 ................................ ................................ .............. 83 3 6 Cross valida tion results of IDW interpolation analysis for B. tabaci means and final TYLCV incidence, 2007 2009 ................................ ................................ ..... 84 3 7 Weekly SADIE aggregation indices of adult B. tabaci counts, Fall 2007 ............ 85 3 8 Weekly SADIE aggregation indices of TYLCV incidence, Fall 2007 ................... 86 3 9 Weekly SADIE aggregation indices of adult B. tabaci counts, Spring 20 08 ........ 87 3 10 Weekly SADIE aggregation indices of TYLCV incidence, Spring 2008 .............. 88 3 11 Weekly SADIE aggregation indices of adult B. t abaci counts, Fall 2008 ............ 89 3 12 Weekly SADIE aggregation indices of TYLCV incidence, Fall 2008 ................... 90 3 13 Weekly SADIE aggregation in dices of adult B. tabaci counts, Spring 2009 ........ 91 3 14 Weekly SADIE aggregation indices of TYLCV incidence, Spring 2009 .............. 92 4 1 Corre lations of adult B. tabaci weekly means to tomato plants with new incidence of TYLCV infection three weeks later, Fall 2007 ............................... 118 4 2 Regression of B. tabaci adults to incidence of tomato plants with symptoms of TYLCV infection three weeks later, 2007 2009 ................................ ............ 119 4 3 Indices of spatial association ( X ) for adult B. tabaci and tomato plants with new symptoms of TYLCV infection three weeks l ater, Fall 2007 ...................... 120 4 4 Correlations of adult B. tabaci weekly means to tomato plants with new incidence of TYLCV infection three weeks later, Spring 2008 .......................... 121

PAGE 7

7 4 5 Indices of spatial association ( X ) for adult B. tabaci and tomato plants with new symptoms of TYLCV infection three weeks later, Spring 2008 ................. 122 4 6 Correlation s of adult B. tabaci weekly means to tomato plants with new incidence of TYLCV infection three weeks later, Fall 2008 ............................... 123 4 7 Indices of spatial association ( X ) for adult B. tabaci and tomato p lants with new symptoms of TYLCV infection three weeks later, Fall 2008 ...................... 124 4 8 Correlations of adult B. tabaci weekly means to tomato plants with new infection of TYLCV incidence three weeks l ater, Spring 2009 .......................... 125 4 9 Indices of spatial association ( X ) for adult B. tabaci and tomato plants with new symptoms of TYLCV infection three weeks later, Spring 2009 ................. 126 4 10 Cross validation results of IDW interpolation analysis for B. tabaci weekly means and tomato plants with new symptoms of TYLCV incidence three weeks later, Fall 2007 2008 ................................ ................................ .............. 127 5 1 Description of variables in the CART analysis used to study the influence of environmental, geographical and cropping variables on the abundance of B. tabaci adults and the incidence of tomato plants with symptoms of TYLCV i nfection ................................ ................................ ................................ ............ 152 5 2 Summary of 10 fold cross validation results to predict adult B. tabaci and TYLCV incidence ................................ ................................ .............................. 153 5 3 Ranking of pre dictor variables on B. tabaci abundance and TYLCV incidence in Fall 2007/2008 ................................ ................................ .............................. 154 5 4 Ranking of predictor variables on B. tabaci abundance and TYLCV incidence in Spring 2008/2009 ................................ ................................ ......................... 155

PAGE 8

8 LIST OF FIGURES Figure page 2 1 An example of a variogram used in geostatistical analysis ................................ 64 3 1 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm A and Farm B, Fall 2007. ................................ ................................ .................... 93 3 2 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm C and Farm D, Fall 2007. ................................ ................................ ................... 94 3 3 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm E and Farm F, Fall 2007. ................................ ................................ .................... 95 3 4 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm G, Fall 2007. ................................ ................................ ................................ ....... 96 3 5 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm H and Farm I, Sp ring 2008. ................................ ................................ ................ 97 3 6 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm J and Farm K, Spring 2008. ................................ ................................ .................. 98 3 7 Sp atial interpolation of adult populations of B. tabaci and TYLCV from Farm A and Farm C, Fall 2008. ................................ ................................ ................... 99 3 8 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm L and Farm E, Fall 2008. ................................ ................................ ..................... 100 3 9 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm F, Fall 2008. ................................ ................................ ................................ ...... 101 3 10 Spatial i nterpolation of adult populations of B. tabaci and TYLCV from Farm H and Farm B, Spring 2009. ................................ ................................ ............. 102 3 11 Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm I and Farm J, Spring 2009. ................................ ................................ ................. 103 4 1 Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later, Week of 20 August 2007 ................................ ......... 128 4 2 Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later, Week of 3 September 2007 ................................ .... 129 4 3 Regression of adult B. tabaci counts o n tomato plants with new symptoms of TYLCV three weeks later, Week of 10 September 2007 ................................ .. 130

PAGE 9

9 4 4 Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks late r, Week of 17 September 2007 ................................ .. 131 4 5 Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later, Week of 22 October 2007 ................................ ....... 132 4 6 Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later, Week of 1 September 2008. ................................ ... 133 4 7 Spatial interpolatio n of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms A, B, and D from the week of 20 August 2007. ................................ ................. 134 4 8 Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms C and E from the week of 3 September 2007. ................................ .................. 135 4 9 Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms C and E from the week of 10 September 2007. ................................ ................ 136 4 10 Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms E and F from the week of 17 September 2007. ................................ ................ 137 4 11 Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms D and F from the week of 22 October 2007. ................................ ..................... 138 4 12 Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms C and E from the week of 1 September 2008. ................................ .................. 139 5 1 Regression tree of the variables influencing populations of B. tabaci in Fall 2007/2008. ................................ ................................ ................................ ....... 156 5 2 Regression tree of the variables influencing populations of B. tabaci in Spring 2008/2009. ................................ ................................ ................................ ....... 157 5 3 Regression tree of the variables influencing TYLCV incidence in Fall 2007/2008. ................................ ................................ ................................ ....... 158 5 4 Regression tree of the variables influencing TYLCV incidence in Spring 2008/2009. ................................ ................................ ................................ ....... 159

PAGE 10

10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfil lment of the Requirements for the Degree of Doctor of Philosophy THE DISTRIBUTION OF, RELATIONSHIP BETWEEN, AND FACTORS INFLUENCING THE ABUNDANCE OF BEMISIA TABACI AND THE INCIDENCE OF TOMATO YELLOW LEAF CURL VIRUS IN SOUTHERN FLORIDA TOMATO By James E. Taylor August 2011 Chair: David J. Schuster Major: Entomology and Nematology Biotype B of the sweetpotato whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae), also known as the silverleaf whitefly, B. argentifolii Bellows and Perring, is a ser ious pest of many agricultural crops around the world. In Florida, B. tabaci has become a limiting pest species in tomato due to its ability to vector Tomato yellow leaf curl virus (TYLCV) (family Geminiviridae, genus Begomovirus ). TYLCV is vectored in a persistent circulative manner and symptoms of infection in tomato include upward curling of leaflet margins, reduction of leaflet area, yellowing of young leaves abscission of flowers and stunting of plants The sampling of adult B. tabaci and TYLCV a cross commercial Florida tomato farms in four seasons from fall 2007 through spring 2009 combined with mapping of their distribution by Geographical Information Systems (GIS) and analyses by Spatial Analysis by Distance IndicEs (SADIE) and classification a nd regression tree (CART) analysis have produced a much more detailed explanation of in field distribution, vector/disease relationship and influencing factors than previously reported. B. tabaci is a mobile pest and has been shown in the present study to have varying aggregation in

PAGE 11

11 both space and time. D istributions of B. tabaci were significantly aggregated in every season but spring 2009 when the population of whitefly was very low Weekly fluctuations throughout the study area suggest that within t he earlier sampling dates of each season, whiteflies were more likely to be aggregated. I nverse distance weighted (IDW) maps created by a GIS program, showed that populations of B. tabaci and incidence of plants with symptoms of TYLCV infection were assoc iate d more closely with the edges of tomato fields Early fall season and late spring season populations of adult B. tabaci had stronger correlations to incidence of symptomatic TYLCV infected plants. SADIE spatial association tests indicated similar con clusions. There were indications that B. tabaci may have migrated from areas in which possible whitefly hosts were destroyed or disturbed. CART analysis confirmed the assumptions that environmental variables such as temperature wind speed and wind direc tion influence populations of B. tabaci and TYLCV incidence. Geographical variables such as buffer distance and block size also influence populations of B. tabaci and TYLCV. S horter buffer distances and smaller block sizes had consistently larger counts of B. tabaci and higher incidence of TYLCV. Rainfall and c ropping factor variables such as mulch type and tomato type did not have as much influence as previously thought.

PAGE 12

12 CHAPTER 1 INTRODUCTION Overview Biotype B of the sweetpotato whitefly, Bemisia tabaci (Gennadiu s) (Hemiptera: Aleyrodidae), also known as B. argentifolii Bellows and Perring (Bellows et al. 1994) is an important economic pest in tropical and subtropical climates around the worl d (Perring et al. 1993) The no menclature of Bemisia spp. has been widely discussed and current research suggest s there are multiple unique species worldwide (Dinsdale et al. 2010) The wide host range of over 600 plant species (Mound and Halsey 1978, Greathead 1986, Secker et al. 1998) including weed hosts th at vary in their importance, depending on the cropping system, makes management even more challenging (Cohen et al. 1988, Schuster et al. 1992, Ucko et al. 1998, Bezerra et al. 2004) The polyphagous nature of the sweetpotato whitefly leads to management problems and could be associated with its high value pest status in many commodities (Naranjo and Ellsworth 2001) The sweetpotato whitefly produce s direct feeding damage in tomato Feeding by nymphs c an cause irregular ripening of fruit (Maynard and Cantliffe 1989, Schuster et al. 1990) a nd inhibition of fruit softening (Hanif Khan et al. 1999, McCollum et al. 2004) along with general reduction in plant vigor. The sweetpotato whitefly cause s considerable yield loss in many areas of the world due to its capability of vectoring plant viruses (Markham et al. 1996, Naranjo and Ellsworth 2001) One important sweetpotato whitefly vectored virus is T omato yellow leaf curl virus (TYLCV: genus Begomovirus family Geminiviridae ) (Polston et al. 1999) It is vectored in a persistent, circulative manner and causes extensive dama ge to tomato worldwide (Czosnek and Laterrot 1997) TYLCV can cause yield losses of up to 100% in tropical

PAGE 13

13 and subtropical regions and in some regions is the limiting factor in commercial tomato production (Czosnek and Laterrot 1997) TYLCV was first recorded in the Middle East North and Central Africa, and Southeast Asia, and has recently spread to Europe (Czosne k et al. 1990, Moriones and Navas Castillo 2000) It has also been recorded in the Caribbean (Nakhla et al. 1994) and Mexico (Ascencio Ibanes et al. 1999) In 1999, it was reported in the United States in Florida (Polston et al. 1999) and Georgia (Momol et al. 1999) and in 2001 in Louisiana (Valverde et al. 2001) Symptoms include leaf curling, chlorosis of leaf margins, reduction of leaf size, mottling, flower abscission, plant stunting and yield reduction (Polston et al. 1999, Mohamed 2010) TYLCV symptoms appear 2 3 weeks after infection and the virus can be acquired by the adult whitefly in 15 30 minutes (Cohen and Nitzany 1966, Rom et al. 1993) During symptom e xpression there is considerable loss in plant vigor and significant yield loss particularly if plants are infected during early growth. Management of TYLCV is difficult and requires a multi faceted approach (Momol et al. 2001) Un fortunately, growers often rely heavily on the use of insecticides to control T YLCV by targeting the whitefly vector a nd insecticide resistance is widespread in B. tabaci (Palumbo et al. 2001, Horowitz et al. 2007) Geographic information systems (GIS) and global positioning systems (GPS) might be useful in monitoring and predicting the distribution of whiteflies and TYLCV. GI S are software tools which allow for storage, analysis, synthesis, and output of spatial data (Bolstad 2005) Historically, applications of GIS in entomology have been limited to forest and rangeland entomology (Kemp et al. 1989, Schotzko and O'Keeffe 1989) Recently, it has been applied to study insect pests in agricultural systems (Barnes et al.

PAGE 14

14 1999, Park and Obrycki 2004, Carriere et al. 2006, Garcia 2006, Reay Jones et al. 2010) S patial dependence in insects shows that there can be value at interpolating counts at un sampled locations (Borth and Huber 1987, Schotzko and O'Keeffe 1989, Setzer 1995) Spatial interpolation methods include nea rest neighbor, inverse distance weighting and spatial prediction models that include k riging (Bolstad 2005) Some insects, including whiteflies, are able to migrate over large areas; therefore, monitoring movement on a regional or area wide scale could be beneficial (Son ka et al. 1997) GIS and geostatistics have also been used to implement management plans for plant viruses (Nelson et al. 1994, Barnes et al. 1999, Nelson et al. 1999) and to analyze spatial patterns of plant dis eases (Fargette et al. 1985, Chellemi et al. 1988) Other statistical program s such as Spati al Analysis by Distance IndicEs (SADIE) have been developed to quantify spatial pattern s of organisms (Perry 1998) SADIE measures the degree of aggregation in spatially referenced data and is based on discrete count data (Xu and Madden 2004) Classification and regression tree (CART) ana lysis was designed to explore and model ecolog ical data and can deal with non linear, complex and missing data values (Breiman et al. 1984) CART based models can handle categorical and continuous data CART analysis uses trees to explain variation of target variables by repeatedly splitting the data into homogeneous partitions. Purpose of t he Study Current management tactics for sweetpotato whitefl y and TYLCV in Florida include the use of a crop free summer period, virus free transplants resista nt cultivars, ultraviolet reflective mulch chemical control of whiteflies, sanitation or r ogu ing of TYLCV infected plants and removal of old plant material (Polston et al. 1999, Schuster et al. 2007a) The use of GIS can lead to a more thorough understanding of the

PAGE 15

15 dispersal of the sweetpotato whitefly a nd of the whitefly and virus reservoirs at both the field and regional level. SADIE analysis of B. tabaci adult counts and symptomatic TYLCV infected tomato plants can express the nature of the distribution of the populations. Using CART analysis enviro nmental, geographical and cropping factor variables can be evaluated for their importance in influencing B. tabaci and TYLCV. With a greater understanding of the distribution of B. tabaci and TYLCV their relationship, and variables influencing population s result s may lead to the development of new management recommendations. The specific objectives for research were: 1) To evaluate seasonal abundance of B tabaci and incidence of TYLCV in Florida tomatoes 2) To investigate the spatial and temporal distribution of B tabaci adults and TYLCV infected plants in Florida tomatoes 3) To investigate the relationship between the abundance of B. tabaci and incidence of TYLCV in Florida tomatoes 4) T o investigate the environmental, geographical, and cropping factor variables i nfluencing the abundance of B tabaci and incidence of TYLCV in Florida tomatoes The overall hypothesis for thi s research was that non tomato hosts in west central Florida can influence populations of B tabaci and subsequent incidence of TYLCV in tomato

PAGE 16

16 CHAPTER 2 REVIEW OF LITERATURE Bemisia tabaci (Gennadius) Taxonomy The B biotype of the sweetpotato whitefly, B emisia tabaci also known as the silverleaf whitefly, B. argentifolii ( Bellows and Perring ) was first described as Aleyrodes tabaci by Genna dius (1889) in Greece. It was first recorded in the United States in 1897 (Russell 1957, Mound and Halsey 1978) There has been confusion in the literature on the nomenclature of this insect as Mound and Halsey (1978) listed 22 synonyms for B. tabaci. Gill (1992) suggested that the A (cotton biotype) and the B (poinsettia biotype) of the sweetpotato whitefly were two disti nct species. Perring et al. (1993) suggested a common name of silverleaf whitefly for the introduced sweetpotato B biotype and concluded that there were two distinct species based on the absence of inter biotype copulation along with genotypic and phenotypic differences. Based on light microscopy (Bedford et al. 1994) and t ransmission electron microscopy inspections of B. tabaci no distinctive characteristics were found to determine diff erences in Bemisia spp. (Rosell et al. 1997) These examinations along with similarities in morphological characters and evidence of biotic and genetic polymorphism (Costa and Brown 1991, Burban et al. 1992, Perring et al. 1 993, Bedford et al. 1994, Brown et al. 1995, De Barro and Driver 1997) have led some researchers to hypothesize that B. tabaci is a cryptic species or species complex (Bedford et al. 1994, Brown et al. 1995, Rosel l et al. 1997, Frohlich et al. 1999, Brown et al. 2000) Perring (2001) reviewed the species complex and hypothesized that there are seven groups within the B. tabaci species. Dinsdale et al. (2010) examined mitochondrial cytochrome oxidase 1 to determine species

PAGE 17

17 differentiation among B. tabaci and concluded that at >3.5% divergence there could be up to 24 species worldwide. The designation of this insect as multiple unique species is not universally accepted, so B tabaci B biotype (= B. argentifolii ), sweetpotato whitefly will be used in the present treatise Biology B tabaci (Hemiptera: Aleyrodidae) is a cosmopolitan polyphagous pest and has become one of the most important pests of world agriculture (Naranjo and Ellsworth 200 1) There are approximately 150 whitefly species in the United States and over 1500 species worldwide (Miller et al. 2001) General whitefly biology was reviewed by Byrne and Bellows (1991) Whiteflies are plant feeders with piercing, sucking mouthparts and undergo incomplete m etamorphosis (Byrne and Bell ows 1991) Whitefly adults are small insects 2 3 mm in length and range from pale to completely pigmented in color (Miller et al. 2001) B. tabaci adults are approximately 2mm in length (Byrne and Bellows 1991) B. tabaci produce offspring based on haplodiploidy, i.e. males are produced from u nfertilized, haploid eggs, and females are produced from fertilized, diploid eggs (Denholm et al. 199 8, Klowden 2002) Whitefly eggs are usually attached to the underside of the leaves, may be smooth or sculptured, and can be laid in patterns or scattered over the leaf (Byrne and Bellows 1991) B. tabaci eggs possess a pedicel, are elliptical in shape and are laid indiscriminate ly (Byrne and Bellows 1991) Oviposition rates vary greatly and are dependent on environmental conditions and host plants (Powell and Bellows 1992, Muniz 2000, Gruenhagen and Perring 2001, Omondi et al. 2005) There are four immature stage s between the egg and adult, with the first three being larval instars and the fourth stage labeled the pupae stage or puparium (Byrne and Bellows 1991) The first immature stage is called the crawler because it is

PAGE 18

18 the only mobile immature stage. The second and third instars are o val in shape and are sessile. The fourth stage is elliptical in shape, is the most common stage to identify species differentiation, and is characterized by the pair of eyes which show up as red spots in B. tabaci (Lopez Avila 1986) Emergence of adult whiteflies takes 5 15 minutes W ing expansion occurs on or near the pupal case and takes approximately 40 50 m inutes (Azeb et al. 1972) The length of the B. tabaci life cycle can vary greatly depending on climatic and host plant conditions (Russell 1975, Coudriet et al. 1985) Under field conditions the life cycle can last from 14 to 75 days (Azeb et al. 1971) Coudriet et al. (1985) found B. tabaci development from egg to adult on tomato to b e 27.3 1.0 days at 26.7 1.0 C whereas Salas a nd Mendoza (1995) observed development on tomato to be 22.3 days at 25C and 65% R. H. Lopez Avila (1986) determined development time on tomato was 23.5 days. On tomato at 25C and 65% R. H. egg incubation took an average of 7.3 0.5 d ays (Salas and Mendoza 1995) and 7.3 d ays at 25C and 75% R. H (Lopez Avila 1986) Egg incubation periods can vary from 3 to 33 days depending on temperature and humidity (Husain and Trehan 1933, Avidov 1956, Azeb et al. 1972, Butler et al. 1983, Powell and Bellows 1992, Liu and Stansly 1998) The first instar stage duration is approximately 4.0 1.0 days; second instar 2.7 1.1 days; third instar 2.5 0.7 days ; and fourth instar/pupa 5.8 0.3 days on tomato (Salas and Mendoza 1995) In tropical field conditions there can be 10 to 16 generations per year (Avidov 1956, Azeb et al. 1972, Salas and Mendoza 1995) History The sweetpotato whitefly has been associated with many agricultural losses and is the limiting pest species in many field and vegetable crops around the world. It has

PAGE 19

1 9 benefited from international trade movement and is now found on every continent except A ntarctica (De Barro 1995, Martin et al. 2000) B. tabaci is a cosmopolitan pest that has (Barinaga 1993) Sweetpotato whitefly B biotype is tho ught to have originated in the northeast Africa/Middle East/Arabian peninsula region (Frohlich et al. 1 999, De Barro et al. 2000) Global outbreak s over recent years have been correlated with whitefly geminiviruses (Polston a nd Anderson 1997, Rubinstein et al. 1999) In the United States, B. tabaci has shown the potential to cause millions of dollars in crop damage and lost yields (Perring et al. 1993, Birdsall et al. 1995, Ellsworth et al. 1999) Since the introd uction of the B biotype in 1986 the sweetpotato whitefly has become a problem in Florida (Price 1987) In a survey by McKenzie et al. (2004) in Florida, researchers co ncluded that the B biotype of B. tabaci has exclude d the native non B biotypes. B. tabaci has been linked to tomato irregular ripening disorder and squash silverleaf disorder (Schuster et al. 1990, Schuster et al. 1991) in Florida and has become a limiting pest species in tomato production due to its ability to transmit TYLCV (Polston et al. 1999) Hosts B tabaci has a host r ange of over 600 plant species (Mound and Halsey 1978, Greathead 1986, Secker et al. 1998, Evans 2007) It has been suggested that B. tabaci has a much wider host range than other Bemisia biotypes indicating its s uccess as a cosmopolitan pest is due to a large host range (Brown et al. 1995, Perring 2001) Weed hosts vary in their importance depending on the cropping system (Cohen et al. 1988, Schuster et al. 1992, Ucko et a l. 1998, Bezerra et al. 2004) B. tabaci B biotype has been shown to be much better at developing and surviving on multiple hosts than a native biotype in China (Zang et al. 2006) This whitefly has shown the ability to quickly

PAGE 20

20 acclimatize to alternative hosts which may give it advantages over non B biotypes (Gerling and Kravechenko 1996) Even though alternative host sources are important for insect pe sts including B tabaci most of the current research has focused on cultivated crops (Avidov 1956, Gerling 1984, Schuster et al. 1992, Muniz 2000, Simmons et al. 2008) There has been considerable research on B. tabaci survival (Costa et al. 1991, Tsai and Wang 1996, Liu and Stansly 1998, Omondi et al. 2005, Bayhan et al. 2006, Walker and Natwick 2006, Zang et al. 2006, Jindal et al. 2008, Mansaray and Sundufu 2009) devel opmental rates (Coudriet et al. 1985, Powell and Bellows 1992, Tsai and Wang 1996, Liu and Stansly 1998, Omondi et al. 2005, Bayhan et al. 2006, Zang et al. 2006, Jindal et al. 2008, Mansaray and Sundufu 2009, Baldi n and Beneduzzi 2010) and fecundity (Costa et al. 1991, Chu et al. 1995, Tsai and Wang 1996, Liu and Stansly 1998, Toscano et al. 2002, Omondi et al. 2005, Bird and Kruger 2006, Walker and Natwick 2006, Zang et al. 2006, Boica Junior et al. 2007, Jindal et al. 2008, Mansaray and Sundufu 2009, Baldin and Beneduzzi 2010) on cultivated plant species. Until recently, research on weeds as hosts of B. tabaci h as been relatively ignored in the literature even though weed hosts are considered important for its management (Gerling 1984, Cohen et al. 1988, Schuster et al. 1992, Hilje et al. 2001) Weeds can act as important reservoirs for whiteflies and their natural enemies and in some systems can be hosts for c ultivated crop pathogens (Bezerra et al. 2004, Naveed et al. 2007) In Italy, Calvitti and Remotti (1998) examined developmental rate s of B. tabaci biotype B under laboratory c onditions on herbaceous weeds found around greenhouses. Sonchus oleraceus L. (sowthistle) and Solanum nigrum L. (black

PAGE 21

21 nightshade) were found to be preferred hosts of B. tabaci and also had the highest intrinsic rate of increase (Calvitti and Remotti 1998) In Spai n, Muniz (2000) concluded there were differences in compo sition and host suitability to the sweetpotato whitefly in overwintering and over summering host species. In winter weeds, Malva par viflora L. (cheeseweed) had significantly more eggs laid per female, higher numbers of pupae and greater percentages of ad ult emergence than Capsella bursa pastor is ( L. ) Medik. Brassica kaber (DC) (wild mustard), and Lactuca serriola L. (prickly lettuce) (Muniz 2000) Summer weeds Datura stramonium L. (jimsonweed) and S. nigrum were more suitable hosts in term s of numbers of infested plants and higher numbers of pupae and adul ts than Amaranthus retroflexus L. (amaranth), Chenopodium album L. (lambsquarters) and Echinochloa crus galli ( L. ) P. Beauv. (barnyardgrass). Host suitability studies in Ghana demonstrated that Desmodium tortuosum (Sw.) DC. (Dixie ticktrefoil) and Euphorb ia heterophylla ( L. ) (Mexican fireplant) were good hosts of B. tabaci based on fecundity and survival compared to A. retroflexus Chromolaena odorata ( L. ) R. M. King and H. Rob. (jack in the bush) and Malvastrum coromandelianum ( L. ) Garcke (threelobe false mallow) (Gachoka et al. 2005) Field studies conducted in Brazil by Bezerra et al. (2004) found higher B. tabaci populations on Ancanthospermum hispidum de Amaranthus deflexus L. (largefruit amaranth), D. stramonium and E. heterophylla In the presence of weeds, tomato infestations by B. tabaci were reduced (Bezerra et al. 2004) suggesting that there might have been a reduction of attractiveness of tomato by a phenomenon discussed by Bernays (1999) Bird and Kruger (2006) found similar

PAGE 22

22 results to Bernays (1999) in a choice test of similar hosts of mixed cultivated crops and cultivars. In the United States, understanding weed hosts is important for management of both B. ta baci and the ir associated viral epidemics (Schuster et al. 1992, Hilje et al. 2001) F lorida weed hosts listed by Stansly and Schuster (1990) were similar in species composition to those listed by other researchers worldwide. Early work by Gerling (1967) found B. tabaci on M parviflora in the Imperial Valley of California. Other Imperial Valley weeds were evaluated as hosts of B. tabaci by Coudriet et al. (1986) L serriola ( prickly lettuce) was the best host in terms of developmental time but S. oleraceus ( sowthistle ) and S. asper Hill were good hosts as well althoug h M. parviflora only differed in adult development time by 2 to 3 days from the best host (Coudriet et al. 1986) C rop plants in the Imperial Valley of California may have as much influence on overwintering whitefly populations as weed hosts (Coudriet et al. 1985, Coudriet et al. 1986) In southwest Florida, Stansly (1995) determined weeds were poor intermediate hosts for whiteflies during suggested crop free periods in the summer. In another Florida study B. tabaci populations in weeds paralleled those found in neighboring tomato fields (Schuster et al. 1 992) Hence, the relationship of weeds to overall whitefly population dynamics is unclear To determine hosts, insects use both visual and olfactory cues (Visser 1988) Color plays a main role in visual determination of a host an d B. tabaci is attracted most strongly to yellow/green in the range of 500 700 nm (Husain and Trehan 1940, Berlinger 1986) Olfactory cues are not considered key factors in host determination in B. tabaci (Berlinger 1986, Van Lenteren and Noldus 1990) Based on olfactory cues

PAGE 23

23 from five different hosts Jing et al. (2003) found host plant preferences a ffected attraction of B. tabaci but they were unsure of the B biotype label of their experimental colony. Alt hough not m uch data suggests that olfaction plays a role in attraction of B. tabaci to hosts, there are current research projects to determine the roles of volatile semiochemicals (Bleeker et al. 2009) and ginger oil (Zhang et al. 2004) on repellency. Once landed whit efl y adults determine host acceptance by contact cues, touch and taste (Berlinger 1986) If the plant is found unfavorable the whitefly will leave or have reduced fecundity which could be due to many factors such as leaf hairiness (Mound 1965, Butler and Henneberry 1984, McAuslane 1996, Gruenhagen and Perring 2001, Mansaray and Sundufu 2009, Baldin and Beneduzzi 2010) leaf age (Bentz et al. 1995c, Liu and Stansly 1995b, Cardoza et al. 2000) pH (Berlinger et al. 1983) secondary metabolites (Baldin and Ben eduzzi 2010) nitrogen availability (Bentz et al. 1995a, Bentz et al. 1995c) and amino acid composition (Blackmer and Byrne 1999) Nitrogen availabil ity in the host plant has been shown to affect B. tabaci populations on some crops including cotton (Blua and Toscano 1994, Bi et al. 2001) Management Successful control of the sweetpotato whitefly requires flexible management programs. Cultural control for sweetpotato whitefly has been reviewed by Hilje et al. (2001) Ellsworth and Martinez Carrillo (2001) and Gerling and Mayer (1996) The most commonly used practice to contro l B. tabaci is chemical control and most whitefly control is dependent on insecticides (Palumbo et al. 2001) Therefore, resistance to many different insecticide classes has been documented (Palumbo et al. 2001 ) With the loss of efficacy of certain chemical classes and chemistries, research has been conducted on resistance management (Palumbo et al. 2001) Concurrent research has

PAGE 24

24 been directed at evaluation of sampling methods (Ekbom and Rumei 1990, Naranjo 1996) and action thresholds (Ellsworth and Meade 1994, Riley and Palumbo 1995, Naranjo et al. 1998) for whitefly control The widespread use of broad spectrum insecticide s in many crops has limited the contribution of predators and parasitoids to control B tabaci With the increased use of selective insecticides and a greater adoption of Integrated Pest Management (IPM) practices, biological control for B. tabaci has had a growing interest in the literature (Naranjo 2001) IPM is the integration of multiple control tactics as part of an overall management plan. Unfortunately, IPM programs are based on local conditions and cropping systems so they are temporally and spatially relative, which complicates adoption on a large scale. Cultural c ontrol Cultural practices to control sweetpotato whitefly ha ve been reviewed by Hilje et al. (2001) Hilje et al. (2001) designated four categories to define whitefly management str ategies using cultural practices: avoidance in time or space, behavioral manipulation of the insect, host suitability and insect removal. Avoidance in time/space would consist of separating the crop from sources of the insect. Planting and termination da te manipulation can be a useful control tactic for both individual growers and for regional control (Ellsworth and Martinez Carrillo 2001) Crop free periods have been used in multiple agricultural systems to reduce sweetpotato whitefly pressure (Stansly and Schuster 1990, Nuessly et al. 1994, Alvarez and Abud Antun 1995, Ucko et al. 1998, Villar et al. 1998) Planting date m anipulation has been used in multiple cropping systems to avoid B. tabaci problems (Patel and Patel 1966, Borah 1994, El Gendi et al. 1997 Hernandez and Pacheco 1998, Mewally 1999) Spatially avoiding associated weed hosts of B. tabaci and vector reservoirs can reduce problems associated with both

PAGE 25

25 (Cohen et al. 1988) Watson et al. (1992) determined that whitefly populations could be affected by disrupting spatial and temporal relationships of neighboring crops. Physical exclusion of w hiteflies from plant material has been used to reduce damage from whiteflies and could include greenhouse structures (Cohen and Berlinger 1986, Horowitz and Ishaaya 1994, Antignus et al. 1996, Berlinger and Lebiush Mordechi 1996, Ausher 1997, Antignus et al. 1998, Costa and Robb 1999) row covers (Natwick and Durazo III 1985, Cohen and Berlinger 1986, Perring et al. 1989, Webb and Linda 1992, Costa et al. 1994, Orozco Santos et al. 1994, Farias Larios et al. 1995, Orozco Santos et al. 1995, Farias Larios et al. 1996, Avilla et al. 1997) barriers (Cohen et al. 1988, Smith and McSorley 2000, Hilje et al. 2001) and high planting densit y (Fargette and Fauquet 1988, Fargette et al. 1990, Ahohuendo and Sarkar 1995) Whitefly behavioral modifications with varying levels of control can be achieved by mulches (Cohen 1982, Cohen and Berlinger 1986, Suwwan et al. 1988, Orozco Santos et al. 1994, Csizinszky et al. 1995, Orozco Santos et al. 1995, Csizinszky et al. 1997, Hooks et al. 1998, Csizinszky et al. 1999, Simmons et al. 2010) and intercropping (Al Musa 1982, Stansly et al. 1998, Smith et al. 2000, Schuster 2004) Host suitabili t y alterations, e.g., changes in fertilization (Blua and Toscano 1994, Bentz et al. 1995b, Blackmer and Byrne 1999, Bi et al. 2001) and irrigation (Mor 1987, Leggett 1993, Flint et al. 1994, Flint et al. 1995, Flint et al. 1996) can affect whitefly reproduction and survival. Adult whitefly counts decline afte r a rain event (personal observation) and other authors have documented a similar decline in whitefly populations in other regions (Zalom et al. 1985, Henneberry et al. 1995) R esearchers such as Castle et al. (1996)

PAGE 26

26 and Castle (2001) have shown reductions in whitefly eggs and nymphs in overhead irrigated cotton and cantaloupe as compared to furrow irrigation treatments. C ultural practices have been used and tested worldwide but widespread adoption of practices is limited due to conventional cropping systems not allowing substantial change in grower acceptance, regional scale necessary for implementation, research experimen tation difficulty and the dependence of cultural practices on other control tactics. Chemical c ontrol The use of insecticides has been the primary strategy to control B. tabaci in many agronomic and vegetable crops around the world (Dennehy et al. 1996, Horowitz and Ishaaya 1996, Ellsworth and Martinez Carrillo 2001, Palumbo et al. 2001) Unfortunately, B. tabaci ha s developed resistance to all chemical c lasses applied for its control as reviewed by Dittrich et al (1990a) and Palumbo et al. (2001) Palumbo et al. (2001) reviewed the then current literature and determined that synergized pyrethroids were the most efficacious of the neurotoxic insecticides and combining pyrethroids with other chemical classes (tank mix ing ) could be more efficacious (Watson 1993, Ellsworth et al. 1994, Hor owitz and Ishaaya 1996, Prabhaker et al. 1998) The increased efficacy of tank mixing can be linked to the inhibition of insecticide resistance mechanisms due to increased esteratic activity and insensitive acetylcholinesterase towards inhibitors (Ishaaya et al. 1987, Prabhaker et al. 1988, Dittrich et al. 1990a, Byrne and Devonshire 1993, Denholm et al. 1998) Synergized pyrethroid sprays are more effective on adult whiteflies through contact action (Horowitz and Ishaaya 1996) but there is some efficacy against nymphs (Prabhaker et al. 1989)

PAGE 27

27 Newer insecticides with novel or less exploited modes of action are becoming more important for B. tabaci control around the world (Horowitz and Ishaaya 1994, Denholm et al. 1996) The chloronicotinyls or neonicotinoids (imidacloprid, acetamiprid, nitenpyram, and thiamethoxam) have shown good efficacy in controlling whiteflies and other insects (Elbert et al. 1990, Bethke and Redak 1997, Palumbo et al. 2001, Bacci et al. 2007) The se compounds most likely target the nicotinic acetylcholine receptors in the post synaptic region of insect nerves and because of their systemic activity, they can be used as soil applications or used as foliar sprays (Bai et al. 1991) Prabhaker et al. (1997) was able to create an imidacloprid resistant strain under laboratory conditions and Cahill e t al. (1996) found resistance to imidacloprid in greenhouse conditions in Southern Europe. These new chemistries have been great tools for controll ing whiteflies in recent years, but overuse and cross resistance between compounds within the neonicotinoid class threatens the continued efficacy of these products in the future (Palumbo et al. 2001) Though oils and soaps have be en available for control of whiteflies for a hundred years, synthetic organic insecticides have been in greater use. With the reduction in use of these synthetic insecticides other alternatives such as soaps and oils have either been discovered or redisc overed. Oils show toxicity against nymphs (Butler et al. 1993, Stansly et al. 1996, Liu and Stansly 2000) and adults (Stansly et al. 1996) but there is some risk of phytotoxicity. C ertain oils have also reduc ed oviposition (Liu and Stansly 1995b, Fenigstein et al. 2001, Schuster et al. 2009) and landing/settling (Liu and S tansly 1995a, Fenigstein et al. 2001, Schuster et al. 2009) Soaps in the form of surfactants or household detergents have efficacy against whitefly adults and nymphs

PAGE 28

28 (Butler et al. 1993) The mode of action of soaps and oils is a combination of physical action, suffocation, or repellency (Larew and Locke 1990, Stansly et al. 1996, Fenigstein et al. 2001) Soaps and oils are effective whitefly insecticides and are generally safer for non target organisms than conventional insecticides, but good plant coverage is required and the risk of phytotoxicity is increased Other novel insecticides such as insect growth re gulators (IGRs), diafenthiuron, and pymetrozine have expanded the list available for sweet potato whitefly control. IGRs a ffect normal insect physiology and can include chitin synthesis inhibitors (buprofezin) and juvenile hormone mimics (pyriproxyfen) (Horowitz and Ishaaya 1996, Palumbo et al. 2001) These compounds require the user to know and understand basic insect biology and ecology because they have selective efficacy on certain life st ages of the insect (Ellsworth and Martinez Carrillo 2001, Palumbo et al. 2001) Diafenthiuron is a thi ourea derivative and has a unique mode of action (Kadir and Knowles 1991, Ishaaya et al. 1993) Pymetrozine is in the pyridine azomethine class of insecticides and is selectively active against sucking insects within Homoptera (Nicholson et al. 1996) The mode of action of pymetrozine has been described as neural inhibition of feeding behavior by affecting the activity of cibarial and salivary pumps (Fluckiger et al. 1992, Kayser et al. 1994, Harrawijn and Kayser 1997) Th e se newer chemistries along with their divers e modes of action, can fit well with current management practices in controlling whiteflies and resistance issues (Palumbo et al. 2001, Ishaaya et al. 2007) Resistance monitoring has been conducted all over the world and will continue to be a central theme in ins ecticide research for Bemisia sp p control (Perry 1985, Ahmed

PAGE 29

29 et al. 1987, Dittrich et al. 1990b, Dittrich et al. 1990a, Prabhaker et al. 1992, Perez et al. 2000, Schuster 2007) With c ontinued resistance monitori ng and implementation of resistance management techniques such as non chemical control, limited use of chemical control tactics, rotation of chemistries, selective use of certain chemistries and the balanced use of insecticides across commodities, chemical control can be continued as a main control tactic for B tabaci (Ellsworth and Martinez C arrillo 2001) Biological c ontrol The widespread use of broad spectrum insecticides in many crops has limited the contribution of predators and parasitoids to control of B tabaci With the increased use of selective insecticides and a greater adoption of IPM practices, interest in biological control for B. tabaci has grow n in the literature (Naranjo 2001) In a recent review by Gerling et al. (2001) 114 arthropod predators from 9 orders and 31 families were listed T here are also parasitoids ( Hymenoptera ) attacking B. tabaci and Gerling et al. (2001) noted 34 species of Encarsia 14 species of Eretmocerus and several species in the genera Amit u s and Metaphycus based on a review of current literature. There are also 9 described and 2 undescribed species of fungi shown to n aturally occur in B. tabaci populations (Faria and Wraight 2001) Though biological control agents have been identifi ed and studied for many years, only in the last 20 years have researchers attempted to apply them for control of B. tabaci Predators have a unique advantage in biological control systems because many species are generalists and exhibit behavioral plasti city. They are able to feed and change prey species as prey availability changes thus making them important biological control agents (Gerling et al. 2001) Predators of B. tabaci include insects in the orders Coleoptera Diptera Hemiptera Hymenoptera Neuroptera Odonata and

PAGE 30

30 Thysanoptera and other orders of non insect ar thropods including Acari and Araneae Some important predators include beetles from the Coccinellidae family (Gerling 1986, Obrycki and Kring 1998) ; true bugs from Anthocoridae (Gerling 1986) Lygaeidae (Hagler and Naranjo 1 994) Miridae (Hagler and Naranjo 1994, Van Schelt et al. 1996, Jones and Snodgrass 1998) ; and Neuroptera from Chrysopidae (Gerling 1986, Dean and Schuster 1995) Predatory mites from Acari feed on B. tabaci and s ome are commercially available (Nomikou e t al. 2001) Predators of B. tabaci have been shown to control damaging populations in some systems though their potential is limited by the efficacy of insecticides against them (Gerling and Kravechenko 1996) K nowledge of B tabaci parasitoids has been hampered by taxonomic complexity of both B. tabaci and its parasitoid s, parasitoid biology, host range of B. tabaci and diversity of cropping patterns for B. tabaci crop hosts (Hoelmer 1996) Parasitoids of Bemisia spp belong to genera in the order Hymenoptera and include Encarsia Eretmocerus and Amitus as reviewed by Gerling et al (2001) These Hymenoptera parasitize whitefly nymphs and complete their development on fourth instars, so they do not control adult whiteflies di rectly (Gerling et al. 2001) Many whitefly parasitoids are oligophagou s allowing for control of new introduced whitefly species ; however, this may limit their efficacy as biological control agents (Gerling et al. 2001) Under certain conditions entomopathogenic fungi can control B. tabaci populations. Rainy seasons or cool, humid conditions can lead to epizootics of fungi to control whitefli es ; however, fungi usually can not be relied on for complete control because the development of epizootics relies on environmental conditions and crop production practices (Faria and Wraight 2001) Fungi controlling populations of whitefly

PAGE 31

31 usually lag behind build up of the pest insect and usually do no t control adults (Faria and Wraight 2001) S ome commercially available entomopathogens are suggested for use in protected systems (Dowell 1990) Since the expansion of B tabaci as a global pest, biolo gical control has been researched and applied in some agricultural systems. This application and research will c ontinue but much more work will need to be done before widespread adoption of biological control tactics will become ubiquitous for control of this noxious pest. Many issues complicate biological control and in some systems the efficiency of Bemisia sp p. to transmit plant viruses damper adoption of biological control tactics. Spatial Distribution Like most insects B tabaci is aggregated both within individual leaves and within plants at all life stages (Naranjo 1996) Dispersion is typically described with models (Taylor 1961) (Iwao 1968) Naranjo and Flint (1994) used law to describe aggregation of B. tabaci eggs and immatures in cotton on individual plants. Adult B. tabaci populations were shown to be aggregated on cotton plants within fields (Naranjo a nd Flint 1995) Using mean crowding index a mean Von Arx et al. (1984) determined aggregation of B. tabaci in cotton within fields In Florida, B tabaci was shown to be agg regated in commercial tomato using Moris i al though the population distribution fluctuated throughout the season (Polston et al. 1996) Distribution of whitefl y adults changed seasonally in Texas from ag gregated in the spring to more diffuse in the summer and early fall using the Taylor power law and the Morisita index (Riley and Ciomperlik 1997)

PAGE 32

32 and the greenhouse whitefly, Trialeurodes vaporariorum (Westwood) were shown to be aggregated between plants in all life stages on greenhouse ornamentals (Liu et al. 1993a) In another study greenhouse whitefly adults were shown to be aggregated in (Kim et al. 2001) In a study comparing B and Q biotypes of B. tabaci both biotypes were reluctant to move from tomato onto other tomato in glasshouse production though if dispersed the whitefly adults from both biotypes moved equally over a short range (Matsuura and Hoshino 2008) Distribution of whiteflies has bee n shown to be altered by insecticide application as well (Liu et al. 1993b, Tonhasca et al. 1994) Liu et al (1993b) examined distribution of B. tabaci on poinsettia in the greenhouse and those immatures surviving an insecticide application we re less likely to be aggregated than those that were not sprayed with an insecticide In field populations in cantaloupe, whiteflies of all stages were aggregated although the results for individual aggregation indices differed (Tonhasca et al. 1994 ) Other indices used to examine dispersion of insects and plant diseases are discussed in further detail in subsequent sections. Improved methods for studying distribution of B. tabaci in and around c rops need to be developed and incorporated into a la ndscape level analysis similar to those applied to other pests (Carriere et al. 2006, Reay Jones et al. 2010) Migration and Dispersal Migration is defined persistent and straightened out movement effected by the by its active embarkation on a vehicle. It depends on some temporary inhibition of station keeping responses, but promotes their (Kennedy 1985) Flight that is not migratory is

PAGE 33

33 termed trivial flight and is associated with short duration fligh ts between hosts. Dispersal is movement that increases distance between individuals and can encompass both migratory and tri vi al flight. Insect m igration and dispersal research has been focused on long range migrations and movement of moths and locusts. Long range dispersal is widely understood as an important economic threat in some agricultural systems (Showers 1997) Short range dispersal (< 5 10km) o f insects classified as weak fl i ers has become important for pests such as homopterans (Aleyrodidae and Aphididae) because of their economic importance in many agricultural areas. Fortunately, much short range dispersal research has been conducted on B. tabaci in order to develop IPM programs for the insect. Long distance flight peaks during morning hours between 06.00 and 10.00 hou rs (Gerling and Horowitz 1984, Blackmer and Byrne 1993b) Both males and females are capable of sustaining flights longer than 2 hours and up to 7km (Byrne 1999) The distribution o f whitefly adults in flight has been described as bimodal with most of the population of dispersing B. tabaci traveling under 2.7 km. These trivial flying insects traveling under 2km responded to vegetative clues and landed, indicating that local populati ons of whiteflies surrounding agricultural fields are important for managing B. tabaci (Byrne 1999) In some areas, whitefly hosts are found in close proximity to planted crops which could allow for very short dispersal distance and account for the majority of the immigrating population (Coudriet et al. 1985, Cohen et al. 1988, Byamukama et al. 2004) Cohen et al. (1988) captured whiteflies in a mark recapture study up to 7 km away, suggesting that the dispersal range may be much further than described by Byrne et al. (1996) Though vertical whitefly flight is strongly biased towards low altitudes (< 2 meters) regardless of

PAGE 34

34 sex (Isaacs and Byrne 1998) whiteflies have been captured at heights above 36 meters (Byrne 1999) These high flying whiteflies could be influe nced by prevailing winds resulting in higher concentrations of insects leeward of windbreaks (Cohen et al. 1988, Pasek 1988) Whitefly flight activity is influenced by endogenous factors such as age, sex and nutritional status. Age effect studies on B. tabaci showed that younger insects between 3 5 days of age had a greater propensity to takeoff, and exhibited phototactic orientation and increased flight duration (Blackmer and Byrne 1993b) Deterioration of indirect flight muscles and mitochondria in the thorax has been suggested as a cause of reduce d flight propensity in older individuals (Blackmer et al. 1995a) Though long duration flights (> mmon in the study population ( 10%) there were differences between the sexes in long duration flight (Blackmer and Byrne 1993b) L ong duration flights occurr ed only in the morning with females, whereas males flew throughout the day (Blackmer and Byrne 1993b) Blackmer and Byrne (1993b) also looked at the effect of host quality on flight and found that diet did influence flight patterns. Adult insects that were raised on poor quality hosts were more likely to take flight earlier, had a narrower window of flight age and initi ated longer phototactic flights. I nsects reared on high qualit y hosts had a longer period for flight and fl ew longer but were less responsive to phototactic cues (Blackmer and Byrne 1993b) Exogenous factors such as temperature, wind speed an d solar radiation also have an e ffect on flight in B. tabaci Temperature was the best predictor of flight activity which increased with rising tem perature and peaked at 24 27 C (Blackmer and Byrne 19 93a, Riis and Nachman 2006) Higher wind speed is deleterious to whitefly flight,

PAGE 35

35 although whitefly adults are able to fly against head wind s up to 30 cm/s (Isaacs et al. 1999, Riis and Nachman 2006) Light is also a cue in initiating flights in B. tabaci and flight propensity increases with solar radiation up to 0.73 kW/m (Riis and Nachman 2006) Blackmer and Byrne (1993b) recorded the longest flights of whitef lies in the early morning hours and other researchers have suggested that flight is retarded during night hours between 18.00 and 07.00 h (Byrne and von Bretzel 1987, Bellows et al. 1988) B tabaci has other cha racteristics suggesting that it could be considered a migratory species. For example, in some insects a polymorphic population emerges with migratory individuals (Palmer 1985) Byrne and Houck (1990) found wing polymorphisms in fi eld populations of B. tabaci in which males that left the host plant flight. Further research by Blackmer et al. (1995b) contradicted Byrne and Houck (1990) but did indicate that males with larger wings were more l ikely to take part in long duration flights than males with smaller wings T here were no wing morphology differences of females in either study. Another principle of a migratory species is the presence of oogenesis flight syndrome which is characterized by the delay of reproductive activity (usually the production of eggs) in favor of using resources for flight (Johnson 1969, Liquido and Irwin 1986) There has b een no evidence of oogenesis flight syndrome found in B. tabaci (Tu et al. 1997, Byrne 1999) as in other insects considered migratory (Sappington and Showers 1992) Bemisia adults could be considered migratory b ecause they can: take off and ignore vegetative cues, have flights over 2 hours and fly against the wind (Byrne 1999) B adult flight distances

PAGE 36

36 however, are not as great as those of some insects (Ritchie and Pedgley 1989, Showers 1997, Westbro ok et al. 1997) Tomato yellow leaf curl virus Biology and Distribution Tomato yellow leaf curl virus (TYLCV) has a circular, single stranded DNA (ssDNA) genome that has a single genomic component of approximately 2.8 kb and is transmitted by B tabaci in a circulative and persistent manner. Symptoms appear on tomato 2 4 weeks after inoculation and can vary based on virus isolate, host genetic background, environmental conditions, the growth stage and physiological condition of the plant (Io annou 1985, Rom et al. 1993) Symptoms in tomato include upward curling of leaflet margins, reduction of leaflet size, yellowing of younger leaves, stunting and flower abortion (Pico et al. 1996, Moriones and Navas Castillo 2000) These symptoms can lead to reduction of yields and, if the plant obtains the virus early in the gro wth cycle, production is lost almost entirely due to reduction of leaf surface and flower abscission (Levy and Lapidot 2008, Mohamed 2010) TYLCV has had a highly economic impact on tomato production around the world. Since being first observed in Israel in the early 1940s and with the d estruction of the tomato crop in the Jordan Valley in the 1960s, TYLCV has been a limiting factor of tomato production across the globe (Varma and Malathi 2003) From Israel through the Middle East and Asia (Makkouk 1978, Navot et al. 1989, Czosnek et al. 1990) TYLCV spread into Africa (D'H ondt and Russo 1985, Czosnek et al. 1990) Europe (Ioannou 1985, Czosnek et al. 1990, Moriones et al. 1993) and then to the Americas (Nakhla et al. 1994, Polston et al. 199 4, Ascencio Ibanes et al. 1999) In the United States, it was first reported by Polston et al. (1999) in Florida and later was found in Georgia (Momol

PAGE 37

37 et al. 1999) and Louisiana (Valverde et al. 2001) The spread of geminiviruses including TYLCV is associated with increased outbreaks of the B biotype of B. tabaci (Polston and Anderson 1997, Rybicki and Pietersen 1999) Hosts TYLCV has been shown to infect tomato and at least 30 other plant species in over 12 plant families (Polston and Lapidot 2007) Host s range from cultivated crops to weeds and their importance for TYLCV management is subject to availability in tomato growing regions. Identification of hosts both in cultivated crops and weeds is important for management of TYLCV. Tomato is the most im portant host of TYLCV but other cultivated crops have been shown to be hosts. Lisianthus, Eustoma grandiflorum (Raf.) Shim. is a host of TYLCV in Israel and TYLCV has become a limiting factor of lisianthus cultivation (Cohen and Gera 1995) Common bean P haseolus vulgaris L., has been shown to be a host of TYLCV Is in Spain (Navas Castillo et al. 1999) Also in Spain, toba cco, Nicotiana tabacum L. has been labeled as a host of TYLCV (Font et al. 2005) Tomatillo Physalis philad e lphica ( Lam ) has been identified as a host of TYLCV in Sinaloa, Mexico (Gamez Jimenez et al. 2009) With regard to pepper, t here has been some debate in the literature concerning its host status Morilla et al. (2005) were unable to transmit TYLCV from infected pepper ( Capsicum spp.) plants using B. tabaci biotype Q On the other hand Polston et al. (2006) demonstrated some genotypes of pepper could serve as symptomless reservoirs fo r TYLCV transmission to tomato with B. tabaci biotype B These latter authors suggested the differences in results between the two studies could be based on the differences in cultivars used the number of viruliferous whiteflies used for inoculation and the ability of biotype Q and B to feed and acquire TYLCV in pepper.

PAGE 38

38 Anfoka et al. (2009) showed that B. tabaci could tran smit TYLCV M ld from cucumber, Cucumis sativus L. to tomato, al though transmission with TYLCV I s was unsuccessful. I t was recommended that trap crop systems using cucurbits could a ffect tomato yellow leaf curl disease ( TYLCD ) epidemics. S quash Curcubita pepo ( L. ) wa s determined to be a host of TYLCV in Cuba (Martinez Zubiaur et al. 2004) In Florida, a survey was conducted to identify weed reservoirs of TYLCV and no weed samples were found to harbor the virus (Polston et al. 2009) W eeds c an also play a role in TYLCV epidemiology. In Israel Cynanchum acutum L. was identified as the most important host of TY LCV and a good source of inoculum (Cohen et al. 1988) Datura stramonium L. has been i dentified as a symptomatic host of TYLCV in the Dominican Republic (Salati et al. 2002) and other tomato production areas (Cohen and Nitzany 1966, Mansour and Al Musa 1992, Cohen and Antignus 1994) Naturally infec ted M parviflora is an annual weed host in Israel (Cohen et al. 1988) and also in the Dominican Republic al though the viral titer was very low in tested plants (Salati et al. 2002) Other weeds shown to be hosts of TYLCV around the world include: Hyoscyamus desertorum (Asch.) Eig, Nicotiana benthamiana N. glutinosa Solanum nigrum Mercuri alis ambigua Cleome viscosa Croton lobatus Physalis spp., Macroptilium spp., Bastardia spp., Eupho r bia spp., and Polygonum spp. (Mansour and Al Musa 1992, Cohen and Antignus 1994, Sanchez Campos et al. 2000, Gilb ertson et al. 2007) Other hosts such as Chaerogphyllum spp., Lens esculenta ( Moench ) M nicaensis ( All. ) N tabacum P vulgaris and S oleraceus are considered hosts a l though these plants were not infected under natural conditions (Cohen and Antignus 1994) Weeds can serve as good source s of inoculum of TYLCV and be important

PAGE 39

39 reservoirs of genetic diversity for TYLCD associated virus populations (Garcia Andres et al. 2006) TYLCV Pla nt Relationship After injection into the phloem by B. tabaci, TYLCV repli cates in infected cell nuclei and spreads systemically th r ough the plant. After the whitefly injects its stylets intercellularly between epidermal cells, virions are usually deposite d into the sieve elements (SE), al though in some cases they are deposited into companion cells or vascular parenchyma cells (Pollard 1955, Wege 2007) For replication, the genomic D NA must enter a nucleus via coat protein (CP) mediat ion of the TYLCV genome (Kunik et al. 1998, Rojas et al. 2001) After entering the nucleus viral DNA replicates by a rolling circle mechanism (Laufs et al. 1995) and moves systemica lly through the plant via sieve tubes assisted by the capsid and movement protein s (Gronenborn 2007, Wege 2007) TYLCV moves systemically through the plant and accumulates preferentially in the tissues containing dividing cells. Viral DNA can be detected at the inoculat ion site 4 5 days post infection and reaches a peak 12 15 days post inoculation (Rom et al. 1993) Maximum viral accumulation occurs in the developing leaves close to the ap ex and the axillary shoots while lower viral accumulation occurs in the stems, roots and expanding leaves (Pico et al. 1996, Wege 2007) Symptoms are expressed tomato 2 3 weeks post infection (Rom et al. 1993) TYLCV B. tabaci Relationship Cohen and Harpaz (1964) described the TYLCV B. tabaci interaction as periodic acquisition due to the loss of inoculative potential of the insect after initial acquisition of the virus. It is now understood that age and sex of B. tabaci affects the efficacy of

PAGE 40

40 a cquisition and transmission of TYLCV. Female whiteflies transmit TYLCV with higher efficiency than males (Cohen and Nitzany 1966) Also, as the insect ages TYLCV inoculation efficiency decreases (Cohen and Nitzany 1966, Mans our and Al Musa 1992, Rubinstein and Czosnek 1997) Nymphs can acquire the virus and transmit TYLCV in the adult stage thus support ing the circulative mode of transmission (Cohen and Nitzany 1966, Mehta et al. 1994) A ssociation of B. tabaci with TYLCV was shown to lead to a reduction in life expectancy (Rubinstein and Czosnek 1997) Mehta et al. (1994) demonstrated that virus titers increase over time in B. tabaci and virus concentration is higher in the whitefly than in the plant TYLCV could be a pathogen of B. tabaci because of virus replication in the insect (Moriones and Navas Castillo 2000) The adverse effects of the virus on the insect includ ed reduced life expectancy and fecun dity (Czosnek et al. 2002) F or successful TYLCV acquisition B. tabaci must be feeding in the phloem where the virus is located. Using electrical penetration graph (EPG) te chniques strong correlations were found between E(pd) 1 (regarded as ingestion from phloem, (Jiang et al. 1999) ) and virus inoculation (Jiang et al. 2000) Transmission rate increases with numbers of viruliferous whiteflies feeding on the plant with a minimum of ten to 20 insects needed for 100% transmission (Pico et al. 1996) Mehta et al. (1994) concluded that the efficiency of transmission increased fourfold after the number of adults were increased to five per plant. The minimum acquisition access per iod (AAP) for transmission of TYLVC by B. tabaci was 15 30 min (Ioannou 1985, Mansour and Al Musa 1992, Mehta et al. 1994, Caciagli et al. 1995, Muniyappa et al. 2000, Czosnek et

PAGE 41

41 al. 2002) W ith longer AAPs ther e were higher rates of transmission peaking at 24hrs (Mehta et al. 1994) After ingestion, TYLCV is n o t immediately available for transmission due to an approximately 8 h latent period (Ghanim et al. 2001) The latent period allows for translocation of the virus particles out of the digestive tract and into the salivary glands. The latent period is not solely based on the rate of movement of virus in the insect but rather defi ned as the time from acquisition to transmission of the virus into plant material (Czosnek 2007) Ghanim et al. (2001) stated that TYLCV DNA could be found in the insect s head, midgut and hemo lymph using PCR within 10, 40, and 90 min, respect ive ly h after AAP. The latent period was reported by Cohen and Nitzany (1966) to be 21 h As virus particles move through the body of the insect they pass through the stylets, esoph agus, filter chamber, midgut, h emolymph and finally into the salivary system. This pathway is dependent on the transition through the midgut wall and into the saliva ry glands. To cross into the h emolymph from the midgut it is suggested that the microvilli along the epithelial cells wi thin the gut wall have begomovirial receptors (Ghanim and Medina 2007, Ohnishi et al. 2009) While in the h emolymph, TYLCV may be protected from degradation by an endosymbiotic bacteria or GroEL homologue (Morin et al. 1999) Transition into the salivary glands is less understood and the method of transportation from the haemolymph into the salivary glands is unclear (Ghanim and Medina 2007) Entry of virus particles into the salivary gland s does not guarantee transmission (Caciagli et al. 2009)

PAGE 42

42 It has been shown that the sex and age of the whitefly affect the transmission efficiency of TYLCV. Female wh iteflies are more efficient at transmit ting TYLCV ; as females 1 to 2 weeks old infected of the plants during a 48 hour inoculation access period (IAP) while males infected 20% of the plants during the same time span (Czosnek et al. 2001) ation times of TYLCV particles within the whiteflies but may be due to differences in viral titers in the salivary glands (Ghanim et al. 2001) Age also affected the efficiency of transmission of TYLCV wit h the younger adults being better at transmitting the virus (Ioannou 1985, Mansour and Al Musa 1992, Caciagli et al. 1995, Czosnek et al. 2001, Czosnek 2007) In females the percentage of infection went down from to 60% in 3 we ek old and 20% in 6 week old adults The percentage of infection by males went from 20% for w eek 1 to 2 week old males to 0% in 3 week old males (Czosnek et al. 2001) Other researchers have found that B. tabaci retains the ability to transmit TYLCV until around 8 12 days after AAP (Ioannou 1985, Mansour and Al Musa 1992, Caciagli et al. 1995) Rubinstein and C zosnek (1997) measured the amount of virus acquired by different aged individuals and found that as the whitefly aged (10 to 17 days) the amount of virus accumulat ed by feeding was reduced by half. TYLCV likely remains associated with B. tabaci during the entire life of the vector (Rubinstein and Czosnek 1997) The association of TYLCV with B. tabaci is ultimately harmful to B. tabaci and resulted in a reduction of 17 23 % in life expectancy and a 40 50% reduction in fecundity (Rubinstein and Czosnek 1997) Whether or not TYLCV actually replicates within the whit efly is subject to debate. Harrison (1985) suggested that geminiviruses do not replicat e in their insect vectors. Other researchers have

PAGE 43

43 shown that TYLCV DNA is detectable in the adult body longer than the whitefly might actually be infected (Caciagli et al. 1995, Rubinstein and Czosnek 1997, Siniste rra et al. 2005) Mehta et al. (1994) concluded TYLCV accumulated and replicated in B. tabaci Czosnek et al. (2001) noticed an increase in TYLCV DNA up to 16 h and Czosnek (2007) su ggested that the increase of TYLCV DNA was due to the ingestion of viral replicative complexes. TYLCV can also be transmitted transovarially for at least two generations (Ghanim et al. 1998) al though other authors have suggested that transovarial transmission is more complicated than previously thought (Goldman and Czosnek 2002, Bosco et al. 2004) Also TYLCV can be transmitted horizontally between individuals of the same biotype through contamination of the h emolymph without passing through the midgut barrier (Ghanim et al. 1998, Ghanim et al. 2007) Management In most tomato growing areas TYLCV management is highly dependent on chemical control targeting the adult B. tabaci (Polston and Lapidot 2007) Conventional insecticides such as organochlorines, organophosphates, c arbamates, pyrethroids, formamidines and cyclodienes have been used historically for whitefly control (Sharaf 1986, Palumbo et al. 2001) Most insecticides used f or control of B. tabaci feeding quickly enough to stop transmission of TYLCV T he number of insects necessary for field epidemics is usually very low and the efficiency of TYLCV transmission is high (Pico et al. 1996) Development of newer chemistries with novel modes of action like the neo nicotinoids, and pymetrozine have given growers more choices for managing B. tabaci adults (Palumbo et al. 2001) The neonicotinoid, i midacloprid, i s an effective chemical for early season control of TYLCV due to its long lasting systemic activity (Ahmed et al. 2001, Attard 2002, Polston and Lapi dot 2007)

PAGE 44

44 Another neonicotinoid, thiamethoxam has been shown to reduce TYLCV incidence by preventing virus transmission by B. tabaci (Mason et al. 2000) Unfortunately, B. tabaci adults have developed t olerance to both of these neonicotinoids in Florida (Schuster and Caballero 2010) and elsewhere (Byrne et al. 2003, Horowitz et al. 2004) Pymetrozine a pyridine azomethine has the ability to interfere with whitefly feeding behavior and can reduce TYLCV infection by up to 7 days (Polston and Sherwood 2003) Unfortunately, with the judicious use of insect icides long term control of B. tabaci has been difficult to maintain with certain chemistries and resistance issues have hampered TYLCV pesticide control tactics (Horowitz and Ishaaya 1996, Horowitz et al. 2007) Use of resistant or tolerant tomato varieties is one of the best approaches to reduce losses due to infection of TYLCV (Pico et al. 1996, Moriones and Navas Castillo 2000, Lapidot and Friedmann 2002) Since searching for resistance in the cultivated tomato ( S. lycopersicum ) failed, breeding programs have been based on the in trogression of resistance from wild Solanum species including; L. peruvianum (Rom et al. 1993, Friedmann et al. 1998) L. chilense (Michelson et al. 1994, Zamir et al. 1994, Scott et al. 1996, Pico et al. 1999) L. pimpinellifoliu m (Vidavsky et al. 1998) and L. hirsutum (Vidavsky and Czosnek 1998) Vidavsky et al. (2008) pyramided genes from wild tomato species in order to have more than one source of resistance. Data suggests that under high inoculum pressure some resistance can be overcome (Michelson et al. 1994, Pico et al. 1996) R esistant cultivars must be carefully managed in order to reduce the selection of TYLCV variants that could possibly overcome current resistant cultivars and alter the virus population structure (Seal et al. 2006)

PAGE 45

45 Unfortunately in most tomato production areas including Florida, growers are reluctant to use currently available resistant cultivars due to lack of bacterial and fungal resistance, lack of horticultural properties and lower than expected yi elds (Polston and Lapidot 2007) Also, current resistant cultivars are tolerant, deteriorat e as badly as hig hly susceptible cultivars, and harbor the virus. T hus, the extent of virus inoculum in the field may be underestimated and the c ultivars could serve as virus reservoir s (Lapidot et al. 2001) Genetic engineering approaches show good promise for the future and research is being conducted in order to better understand the mec hanisms and approaches for incorporating genetically engineered resistance into tomato. These genetic techniques confer resistance through expression of viral capsid protein (Kunik et al. 199 4) altered viral Rep protein (Brunetti et al. 1997, Brunetti et al. 2001, Antignus et al. 2004, Yang et al. 2004, Praveen et al. 2005) GroEL gene (Akad et al. 2007) and post transcriptional gene silencing (Abhary e t al. 2006) Cultural control tactics are useful in managing TYLCV and have been used in tomato production areas in both greenhouse and open field production. T he use of virus free planting material along with avoid ance of adjacent TYLCV reservoirs thr ough time and space can limit the amount of initial virus inoculum (Ioannou 1987, Cohen et al. 1988, Lapidot et al. 2001, Polston and Lapidot 2007) To avoid TYLCV inoculum sources, m andatory crop and host free pe riods were established in the arid Arava region in Israel (Ucko et al. 1998) and the Dominican Republic (Polston and Anderson 1997, Salati et al. 2002) Ph ysical barriers have been used for tomato protection in the Mediterranean since 1990 (Berlinger and Lebiush Mordechi 1996, Berlinger et al. 2002) and other areas (Bethke et al. 1994, Arsenio et al. 2002) Screening greenhouses with

PAGE 46

46 whitefly exclusion material in Israel was very cost effective (Taylor et al. 2001) Ultraviolet absorbing plastic films have been used in protected culture and have shown good results (Antignus et al. 1998, Antignus et al. 2001) Unfortunately these p rotective covers can cause problems in tomatoes in hot temperatures and can increase the spread of foliar diseases. Proper ventilation and cooling systems are needed to avoid overheating (Weintraub and Berlinger 2004) In open field product ion areas which currently includes Florida, UV reflective soil mulches have been widely adopted as a control tactic for reducing TYLCV settling of B. tabaci adults on tomato plants (Csizinszky et al. 1997, Csizinszky et al. 1999) In Israel, yellow plastic mulch reduced the number of B. tabaci adults on tomato plants because adults were attracted to the yellow color and were then rapidly dehydrated by the high temperature of the mulch (Cohen 1982, Cohen and Berlinger 1986) The effectiveness of yellow mulch was not corroborated in Florida tomatoes however (Csi zinszky et al. 1997) Weeds can also serve as a host of TYLCV and good in field and field edge weed management is suggested though beneficial insects and pathogens of B. tabaci can inhabit these areas and can provide control in certain situations. Rogu ing is also recommended and plants with early symptoms can be removed to reduce the inoculation source within the field and reduce secondary spread (Polston and Lapidot 2007) Bait (trap) crops have also been used as control tactics and altern at ing rows of cucumbers and tomatoes delayed the spread of TYLCV for 2 months (Al Musa 1982) In Florida, similar results were found when squash was used as a bait crop (Schuster 2004) Cucurbit crops can serve as reservoir s of Tomato yellow leaf curl Sardinia virus

PAGE 47

47 (TYLCSV) and TYLCV Mld and in areas of the world where these viruses exist, it is suggested that growers do not use trap crop systems (Anfoka et al. 2009) Geographic Information Systems and Spatial Statistics GIS Geographic information s ystems (GIS) are tools for studying and mapping the spatial relationship of unknown variables Data are stored in a GIS using both vector and raster based models and combined with define d location s in a geographic coordinate system (Kennedy 2000) GIS can al so be used to create maps that show a picture of pest populations and visually express point data (Nelson et al. 1999) Area with the influence of GIS and geostatistics ecological, pathological and entomological questions on large r scales will be easier to identify and research GIS and geostatistics have been used to monitor and predict pest populations and amend pest management strategies in medical, entomological and plant disease systems. This review will focus on the use of GIS in the entomology and plant disease management context For review of GIS uses in medical and veterinary entomology see Thomson and Connor (2000) Noonan (2003) and Patz and Confalonieri (2005) To be considered a GIS, the system must be able to input, store, retrieve, manipulate and report spatial data (Bolst ad 2005) Spatial data can be collected in many forms and may include paper maps, remotely sensed data, digital line graphs or field acquired point data (Liebhold et al. 1993) Spatial data input is essential to a GIS and most systems can import from a variety of data formats. Once in a GIS data are stored and retrieved in two main models, vector and raster. Raster data use a grid cell data structure where each cell has a value and groups of adjacent cells with identical

PAGE 48

48 values define a spatial object. The raster data format is good for mathematical modeling and representing continuous variables. Unfortunately, the size of the grid cells determine the resolution at which data are represented and the boundaries can tend to look blocky rather than smooth like vector data representations Vector data represents geographic features with points, lines, curves or areas. Vector data use less storage space and are good at representing linear features such as roads, data points and boundaries between objects. Image data can also be imported into a GIS and can include background information for spatial maps. A GIS defines locations with a geographic coordinate system or x, y coordinates for a given map layer. Coordinates are usually acquired by a global positioning system (GPS) device. GPS units are usua lly handheld and connect to satellites to calculate location. Latitude and longitude is the most commonly used geographic coordinate system along with Universal Transverse Mercator (UTM) which is an adaptation of the Mercator projection and is based on d istance in meters (Noonan 2003) These coordinates must be expressed on a 2 D flat surface such as a computer monitor or map so they need to be projected, which distorts the map to some degree. Pest density maps can be cr eated in a GIS using interpolation methods to mathematica lly estimat e values at unsampled locations (Fleischer et al. 1999) Some commonly used interpolation methods are inverse distance weighted (IDW) and kriging. The ability of maps to show pest densities and, thus, to better measure and unde rstand spatial variation is valuable to researche r s and crop managers. They can be used to indicate hot spots or drive pest management tactics to control local populations.

PAGE 49

49 Geostatistics Geostatistics quantifies and models spatial and temporal data and predicts the value of a variable at unsampled locations Geostatistics can also model the uncertainty about those unknown values. Isaaks and Srivastava (1989) provide an introduction to applied geostatistics The central theme of geostatistics rests on the expectation that closer objects are more related tha n objects fa rther apart. Dispersion patterns can provide useful information about population structure but ignore spatial location of samples and assume independently distributed data (Rossi et al. 1992) One can also characterize dispersion patterns by quantifying spatial dependence with semivariograms. Se mivariograms use distance between data points to create a graph of the spatial dependence of attribute values and serve as a tool to find the range of spatial autocorrelation. Spatial autocorrelation is a correlation of a variable with itself through spac e and thus helps describe patterns of variables. Interpolation methods estimate predicted values at unsampled locations based on values at sampled locations. IDW interpolated values are determined using a linear weighted combination of observed values a nd those weights are functions of the distance between locations. Unlike other interpolation methods IDW does not require a variogram model and is appropriate for small data sets (Kravchenko 2003) IDW values are estimated by: (2 1) where Z j is the estimated value for the unknown point at location I d ij is the distance from known point i to unknown point j Z i is the value for the known point i and n is a user defined exponent (Equation 2 1) The fa rther away the point the smaller the

PAGE 50

50 weight, so the less influence it has on the estimate of the unknown point. Care must be taken when n (number of samples) is selected because when a larger n is used the closer points become more influential. Cross validation ha s been used to estimate the fi t of the IDW model. Cross validation removes one sample point at a time and compares observed and predicted values for that point (Isaaks and Srivastava 1989) The root mean square prediction error (RMSE) produced by cross validation has been presented as the summary statistic to check the accuracy of the model produced in IDW (B onsignore et al. 2008, Tillman et al. 2009, Reay Jones et al. 2010) IDW is a good tool for initial analysis and because of its simplicity it has been used extensively in the literature to create interpolation maps (Bonsignore et al. 2008, Tillman et al. 2009, Reay Jones et al. 2010) Geostatistical theory was first developed for geology and mining to estimate ore and mineral quantities (Vieira et al. 1983, Goovaerts 1997) Entomologists have focused on d escribing spatial patterns of insects d ue to their spatially heterogeneous populations S ome commonly used dispersion indices ignore valuable information such as the spatial location of samples (Rossi et al. 1992) Geostatistics allo ws for the description of spatial pat terns using s patial locations. To display the spatial dependence of an organism a semivariog ram or variogram is produced. A variogram is a graph that expresses the variance of sample pairs against the distance betwee n sample points and is defined as: (2 2) h ) is the estimated semivariance value for lag h N ( h ) is the number of pairs of points separated by h z( x i ) is the variable as a function of spatial location and z ( x i h )

PAGE 51

51 is the lagged version of the variable (Equation 2 1 ). The variograms can be evaluated as an average over all directions or in a specific direction. A general va riogram is presented in Figure 2 1. T he centra l theorem of geostatistics states that objects closer are more related than those far away ; therefor e as semivariance increases with distance it eventually levels off and becomes constant (Figure 2 1). At the y intercept there should be no variability b etween a sample and itself though when extrapolated to lag zero the y intercept is commonly greater than zero. This value of the y intercept is termed the nugget and was established by gold mining engineers who found gold nuggets not spatial ly associat ed with ore deposits (Liebhold et al. 1993) This nugget represents two sources of variability: spatial variability at a scale smaller than the minimum lag distance and experimental error or the human nugget. Sometimes variograms appear horizontal which indicate s a complete lack of spatial structure These variograms are termed pure nugget variograms. The sill is the point at which the curve levels off and is usually equivalent to the sample variance. The distance at which the sill levels off is called the range and is commonly used to express the di stance at which spatial dependency is lost. The general rule of thumb is that there are at least 30 50 pairs of points needed to create variograms al though the more points the greater the statistical reliability. Also, only half the total distance measu red in any direction over the sampling area may be represented legitimately in a variogram. Variograms may be omnidirectional or calculated for specific directions S imilar spatial continuity with direction is known as isotropy. Variograms can be greatl y affected by outliers as

PAGE 52

52 variograms assume local means and variances are sta tionary across the study area (Rossi et al. 1992, Fleischer et al. 1999) The modeled variogram can be used for many purposes including spatial prediction. Spatial prediction tools such as kriging can estimate values by taking a wei ghted linear average of available samples. Values are determined using a linear weighted combination of observed values and those weights are functions of the distance between locations. Unlike IDW, k riging prediction maps use geostatistical methods and are based on statistical models that include autocorrelation. Kriging can provide a certainty and accuracy of the predictions even though it is more complicated to produce and requires data exploration before a map can be produced One criticism of the utility of geostatistics in applied agriculture is the high level of understanding of mathematical concepts. GIS Uses in Entomology Historically, applications of GIS in entomology have been limited to forest and rangeland entomology (Kemp et al. 1989, Schotzko and O'Ke effe 1989) Recently, GIS has been applied to manage insect pests in agricultural systems (Barnes et al. 1999, Park and Obrycki 2004, Carriere et al. 2006, Garcia 2006) Using GIS, entomologists have able been t o relate insect populations to biological and physiographic features of the landscape (Shepherd et al. 1988, Van Sickle 1989, Bryceson 1991) Spatial dependence in insects shows that interpolating counts at un sam pled locations can be valuable (Borth and Huber 1987, Schotzko and O'Keeffe 1989, Liebhold et al. 1993, Setzer 1995) GIS technology relates questions of insect ecology with their spatial components. In rangelan d entomology locust populations and their subsequent outbreaks can be

PAGE 53

53 modeled and pest density maps can be created (Cigliano et al. 1995, Schell and Lockwood 1997) Locust populations have been studied on an area wide scale and locust counts have been predicted at these scales using kriging and geostatistics (Kemp et al. 1989) GIS have correlated grasshopper populations with their distribution through different landscapes (Kemp et al. 2002) Locust population outbreaks have also been evaluated for dependency on ecological variables (Schell and Lockwood 1997) Fores t pests have been subjected to analysis with GIS including gypsy moths Lymantria dispar ( L.) Data acquired from a GIS by Sharav et al. (1996) quantified spatial variation of the gypsy moth which could be used to develop IPM strategies aimed at monitoring and trapping. The efficacy of aerial applications of insecticides ag ainst gypsy moth was evaluated using a GIS (Liebhold et al. 1996) Liebhold et al. (1991) used semivariograms to model gypsy moth egg masses in forest landscapes and created krig ed prediction and threshold maps. Similar geostatistical tools including semivario gram production and cokriging were used to estimate gypsy moth egg mass abundance in Sardinia (Cocco et al. 2010) S imilar egg mass prediction studies were used for developing regional gy psy moth defoliations maps (Gribko et al. 1995) GIS and geostatistics have also contributed to studies in other forest pests including the southern pine beetle, Dendroctonus frontalis ( Zimmermann ) (Fitzgerald et al. 1994) a nd spruce budworm, Choristoneura fumifera na (Clemons) (Candau et al. 199 8, Lyons et al. 2002) GIS technologies and other geostatistical tools have been used to examine insect populations in agricultural systems as well. Studies have shown that most arthropod species are spatially aggregated (Taylor et al. 1978, Wilson and Room 1983) GIS and

PAGE 54

54 geostatistical programs can help determine the spatial distribution of insects which can help with management and prediction of insects at unsampled locations. Insect distributions have been mapped for statewide monitoring of European corn borer, Ostrinia nubilalis (Hbner) and corn earworm, Helicoverpa zea (Boddie) in New Jersey (Holmstrom et al. 2001) Using kriging analysis t he bollworm H armigera (Hbn er) exhibited an edge effect in a tomato field in Spain (Garc ia 2006) Other researchers showed clustering of H. armigera in cotton, though patterns changed with the population density (Ge et al. 2005) The pink bollworm, Pectinophora gossypiella (Saunders) and its spatial dynamics were ex a mined visually by kriged maps (Borth and Huber 1987) The spatial dynamics of the potato tuberworm, Phthorimaea operculella (Zeller) were described by Debano et al. (2010) In corn, aggregation of the fall armyworm, Spodoptera frugiperda (Smith) diffuses throughout the season (Farias et al. 2008) The western tarnished plant bug, Lygus hesperus (Knight) was shown in lentils to be aggregated in the early season and to be uniformly distributed in mid season (Schotzko and O'Keeffe 1989) G eostatistical tools were also used to determine sample placement f or L hesperus to assist with management (Schotzko and O'Keeffe 1990) Stinkbugs species including; the green stinkbug, Acro sternum hilare (Say); brown stinkbug, Euschistus servus (Say); and the southern green stinkbug, Nezara viridula (L.) demonstrated aggregated spatial patterns (Reay Jones et al. 2010) IDW maps were created to visually express stinkbug populations along edges of cotton associated with peanuts (Tillman et al. 2009) Populations of the sharpshooters Dilobopterus costalimai (Young), Acrogonia sp. and Oncometopia facialis (Signoret) which vector Xylella fastidiosa were fou nd to be aggregated during the summer, winter and spring (Farias et

PAGE 55

55 al. 2004) Aggregated patterns and anisotropy in the direction of rows in vineyards were found for Lobesia botrana (Den is and Schiffermller) in Northern Greece (Ifoulis and Savopoulou Soultani 2006) Byrne et al. (1996) used geostatistics to describe patchy distribution of migrating whiteflies. Much work has been conducted on spatial variation of beetle populations in corn and other crops. Aggregation of the sugarbeet wireworm, Limonius californicus (Manner heim) using geostatistical tools was shown by Williams et al. (1992) Aggregated spatial distribution s of northern corn rootworm, Diabrotica barberi (Smith and Lawrence) and western corn rootworm, D. virgifera virgifera (Le conte) were found in South Dakota corn (Beckler et al. 2005) Other resear chers found similar aggregated spatial structures in D. barberi and D. virgifera virgifera and found the range of spatial dependence to be lower in D. barberi (Ellsbury et al. 1998) Aggregated populations of D. virgifera virgifera were also presented by Midgarden et al. (1993) and Park and Tollefson (2005) Site specific IPM for Colorado potato beetle, Leptinotarsa decemlineata (Say) was designed using kriging and IDW maps (Weisz et al. 1995) Predaceous lady beetles Harmonia axyridis (Pallas), Coleomegilla maculata (DeGeer) and Coccinella septempunctatata ( L. ) and their prey the corn leaf aphid, Rhopalosiphum maidis (Fitc h) are aggregated during peak populations and randomly distributed in the early and late season (Park and Obrycki 2004) The cereal leaf beetle, Oulema melanopus (L.) has an edge effect and is spatially aggregated (Reay Jones 2010) The Mexican bean beetle egg mass distribution indicated an edge effect, but there were no ag gregated semivariogram (Barrigossi et al. 2001) The buprestid, Capnodis tenebrionis (L ) was

PAGE 56

56 spatially and temporally modeled using IDW in an apric ot orchard in Sicily and pest populations were found to be higher in crowns associated with sun exposure (Bonsignore et al. 2008) Environmental factors such as temperature and elevation can affect population dynamics as wel l. Using nonparametric multiplicative regression along with a GIS, increased trap counts of the potato tuberworm corresponded with increased temperature (DeBano et al. 2010) D ecreased trap counts of potato tuberworm corresponded with increas ed elevations and latitudes (DeBano et al. 2010) Crop phenology can affect D. virgifera virgifera spatial patterns in corn (Darnell et al. 1999) L. hesperus populations are affected by habitat types and distance to available habitats (Carriere et al. 2006) The authors used a GIS to evaluate the spatial arrangement of fields to help design and improve area wide management of L. hesperus (Carriere et al. 2006) V egetation types and their interactions with grasshopper species composition and distributio n were evaluated using GIS (Kemp et al. 2002) S tinkbugs demonstrate an edge effect and adjacent crops and landscapes can affect popu lations within a target crop (Tillman et al. 2009, Reay Jones et al. 2010, Reeves et al. 2010) Cereal leaf beetles are found aggregated on field edges in wheat fields, particularly on edges bordering corn (Reay Jones 2010) Strong spatial dependence in insect data indicates that estimating populations at unsampled locations is a possibility (Liebhold et al. 1993) Agroecosystem heterogeneity has a direct effect on pest population dynamics, dispersal and habitat selection and pest problems commonly extend beyond the boundaries of individual growers, thus illustrating the importance for developing regional or are awide pest

PAGE 57

57 management. Spatial distribution of insects is important for developing crop management tactics, and spatial m odels can help researchers and growers target pest insects for a successful integrated pest management system (Legaspi Jr. et al. 1998) However, g eostatistical analysis is complicated and specialized expertise can not be expect ed from every researcher or farm manager ; nevertheless, proper application s of spatial distribution of pest insects could lead to better sampling detection and management GIS Uses in Plant Disease Management GIS can be used in conjunction with geostatistics to better understand insect vectored plant diseases and the factors that influence their epidemics. Spatial patterns of disease can provide information such as direction and distance of spread and importance a nd proximity of the sources of inoculum and vectors (Thresh 1976) Research is commonly undertaken on the small scale plot, but GIS allows for regional scale studies and extrapolation of knowledge of small scale variat ion to a larger geographic area. Disease problems and their associated vectors commonly extend or should be addressed regionally. Area wide management of plant disease is valuable an d with the development of GIS systems, new research efforts can focus on spatial relationships of landscape features and management tactics on a much larger scale. Within field studies focus on fine scale variation and results acquired should set the ba sis for future work on larger regional studies. Understanding this field scale variability allows for extrapolation of knowledge into larger management areas. African cassava mosaic disease (ACMD) is caused by a whitefly vectored geminivirus and early wo rk with geostatistics indicated a spatially dependent structure that was influenced

PAGE 58

58 heavily by wind (Lecoustre et al. 1989) The influence of wind on th e distribution of ACMD was further suggested by Fargette et al. (1985) and corroborated by Colvin et al. (1998) Lack of aggregation of C itrus tristeza virus which is vectored by the aphids Aphis gossypii (Glover), A. spiraecola (Patch), Toxoptera aurantii (Fons.) and T. citricida was confirmed by geostatistical an alysis (Gottwald et al. 1996) Almond leaf scorch disease (ALS), caused by Xylella fastidiosa which is vectored by sharpshooter leafhoppers and spittlebugs had spatial aggr egation patterns in certain almond cultivars (Groves et al. 2005) a limiting disease in gr apes in California and other areas of the country is caused by the same causal agent of ALS In grapes in n s and showed anisotropy consistent with vine to vine spread related to the movement o f its vector a glassy wing sharpshooter, Homalodisca coagulata (Germar) (Tubajika et al. 2004) If plant diseases exhibit spatial autocorrelation beyond the boundary of a single field then regional management would be useful. There have been successful regional management programs for plant pathogens that were aided by the use of GIS and geostatistics. Cotton lea f curl disease caused by Cotton leaf curl virus vectored by a whitefly is under evaluation in Punjab, Pakistan and includes risk analysis of spatial landscape characteristics (Nelson et al. 1999) Similar work in Arizona with C otton crumple virus vectored by B tabaci could help with understanding the distribution of whiteflies and their associated vectored plant viruses (Nelson et al. 1999) There are some pitfalls with creating area wid e pest density or risk maps. If adjacent fields are highly dissimilar with abrupt differences, surface maps over larger regional areas can be misleading (Nelson et al. 1999) T he level of knowledge

PAGE 59

59 and cost associated with area wide management programs based on a GIS can be large (Nelson et al. 1999) A regional plant virus management program based on risk assessment and virus disease incide nce data was developed for the multivirus, multivec tor, disease complex of tomatoes in the Del Fuerte Valley, Sinaloa, Mexico (Nelson et al. 1994) Risk maps created by a GIS were correlated to disease incidence and were used as a decision tool for adapting disease management tactics (Nelson et al. 1994) Although GIS analysis was not required for implementation of management tactics for control of plant viruses and their insect vectors in the Del Fuerte Valley Sinaloa, Mexico the output from the GIS was instrumental in pro viding a regional perspective of the problems (Barnes et al. 1999) In Florida, a decision support system for management of B. tabaci and TYLCV is being developed from regional surveys using a GIS (Turechek 2010) The use of GIS coupled with integration of agroecosystem data at the regional level could improve management of pest problems at a much larger scale. Spatial Analysis by Distance IndicEs (SADIE) Most arthropods ar e spatia l l y aggregated in nature and many have been mapped with GIS technologies which use geostatistics to determine spatial structure. Entomological data sets are often patchy and can include a majority of zero values and be highly dynamic, which limit s the use of geostatistical methods and t hese limitat ions led to the development of SADIE (Perry 1995) SADIE measures the degree of clustering in georeferenced data. Unlike geostatistics, SADIE is based on discrete count data and results are conditional to the observed heterogeneity of the data (Xu and Madden 2004) The basis of SADIE is to quantify the degree of clustering by calculating the distance to regularity ( D ). Distance to regularity is defined as the minimum distance

PAGE 60

60 individuals in a sample would have to move to result in a uniform distribution. The overall aggregation index I a is defined as D/E a where E a is the mean expected distance to regularity for the randomized samples Values of I a > 1 suggest aggregation, I a < 1 suggest a regular pattern and I a = 1 indicates a random pattern (Perry 1998) SADIE was designed to detect clusters in patch or gap form which makes it sensitive to patterns when disease incidence is low (Dallot et al. 2003) SADIE also provides overall clustering indi ces j I indicating positive patch cluster to quantify the degree to which the count for each sample unit contributes towards the overall degree of clustering (Perry e t al. 1999) SADIE has been used to describe spatial patterns in insects and plant diseases. It can also be combined with GIS produced maps t o visualize the aggregation or uniformity of populations. Spatial and temporal dynamics of H coagulata and H. l iturata were examined using SADIE and trap counts were aggregated overall where H. coagulata and H. liturata were associated with their respective hosts citrus and desert saltbush scrub (Park et al. 2006) SADIE was used to determine that carabid beetles were spatially aggregated in winter oats in relation to food availability or microclimate (Korie et al. 2000) The s tink bugs A hilare E servus and N viridula and their associated damage to cotton bolls indicated an aggregated pattern in South Carolina and Georgia (Reay Jones et al. 2010) Other studies with N. viridula and E servus indicate d similar aggregation patterns u sing SADIE analysis (Tillman et al. 2009) In winter oilseed rape, the cabbage seed weevil, Ceutorhynchus assimil is (Payk.) was distributed in an aggregated pattern during colonization as the ins ect moved from the edges toward the center of the crop (Ferguson et al. 2000) Another pest of winter oil

PAGE 61

61 seed rape, the cabbage stem flea beetle, Psylliodes chrysocephala (L.) showed aggregation after immigration from the edges of the crop (Warner et al. 2003) SADIE was also used to evaluat e the spatio temporal dist ribution of carabid beetles [ Trechus quadristriatus (Schrank), Pterostichus madidus (Fabricius) and Nebria brevicollis (Fabricius) ] related to their control of P. chrysocephala in winter oil seed rape (Warner et al. 2003) Carabid beetles were also found to be spatially and temporally aggregated in field and hedgerow areas in the United Kingdom (Thomas et al. 2001) SAD IE was used as a dispersion indices for development of sampling plans for corn rootworm, Diabrotica spp. adults, which were shown to be aggregated (Park and Tol lefson 2006) The western flower thrips, Frankliniella occidentalis (Pergande), demonstrated aggregated spatial patterns in both the adult and immature stage in greenhouse cucumbers (Park et al. 2009) SADIE analysis was used to describe aggregated distribution of strawberries infected with Strawberry leaf blight in fields in Ohio (Turechek and Madden 1999) In France, peach tre es infected with the aphid transmitted, Plum pox virus Strain M varied in aggregation throughout orchards suggesting transmission of the virus was frequent though not systematic and ecological conditions had a major influence on spread of the virus (Dallot et al. 2003) Bean yellow mosaic vi rus vectored by the lupin aphid, Macrosiphum albifrons (Essig) and the peach aphid, Myzus persica e (Sulzer) was shown to have an aggregated spatial distribution in lupins (Korie et al. 2000) A foliar and glume disease of wheat caused by Stagonospora nodorum (E. M was shown to be aggregated in at least one wheat fie ld using SADIE analysis thus

PAGE 62

62 demonstrating that SADIE successfully evaluates discrete data which contain many zero counts (Shah et al. 2001) Classification and Regression Tree Analysis Ecological data are often complex and unbalanced and may b e strongly nonlinear. Classification and regression tree (CART) analysis are statistical techniques designed to explore and model such data and can deal with non linear, complex and missing data values (Breiman et al. 1984) Classical regression methods rely on assumptions of the distribution and variance of data that are usually invalidated w ith ecological data sets. CART based models are non parametric and can use either categorical or continuous data types, or both. The goal of CART models is that each partition is as homogenous as possible and each split is defined by a simple rule based on a single explanatory variable (De'ath and Fabricius 2000) Landscape elements and other factors influencing insect populations and plant pathogens have been identified using CART. CART has been to used to identify mo rtality factors in larvae of a sawfly, Profenusa thomsoni (Konow) (MacQuarrie et al. 2010) Using CART analysis, outbreaks of the southern pine beetle in the southeastern United States were described to be associated with average climatic conditions (temperature and precipitation) instead of extreme conditions (Duehl et al. 2011) Climatic factors were also used to predict historic outbreaks of the spruce beetle, Dendroctonus rufipennis (Kirby) in Utah and Colorado (Hebertson and Jenkins 2008) CART analysis was used to identify several factors influencing c olony collapse disorder (CCD) of honey bees Apis mellifera ( L. ) (vanEngelsdorp et al. 2010) Alt hough CART analysis is a relatively new technique for insect data analysis it i s ideally suited for complex ecological data. It can handle nonlinear relationships, high order interactions

PAGE 63

63 and missing values an d is not constrained by the restrictions placed on widely used multi regression analysis tools.

PAGE 64

64 Figure 2 1. An example of a variogram used in geostatistical analysis

PAGE 65

65 CHAPTER 3 SPATIAL AND TEMPORAL DISTRIBUTION OF BEMISIA TABACI AND TYLCV IN TOMATO Purpose Biotype B of the sweetpotato whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) also known as the silverleaf whitefly, B. argentifolii ( Bellows and Perring ) is a serious pest of many agricultural crops around the world (Perring et al. 1993) Biotype B has become the key insect pest of tomatoes, Solanum lycopersicum (L.) in south Florida (Schuster et al. 1996a) displacing the native non B biotypes (McKenzie et al. 2004) Biotype B of B. tabaci ca n cause direct damage to tomatoes including an irregular ripening disorder of fruit, inhibition of fruit softening and general reduction of plant vigor (Schuster et al. 1996b, Schuster 2001, McCollum et al. 2004) In Florida, B. tabaci has become a limiting pest species due to its ability to vector plant viruses such as Tomato yellow leaf curl virus (TYLCV) (family Geminiviridae, genus Begomovirus ) (Polston et al. 1999) TYLCV causes one of the most devastating diseases of cultivated tomato world wide Infection by TYLCV can result in loss es of up to 100% in tropical and subtropical regions and can be the limiting factor in commerci al tomato production (Czosnek and Laterrot 1997) TYLCV is transmitted in a persistent, circulative manner by B. tabaci and symptoms of infection in tomato include upward curling of leaflet margins, reduction of leaflet area yellowing of young leaves, stunting of pl ants and abscission of flower s (Polston et al. 1999) With these symptoms there is considerable loss in plant vigor and significant yield loss if infect ion occurs during ea rly growth. Symptoms are expressed in tomato approximately 2 3 weeks post infection (Rom et al. 1993)

PAGE 66

66 Like most insects B. tabaci is aggregated both within individual le aves and within plants at all life stages (Naranjo 1996) B. tabaci populations are aggregated on cotton and tomato plants within fields, although a better understanding of the spatial and temporal dynamics at the field scale and larger is needed (Naranjo and Flint 1995, Naranjo 1996, Polston et al. 1996, Naranjo et al. 2010) Spatial and temporal structure of pest populations can be very important and studi es on spatial patterns could provide a better understanding of pest dynamics and refine sampling plans (Naranjo and Flint 1995, Byrne et al. 1996, Naranjo 1996) With the increased significance of B. tabaci and TY LCV as limiting factors on world wide commercial tomato production there is a need for a better understanding of the ir population dynamics; therefore, spatio temporal dynamics of B. tabaci and TYLCV were evaluated in tomato using Geographical Information Systems (GIS) and Spatial Analyses by Distance IndicEs (SADIE). GIS are tools for studying and mapping the spatial relationship of unknown variables. Interpolation can provide population densities at points not sampled. One interpolation method that is simple to use and can be used with limited sampling points is inverse distance weighted (IDW) (Kravchenko 2003) IDW interpolates values using a linearly weighted combination of a set of sampled points and assumes the variable being mapped decreases in influence with distance f rom its sampled location. IDW has been used to interpolate insect populations including stinkbugs, buprestids, cereal beetles and corn rootworms (Beckler et al. 2005, Bonsignore et al. 2008, Tillman et al. 2009, Re ay Jones 2010) SADIE was developed to quantify spatial patterns of organisms and can be used for insect count data. SADIE measures the degree of clustering in geo referenced data

PAGE 67

67 and unlike most geostatistical methods, SADIE is based on discrete coun t data and its results are conditional to the observed heterogeneity of the data (Xu and Madden 2004) SADIE has been used to describe spatial patterns in insects and plant diseases (Turechek and Madden 1999, Dallot et al. 2003, Park et al. 2006, Reay Jones et al. 2010) M ethods and Materials Study Sites Populations of B. tabaci and TYLCV incidence were monitored on commercial tomato farms for four seasons in central Florida. Farm sizes ranged from 23.6 to 27 3 0 ha and were located in a study area of 53.8 km 2 in Manatee Co., Florida. The farms were selected because they were spatially isolated by distances over 10 km from other commercial tomato production. B. tabaci is capable of traveling 7 km with most migration under 2.7 km (Cohen et al. 1988, Byrne 1999) Farms were managed by commercial growers so pesticide sprays and cultural practices were based on standa rd grower practices. There were cultivar differences between and within farms and all samples were taken on plastic culture d, staked and tied tomatoes. Twice weekly sampling by scouts was initiated as each field was transplanted and included adult white fly counts (total number of adults on 6 contiguous plants) and incidence of TYLCV infection (visual inspection of 50 contiguous plants). Before first tie ( 3 4 weeks after transplant ing ) whitefly counts were taken on whole plant samples and after first tie, counts were taken on the abaxial surface from two leaves at the third node from the top of two branches on each of the six plants using a leaf turn technique (Naranjo et al. 1995, Palumbo et al. 1995) The incidence of TYLCV infected plants was based on the presence of characteristic foliar symptom s and was considered

PAGE 68

68 cumulative throughout the growing season. Scouts were trained to only record tomato plants with unequivocal symptoms of TYLCV infection which included upward curling of leaves, reduction of leaflet area and yellowing of young leaves (Polston et al. 1999) The grower s farm s were divided into blocks and sampling points remained constant throughout the season (i.e. the same 6 plants were used for B. tabaci counts and the same 50 plants were evaluated for incidence o f TYLCV infection ). Along with scouting methodology, t he total number of sample sites w as determined by previous work by Schuster et al (2007b) and the numb er of sample points was determined by block size within farms. The area of tomatoes sampled, the number of sample points, the sample density (ha/sample) and the duration of sampling varied by season ( Table 3 1 ) Sampling varied throughout the season beca use certain fields were planted at different times, pesticide applications kept scouts from entering fields or rain events postponed scouting efforts. At the beginning of each season, geographical positioning system (GPS) coordinates were collected using a GeoExplorer 3000 Series (Trimble ). Data coordinates were converted from decimal degrees to Universal Transverse Mercator (UTM) coordinate system using ArcGIS 9.2 (ESRI 2006) UTM is an adaptation of the Mercator projection and is based on distance in meters. Data Analys e s Spatial and temporal distribution of B. tabaci and TYLCV incidence was interpolated using IDW a statistical method in GIS software ArcView 9.2 (ESRI 2006) IDW is based on the assumption that points close by are more closely related than points fa rther apart and estimated predictions are based on valu es at sampled locations Values are determined using a linear weighted combination of observed values and those weights are functions of the distance between locations. The most

PAGE 69

69 commonly used weighed power of two was used. Unlike other interpolation met hods IDW does not require a variogram model and is appropriate for small data sets (Kravchenko 2003) Cross validation was used to estimate the fit of the IDW model. Cross validation removes one sample point at a time and compares observed and predicted values for that point (Isaaks and Srivastava 1989) The root mean square prediction error (RMSE) produced by cross validation is presented as the summary statistic to che ck the accuracy of the model. IDW maps were created using seasonal means of B. tabaci adults and totals of final TYLCV incidence. Underlying d igital images in the form of digital orthographic photos were downloaded from the Land Boundary Information Syst em (LABINS 2011) SADIE (version 1.22) analys e s of adult B tabaci w ere conducted on year end means over the entire season (Perry et al. 1999) TYLCV incidence expr essed as a percentage of 50 plants, was subjected to SADIE analysis using the final cumulative seasonal virus incidence per sampling point. S patial scales were evaluated at the level of the entire sampling area and within farms. Additionally, weekly adul t B. tabaci and TYLCV incidenc e was subjected to SADIE analyse s over the entire tomato growing season to analyze the spatio temporal distribution. The purpose of SADIE is to quantify the spatial pattern by calculating the distance to regularity ( D ) defi ned as the minimum distance to which individuals in a sample would have to move to result in a uniform distribution. The overall aggregation index I a is defined as D/E a where E a is the arithmetic mean distance to regularity for the randomized samples. V alues of I a > 1 suggest aggregation, I a < 1 suggest a regular pattern and I a = 1 indicates a random pattern (Perry 1998) The probability ( P ) is

PAGE 70

70 derived after randomizations as a formal test of randomness: the null hypothesis of spatial randomness is rejected at = 0.1 ( P < 0.05, aggregation or P > 0.95, uniformity). Similar levels of significance are presented in other spatio temporal insec t studies (Kim et al. 2007, Park et al. 2009, Reay Jones et al. 2010) SADIE also provides overall clustering indices j i indicating positive patch cluster to quantify the degree to which the count for each sample unit contributes to the overall degree of clustering. Random spatial pattern has clustering indices around 1. In the convenience of time and effort a total of 3900 randomizations were used for each test. Resul ts Fall 2007 In the fall of 2007, 5351 data points were recorded for B. tabaci adults over the entire study area and season from 23 1 geo referenced sample sites (Table 3 1) There w as a maximum number of whiteflies of 121 per six plants, per sample, per d ay and a mean of 1.354 over all sample sites and sampling dates (Table 3 2 ) Farms had varying levels of adult whiteflies ranging from a daily maximum of 10 to 121 per sample site per day and seasonal means from 0.842 to 2.861 over entire farms and sampli ng dates (Table 3 2 ). The 232 geo referenced sample sites were used to record the final virus per farm and the progression of virus over weeks. Final virus incidence maximums per farm ranged from 30% to 100%. Mean virus incidence averaged from final vir us incidence levels per site ranged from 7.58% to 68.21% from individual farms (Table 3 2 ).

PAGE 71

71 Spring 2008 In the spring of 2008, 7515 data points were recorded for B. tabaci adults from 334 sample sites (Table 3 1) The overall season high adult B. tabaci c ount was 45 per sample site, per day and the overall seasonal mean was 0.377 over all sampling sites (Table 3 3 ) Individual farms varied in whitefly pressure and had daily maximums of 8 to 45 and farm means from 0.076 to 0.569 over all sample dates (Tabl e 3 3 ). In the spring of 2008, 334 sites were used for virus incidence measurement. Virus incidence was much lower than the previous season and farms had final virus incidence per sample site rang ing from 2% to 14% incidence. Final mean virus incidence per farm varied from 0.17% to 3.49% (Table 3 3 ). Fall 2008 In the fall of 2008, 4103 data points were recorded for B. tabaci adults from 226 sample sites (Table 3 1) B. tabaci counts were similar to spring of 2008 with an overall season high of 45 whitef lies per sample site and a seasonal mean of 0.47 (Table 3 4 ) Adult whitefly maximum counts on individual farms ranged from 3 to 45 and means from 0.218 to 1.591 (Table 3 4) The maximum virus incidence was 32% at one sample site over the entire sampling area. Levels of final overa l l virus incidence ranged from 0% to 32% and final mean virus incidence per farms varied from 0% to 6.55% (Table 3 4 ). Spring 2009 During the spring of 2009, 4121 data points were recorded for B. tabaci adults from 372 geo ref erenced samples (Table 3 1) During this season the lowest B. tabaci counts were taken over the entire study (Table 3 5). T he maximum number of whiteflies over all sample sites, over the entire season was 4 and the overall mean was 0.1 (Table 3 5 ) B. t abaci adult maximum counts per farm ranged from 2 to 4 and means from

PAGE 72

72 0.081 to 0.149 (Table 3 5 ). Like B. tabaci counts TYLCV incidence was much lower with a maximum virus incidence per sample site over the entire study area of 2%. Final mean virus inc idence per farm varied from 0% to 0.211% (Table 3 5 ). IDW Interpolation For interpolation methods such as IDW, mean error and root mean square error (R MS E) from cross validation tests can be used to evaluate how precise the method is producing interpolat ion maps. The value of the root mean square error depends on the scale of the data so overall adult whitefly R MS E were lower than TYLCV incidence R MS E (Table 3 6 ). Smaller R MS E indicate a better fit of the model to the observed data. Overall, mean erro rs were low, indicating good estimates of B. tabaci populations and TYLCV incidence (Table 3 6 ) (Figures 3 1 to 3 11 ). Similar RMSE were presented for other IDW interpolation maps of insect counts (Tillman et al. 2009, Reay Jones et al. 2010) SADIE Analys is Significant aggregation (positive I a values) over all sample sites from seasonal means of B. tabaci and seasonal incidence of TYLCV was found in fall 2007, spring 2008, and fall 2008 (Table 3 2 3 3 and 3 4 ). Within those three seasons, populations al so showed a strong presence of gap (negative j ) and patchiness (positive i ) Individual farms had varying levels of overall aggregation and in fall 2007 u sing seasonal means B. tabaci was significantly aggregated in 16.7% of the farms and TYLCV incidence was aggregated in 50% of the farms (Tab le 3 2 ) There was also significant cluste ring into gaps in the fall 2007 In the spring of 2008, only Farm J had significantly aggregated populations of B. tabaci and incidence of TYLCV (Table 3 3 ). In the fall of 2008, 40% of the farms had seasonal me an aggregation of B. tabaci and

PAGE 73

73 none showed significant aggregation of TYLCV incidence Significant patch an d gap clusters varied (Table 3 4 ) In the spring of 2009, there were no significantly aggregated populations of B. tabaci or incidence of TYLCV (T able 3 5 ). Weekly adult B. tabaci means and TYLCV incidence at the last virus incidence per week were subjected to SADIE analysis. In the interest of space only aggregation indices and their associated probabilities are presented. Weekly populations var ied in distribution throughout each season. In the fall of 2007, B. tabaci populations were significantly aggregated in 14.7% and significantly uniform in 2. 1% of farms per weeks sampled (Table 3 7 ). TYLCV incidence was significantly aggregated in 30% an d significantly uniform in 2.5% of farms per weeks sampled (Table 3 8 ). In the spring of 2008, B. tabaci were significantly aggregated in 37.0% of farms per weeks sampled (Table 3 9 ). TYLCV was significantly aggregated in 36.4 % and significantly uniform in 4.6 % of farms per weeks sampled (Table 3 10 ). In the fall of 2008, B. tabaci were significantly aggregated 23. 1 % and significantly uniform in 1.9% of farms per weeks sampled (Table 3 1 1 ). TYLCV was significantly aggregated in 14.8% and significantly u niform in 3.7% of farms per weeks sampled (Table 3 1 2 ). In the spring of 2009, B. tabaci were significantly aggregated in 7.3% and significantly uniform in 7. 4 % of farms per weeks sampled (Table 3 1 3 ). TYLCV was not significantly aggregated or uniform on any farm at any week sampled (Table 3 1 4) Discussion The sampling of adult B. tabaci and TYLCV across commercial Florida tomato farms combined with mapping o f their distribution and analyse s by SADIE has produced a much more detailed explanation of dis tribution than previously reported SADIE confirmed the dynamic distribution of B. tabaci populations and showed B. tabaci

PAGE 74

74 tended to be aggregated within the study area and within individual farms. Results showed that TYLCV distribution tend ed to follow spatio temporal patterns associated with its vector. Strong spatial dependence in insect data indicates that estimating populations at unsampled locations is possible (Liebhold et al. 1993) With the use of IDW, a map a llow s for visual expression of pest populations, which is valuable to re searchers an d farm managers. From IDW maps created by seasonal means visual expression of seasonal populations of whiteflies and TYLCV can be evaluated. In some farms, including Farms A, C, and D from fall 2007, seasonal mean IDW maps indicated little to no evidenc e of adult whiteflies distributing uniformly throughout the farms Rather, p opulations appeared to be aggregated along field edges (Figures 3 1 and 3 2). This edge effect was also apparent in other farms regardless of the scale of whitefly density as was observed in Farms E and F from fall 2007; Farms I and J from spring 2008; and Farms C, L, E, and F from fall 2008 (Figures 3 3, 3 5, 3 6, 3 7, 3 8, and 3 9). Even in years with very low B. tabaci pressure such as in spring 2009 IDW maps faintly indicat ed edge effects as in Farms H and J (Figures 3 10 and 3 11). Some farms suc h as Farms B and G in fall 2007 and Farm B in spring 2009, indicated populations of whiteflies scattered across the farm with no indication of edge effects (Figure 3 1, 3 4 and 3 10 ). Other observations suggested that some adult B. tabaci populations were located on the northwest and southeast corners of farms, as indicated by the following IDW maps : Farms A, D, and E fall 2007; Farm J Spring 2008; and Farm L Fall 2008 (Figures 3 1, 3 2, 3 3, 3 6, and 3 8). S easonal means do not take into consideration daily fluctuations of B. tabaci within the farm ecosystem which could be influenced by surrounding crops or habitats or by events including weather (wind,

PAGE 75

75 rain, etc.), pesticide ap plications or grower cultural manipulations (pruning, staking, tying, etc.) However, maps of seasonal means of B. tabaci describe the overall trends among and across sampled farms. SADIE analysis provide d tools to evaluate population distributions within the study area at multiple spatial scales. Over the entire study area distribution s of whiteflies were significantly aggregated in every season except spring 2009. This corroborates previous research about the aggregated distribution of B. tabaci in to mato and other crops (Naranjo and Flint 1995, Polston et al. 1996) Significant I a and clustering into gaps and patches indicated that populations were highly spatially aggregated and there were large areas with f ew o r no whiteflies (Tables 3 2 3 3 and 3 4 ). Xu and Madden (2004) suggested that the I a index was more influenced by the number and position of clusters rather than the cluster size. A limitation of SADIE is that the geographic position of aggregated clusters is not taken into account. However, w here the aggregation occurs in geographic space (e.g. edge of tomato field) plays a role in determining causes for abundance of B. tabaci and plant expressing symptoms of TYLCV. Recent work (Taylor, unpublished) points to populations of B. tabaci and plan ts expressing symptoms of TYLCV influenced by unknown reservoir sources surrounding tomato fields with populations of both B. tabaci and TYLCV being influenced greatly by distance to field edge. Weekly fluctuations in aggregation indicate either a highly mobile pest species or one that has many dynamic variables influencing distribution. Alt hough aggregation varied throughout each season within farms some interesting conclusions can be draw n from individual farms. In F arms A, D, and G from fall 2007; Farms H and J from

PAGE 76

76 spring 2008 ; Farms C, L, and E from fall 2008 ; and Farms I and J from spring 2009 ; significant aggregation of B. tabaci adults was shown in some of the earliest sampling dates but showed more random trends in later weeks (Tables 3 7 3 9 3 1 1 and 3 1 3 ). Some of those farms that had earlier significant B. tabaci aggreg ated populations had brief period s of significant re aggregation 6 to 10 weeks later. In contrast to this observation the corn leaf aphid, Rhopalosiphum maidis (Fitch), a dynamic pest of field corn, was aggregated during the peak populations in the middle of the crop and randomly distributed early and late in the crop (Park and Obrycki 2004) Though it is unclear biologically what was influencing the periods of re aggregation i n this study, it is possible that re introduction of adult B. tabaci in farm reproduction, or management tactics were influencing population distribution. Pesticides which are a central ta ctic for controlling B. tabaci and the TYLCV it vectors, have bee n shown to alter the dispersion patterns of whiteflies (Liu et al. 1993b, Tonhasca et al. 1994) As expected, the distribution of plants with symptoms of TYLCV was more static than the distribution of B. tabaci TYLCV symptom expression lags behind B. tabaci populations by 2 3 weeks; therefore, significant aggregation indices of TYLCV were delayed. In the fall of 2007, every farm except Farm F had at least one week of significant aggregation of TYLCV incidence ( Tab le 3 8 ). Farms B, C, and E had significant aggregation indices through almost the entire season, suggesting re introduction of whiteflies in the same areas or secondary spread after the initial introduction (Table 3 8 ). Similar results were suggested from Farm J in spring 2008 and Farm F in fall 2008 (Tables 3 10 and 3 1 2 ). Other farms such as Farms A and D from fall 2007 and Farm H from spring 2008 had brief significantly aggregated distributions in

PAGE 77

77 the early to middle of the season before returning to a random distribution (Tables 3 8 and 3 10 ). Farm G in fall 2007, Farm I in s pring 2008 and Farm C in f all 2008 had significant weekly uniform distributions of TYLCV and tended towards uniformity over the entire season (Tables 3 8 3 10, and 3 12 ). T raditional indices for determining non randomness, such as variance mean ratios do not include the spatial patterns of sample points. SADIE analysis which uses spatial patterns was developed for highly dynamic populations and takes into account extreme ly patchy counts, many counts of zeros and dynamic populations over space and time. SADIE analysis demonstrated aggregated populations of B. tabaci in tomato which confirmed earlier work with other indices (Polston et al. 1996) Spatial patterns of B. tabaci and TYLCV demonstrated different patterns within some farms while patterns were very similar in other farms. This could be explained by the movement of B. tabaci populations that were able to transmit TYLCV. Analysis with IDW indicated edge effects and aggregated populations in certain areas of many farms. Future work could encompass more area or increase the level of sampling. Future work could also include geostatistical analysis of this data set and use of semivariograms and kriging rather than deterministic methods such as IDW. Semivariograms express the variance of sample pairs against the distance between sample points and provide important ecological information on the spatial patterns of organisms. These spatial patterns could be used to indicate the distance between sampling locations to derive sampling plans which require independent samples. Unlike IDW, k riging prediction maps use geostatistical methods and are based on statistical models that include autocorrelation or the correlation of a variable with itself through

PAGE 78

78 space Kriging can provide a certainty and accuracy of the predictions even though it is more complicated to produce and requires data exploration before a map can be produced Highly dynamic populations such as those seen in insect count data increase the level of uncertainty in geostatistical analysis and much work will have to be conducted to create validated results. One criticism of the utility of geostatistics in applied agriculture is the high level of understanding of mathematical concepts. B. tabaci is a mobile pest and has been shown in the present study to have varying aggregation in both space and time. The dynamic distribution patterns are more likely driven by h osts outside tomato be cause IDW maps and SADIE analyse s suggest that most populations arose from outside of the sampled farms Cultivated crop hosts of B. tabaci grown relatively close to tomato and old tomato fields have been indicated as hosts of B. tab aci and TYLCV (Polston and Lapidot 2007) Weeds can be important hosts for B. tabaci and TYLCV and in a heterogeneous landscape weeds could be heavily influencing populations in tomato. B. tabaci has a host range of over 600 plant species (Mound and Halsey 1978, Greathead 1986, Secker et al. 1998) Future work will need to be conducted to evaluate which weed hosts are the most important reservoirs of both B. tabaci and TYLCV.

PAGE 79

79 Table 3 1. Total sc outing area and sample sites, 2007 2009 Season Total ha Scouted Sample Sites Mean ha per sample Samples Taken Mean Number of Samples per site Scouting Initiated Scouting Ended Fall 2007 318.9 231 1.4 5351 23.2 8/17/07 12/11/07 Spring 2008 489.2 334 1.5 7 515 22.5 1/15/08 5/1/08 Fall 2008 268.0 226 1.2 4103 18.2 8/28/08 12/4/08 Spring 2009 641.7 372 1.7 4121 11.1 2/19/09 5/22/09

PAGE 80

80 Table 3 2 Summary data for distribution of B. tabaci adults and TYLCV incidence including SADIE analysis, Fall 2007 Farm Va riable a Mean Min Max Std. Dev. N I a b P c j P j c i P i c Overall AW 1.354 0 121 4.459 5351 2.87* < 0.001 3.108* < 0.001 3.03* < 0.001 TYLCV 29.672 0 100 26.650 232 2.59* < 0.001 3.132* < 0.001 2.18* 0.045 A AW 1.051 0 45 3.235 877 2.27* 0.004 2.270* 0.002 1.38 0.110 TYLCV 38.82 1 4 82 22.774 39 1.148 0.251 1.134 0.261 0.117 0.272 B AW 0.468 0 16 1.430 1278 0.865 0.791 0.834 0.847 0.864 0.779 TYLCV 20.536 0 96 21.947 56 1.718* 0.002 1.807* 0.001 1.229 0.116 C AW 0.842 0 10 1.758 317 0.821 0.763 0.9 0.618 0.948 0.537 TYLC V 11.067 0 30 7.554 15 1.164 0.177 1.075 0.290 1.253 0.107 D AW 1.628 0 116 6.469 749 1.311 0.069 1.392* 0.041 1.125 0.192 TYLCV 30.063 4 66 18.854 32 1.486* 0.024 1.882* 0.002 1.368 0.053 E AW 2.861 0 60 6.001 792 1.124 0.238 1.072 0.327 0.961 0.5 11 TYLCV 68.214 10 100 24.162 28 1.486* 0.024 1.882* 0.002 1.368 0.053 F AW 1.928 0 121 6.188 751 0.771 0.446 0.334 0.527 0.69 0.365 TYLCV 32.138 2 96 24.802 29 0.458 0.692 0.399 0.449 1.045 0.198 G AW 0.894 0 16 1.701 587 0.915 0.572 1 0.419 0.8 17 0.820 TYLCV 7.576 0 36 7.981 33 0.936 0.531 0.928 0.536 0.963 0.465 Note: a Variable AW equals adult B. tabaci and TYLCV equals cumulative TYLCV incidence. b Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern (< 1). c P value for null hypothesis of spatial randomness. values indicate significant indices for = 0.1 (P < 0.05).

PAGE 81

81 Table 3 3 Summary data for distribution of B. tabaci adults and TYLCV incidence including SADIE analysis, Spring 2008 Farm Variable a Mean Min Max Std. Dev. N I a b P c j P j c i P i c Overall AW 0.377 0 45 1.732 7515 6.276* < 0.001 8.244* < 0.001 7.567* < 0.001 TYLCV 0.743 0 14 1.967 334 3.52* < 0.001 4.312* < 0.001 2.413* 0.005 H AW 0.076 0 14 0.405 43 15 NA NA NA NA NA NA TYLCV 0.172 0 6 0.764 151 0.977 0.443 0.993 0.431 1.032 0.363 I AW 0.569 0 30 1.920 2027 1.094 0.261 1.075 0.288 1.156 0.181 TYLCV 0.207 0 6 0.763 116 0.974 0.465 0.949 0.522 1.078 0.276 J AW 1.225 0 45 3.516 1051 2.69* < 0.0 01 3.118* < 0.001 1.64* 0.046 TYLCV 3.491 0 14 3.378 55 3.447* < 0.001 4.097* < 0.001 2.723* < 0.001 K AW 0.770 0 8 1.589 122 1.409 0.076 1.611* 0.043 1.410 0.102 TYLCV 0.833 0 2 1.030 12 0.852 0.679 0.835 0.652 0.770 0.817 Note: a Variable AW equals adult B. tabaci and TYLCV equals cumulative TYLCV incidence. b Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern (< 1). c P value for null hypothesis of spatial randomness. values indicate s ignificant indices for = 0.1 (P < 0.05).

PAGE 82

82 Table 3 4 Summary data for distribution of B. tabaci adults and TYLCV incidence including SADIE analysis, Fall 2008 Farm Variable a Mean Min Max Std. Dev. N I a b P c j P j c i P i c Overall AW 0.470 0 45 2.105 4103 2.977* < 0.001 3.595* < 0.001 3.218* < 0.001 TYLCV 2.352 0 32 4.414 226 1.92* 0.005 1.756* 0.037 1.305 0.151 A AW 0.293 0 3 0.219 307 NA NA NA NA NA NA TYLCV 0.000 0 0 0 22 NA NA NA NA NA NA C AW 0.267 0 6 0.670 719 0.995 0.434 0.99 0.458 0.982 0.500 TYL CV 2.100 0 14 3.536 40 0.864 0.800 0.931 0.583 0.817 0.864 E AW 1.591 0 45 4.247 492 1.518* 0.015 1.476* 0.021 1.307 0.076 TYLCV 6.552 0 32 8.798 29 1.329 0.057 1.287 0.070 1.594* 0.011 F AW 0.218 0 15 0.752 898 1.512* 0.045 1.4 0.080 1.605* 0.046 TYLCV 1.509 0 10 2.599 53 1.380 0.091 1.268 0.143 1.467 0.069 L AW 0.225 0 4 0.564 1687 1.564 0.061 1.646* 0.047 1.646 0.449 TYLCV 1.927 0 14 2.905 82 0.777 0.722 0.794 0.719 0.935 0.468 Note: a Variable AW equals adult B. tabaci and TYLCV equal s cumulative TYLCV incidence. b Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern (< 1). c P value for null hypothesis of spatial randomness. values indicate significant indices for = 0.1 (P < 0.05).

PAGE 83

83 Table 3 5 Summary data for distribution of B. tabaci adults TYLCV incidence including SADIE analysis, Spring 2009 Farm Variable a Mean Min Max Std. Dev. N I a b P c j P j c i P i c Overall AW 0.100 0 4 0.370 4121 0.758 0.679 0.778 0.647 0.820 0.591 TYLCV 0.070 0 2 0.368 372 1.488 0.107 1.600 0.074 1.802* 0.047 B AW 0.105 0 4 0.367 840 NA NA NA NA NA NA TYLCV 0.070 0 2 0.371 57 1.107 0.256 1.100 0.264 1.137 0.209 H AW 0.101 0 3 0.363 2073 0.976 0.441 0.964 0.460 1.039 0.376 TYLCV 0.034 0 2 0.260 176 1.048 0.335 1.014 0.386 1.078 0.304 I AW 0.081 0 4 0.336 1047 1.074 0.284 1.080 0.278 1.167 0.175 TYLCV 0.000 0 0 0 120 NA NA NA NA NA NA J AW 0.149 0 4 0.550 161 1.306 0.108 1.353 0.079 0.993 0.948 TYLCV 0.211 0 2 0.631 19 1.063 0.293 1.031 0.346 1.164 0.174 Note: a Variable AW equals adult B. tabaci and TYLCV equals cumulative TYLCV incidence. b Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern (< 1). c P value for null hypothesis of spatial randomness. values indicate significant indices for = 0.1 (P < 0.05).

PAGE 84

84 Table 3 6 Cross validation results of IDW interpolation analysis for B. tabaci means and final TYLCV incidence, 2007 2009 Season Farm Adult B. tabaci Mean Error Adult B. tabaci RSME a TYLCV Incidence Mean Error TYLCV Incidence RMSE a Fall 2007 A 0.013 0.559 0.249 22.670 B 0.008 0.265 0.054 22.090 C 0.029 0.361 0.717 8.496 D 0.092 0.555 1.210 15.310 E 0.227 2.812 1.205 19.510 F 0.061 1.704 0.566 25.680 G 0.069 0.564 0.140 8.940 Spring 2008 H 0.001 0.079 0.007 0. 801 I 0.020 0.549 0.031 0.776 J 0.064 1.015 0.117 2.904 K 0.050 0.408 0.016 1.390 Fall 2008 A 0.000 0.043 NA NA C 0.016 0.235 0.297 3.938 E 0.051 0.820 0.622 8.271 F 0.012 0.242 0.101 3.203 L 0.015 0.165 0.096 3.320 Spring 2009 B 0.012 0.112 0.011 0.387 H 0.008 0.138 0.000 0.295 I 0.004 0.139 NA NA J 0.005 0.206 0.019 0.697 Note: Adult B. tabaci counts were summarized by averages of whiteflies taken over all dates and sampling points from farms. TYLCV Incidence was t aken from the last scouting date of the season and presented as the cumulative virus incidence over the entire season per farm. a Smaller root mean square error (RMSE) indicates a better fit of the model.

PAGE 85

85 Table 3 7 Weekly SADIE aggregation indices of adult B. tabaci counts, Fall 2007 Date Farm A Farm B Farm C Farm D Farm E Farm F Farm G I a a P b I a a P b I a a P b I a a P b I a a P b I a a P b I a a P b 8/13 0.941 0.552 NA NA NA NA NA NA 1.070 0.308 0.771 0.446 NA NA 8/20 2.487* 0.001 0.837 0.772 1.011 0.401 1. 409* 0.037 1.261 0.110 1.072 0.324 NA NA 8/27 2.88* <0.001 1.088 0.258 0.895 0.676 1.379* 0.041 1.073 0.301 0.771 0.446 NA NA 9/3 1.076 0.313 1.062 0.299 1.151 0.179 0.768* 0.956 1.243 0.123 1.210 0.306 NA NA 9/10 2.392* 0.001 1.635* 0.005 1.126 0.243 1 .365 0.059 1.516* 0.019 0.864 0.253 1.857* 0.003 9/17 0.873 0.607 1.117 0.214 0.774 0.933 1.126 0.240 1.228 0.120 0.299 0.865 1.862* 0.003 9/24 1.380 0.107 1.219 0.109 0.936 0.586 0.815 0.892 1.113 0.235 0.502 0.804 1.262 0.132 10/1 1.121 0.274 0.789 0. 938 0.947 0.577 1.023 0.377 1.076 0.314 0.353 0.570 1.497* 0.040 10/8 0.914 0.545 1.336* 0.048 1.045 0.295 0.825 0.849 1.169 0.200 0.453 0.687 0.861 0.682 10/15 1.135 0.257 0.952 0.549 0.947 0.577 1.180 0.142 1.082 0.309 0.971 0.146 1.364 0.084 10/22 2. 604* <0.001 0.917 0.637 0.895 0.702 1.290 0.063 1.439* 0.027 1.464 0.167 0.801 0.820 10/29 1.263 0.158 0.977 0.484 0.985 0.439 1.023 0.367 1.132 0.232 0.548 0.758 1.215 0.146 11/5 1.494 0.106 1.035 0.350 NA NA 0.771* 0.954 1.357 0.062 1.012 0.328 1.204 0 .163 11/12 1.518 0.095 1.107 0.261 NA NA 0.824 0.853 1.187 0.163 0.795 0.235 1.129 0.250 11/19 NA NA NA NA NA NA 1.108 0.259 0.976 0.484 0.631 0.809 1.452* 0.049 11/26 NA NA NA NA NA NA NA NA 0.892 0.668 0.880 0.492 NA NA Note: a Overall index of disp ersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern (< 1). b P value for null hypothesis of spatial randomness. values indicate significant aggregation indices ( P < 0.05) or significant uniform indices at ( P > 0.95).

PAGE 86

86 Ta ble 3 8 Weekly SADIE aggregation indices of TYLCV incidence, Fall 2007 Date Farm A Farm B Farm C Farm D Farm E Farm F Farm G I a a P b I a a P b I a a P b I a a P b I a a P b I a a P b I a a P b 8/13 NA NA NA NA NA NA NA NA NA NA NA NA NA NA 8/20 NA NA NA NA NA NA NA NA NA NA NA NA NA NA 8/27 NA NA NA NA NA NA 1.218 0.114 0.984 0.470 NA NA NA NA 9/3 0.865 0.643 1.064 0.433 NA NA 1.304 0.068 0.866 0.744 NA NA NA NA 9/10 0.871 0.611 0.857 0.779 1.004 0.466 1.389* 0.039 1.098 0.260 0.518 0.282 NA NA 9/17 0.854 0.625 0 .896 0.689 0.814* 0.950 1.320 0.063 1.418* 0.040 0.646 0.762 NA NA 9/24 0.864 0.620 1.035 0.353 0.936 0.586 1.175 0.163 1.659* 0.007 1.223 0.373 1.263 0.132 10/1 0.798 0.728 1.487* 0.016 1.511* 0.012 1.175 0.163 1.676* 0.004 0.994 0.467 0.664* 0.986 10/ 8 1.337 0.128 1.499* 0.016 1.646* 0.005 1.186 0.152 1.559* 0.014 1.049 0.349 1.296 0.102 10/15 1.454 0.086 1.527* 0.014 1.553* 0.009 1.235 0.104 1.455* 0.028 1.161 0.228 0.876 0.660 10/22 1.657* 0.040 1.254 0.092 1.602* 0.008 1.308 0.061 1.58* 0.012 0.84 0 0.489 0.914 0.574 10/29 1.157 0.240 1.458* 0.017 1.403* 0.036 1.240 0.101 1.488* 0.027 0.648 0.522 0.876 0.676 11/5 1.005 0.404 1.443* 0.022 1.307 0.070 1.285 0.070 1.574* 0.016 0.314 0.772 1.012 0.390 11/12 1.148 0.251 1.779* 0.001 1.164 0.177 1.271 0.086 1.515* 0.023 0.586 0.534 1.022 0.367 11/19 NA NA 1.718* 0.002 NA NA 1.284 0.083 1.486* 0.024 0.458 0.692 0.936 0.531 11/26 NA NA NA NA NA NA NA NA NA NA 0.640 0.566 NA NA Note: a Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern (< 1). b P value for null hypothesis of spatial randomness. values indicate significant aggregation indices ( P < 0.05) or significant uniform indices at ( P > 0.95).

PAGE 87

87 Table 3 9 Weekly SADIE aggregation indices of adult B. tabaci counts, Spring 2008 Date Farm H Farm I Farm J Farm K I a a P b I a a P b I a a P b I a a P b 1/21 0.749 0.925 NA NA NA NA NA NA 1/28 1.627* 0.027 NA NA NA NA NA NA 2/4 1.883* 0.010 NA NA NA NA NA NA 2/11 1.532* 0.045 NA NA NA NA NA NA 2/18 1.883* 0.010 1.120 0.233 2.243* < 0.001 NA NA 2/25 0.855 0.702 1.036 0.353 2.642* < 0.001 NA NA 3/3 0.962 0.458 0.840 0.786 1.858* 0.014 0.987 0.506 3/10 0.931 0.521 NA NA 1.807* 0.012 NA NA 3/17 0.837 0.711 0.943 0.524 0.951 0.502 1.093 0.272 3/24 1. 531* 0.044 1.444* 0.050 0.782 0.797 NA NA 3/31 0.967 0.454 1.299 0.102 1.73* 0.026 1.66* 0.017 4/7 1.283 0.114 0.851 0.747 1.375 0.094 0.762 0.952 4/14 1.006 0.405 2.044* 0.001 1.049 0.345 1.359 0.081 4/21 1.309 0.104 1.769* 0.004 1.093 0.288 1.025 0.3 73 4/28 2.48* < 0.001 1.017 0.385 2.174* 0.002 1.130 0.220 5/5 1.019 0.376 1.448* 0.033 2.317* 0.001 NA NA 5/12 1.196 0.178 1.482* 0.030 1.946* 0.010 1.470 0.065 5/19 NA NA 1.284 0.090 1.043 0.354 0.992 0.434 5/26 NA NA 1.319 0.067 NA NA NA NA Note: a Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern ( < 1). b P value for null hypothesis of spatial randomness. values indicate significant aggregation indices ( P < 0.05) or significant uniform i ndices at ( P > 0.95).

PAGE 88

88 Table 3 10 Weekly SADIE aggregation indices of TYLCV incidence, Spring 2008 Date Farm H Farm I Farm J Farm K I a a P b I a a P b I a a P b I a a P b 1/21 NA NA NA NA NA NA NA NA 1/28 NA NA NA NA NA NA NA NA 2/4 NA NA NA NA NA NA NA N A 2/11 NA NA NA NA NA NA NA NA 2/18 0.994 0.418 NA NA 1.783* 0.019 NA NA 2/25 1.136 0.232 NA NA 2.202* 0.002 NA NA 3/3 1.136 0.232 NA NA 2.067* 0.006 NA NA 3/10 1.136 0.232 NA NA 1.672* 0.032 NA NA 3/17 1.403 0.075 0.728* > 0.999 1.407 0.052 NA NA 3 /24 1.7* 0.017 0.728* > 0.999 2.341* 0.001 NA NA 3/31 1.986* 0.004 NA NA 3.014* < 0.001 NA NA 4/7 1.658* 0.022 0.953 0.517 2.418* 0.001 0.835 0.921 4/14 1.091 0.272 1.312 0.083 3.309* < 0.001 0.835 0.921 4/21 1.288 0.118 0.985 0.449 2.385* 0.001 0.835 0.921 4/28 1.327 0.096 0.854 0.738 3.393* < 0.001 0.835 0.921 5/5 0.903 0.584 0.856 0.757 3.687* < 0.001 0.835 0.921 5/12 0.977 0.443 0.974 0.473 3.206* < 0.001 1.155 0.230 5/19 NA NA 1.212 0.148 3.447* < 0.001 0.852 0.679 5/26 NA NA 0.974 0.465 NA NA NA NA Note: a Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern ( < 1). b P value for null hypothesis of spatial randomness. values i ndicate significant aggregation indices (P < 0.05) or signific ant uniform indices at (P > 0.95).

PAGE 89

89 Table 3 1 1 Weekly SADIE aggregation indices of adult B. tabaci counts, Fall 2008 Date Farm A Farm C Farm E Farm F Farm L I a a P b I a a P b I a a P b I a a P b I a a P b 8/25 NA NA 1.096 0.233 NA NA NA NA 1.297 0.066 9/1 NA NA 1.539* 0.006 1.279 0.093 1.033 0.339 1.664* 0.003 9/8 NA NA 1.248 0.088 NA NA 0.902 0.581 2.974* < 0.001 9/15 1.8* 0.031 1.118 0.200 1.748* 0.001 1.176 0.216 1.015 0.411 9/22 0.955 0.665 1.154 0.153 1.035 0.352 0.964 0.450 0.710 0.836 9/29 NA NA 0.977 0.484 0.928 0.584 1.105 0.266 0.928 0.563 10/6 NA NA 1.151 0.163 NA NA 0.831 0.710 1.006 0.414 10/13 NA NA 1.132 0.302 1.331 0.054 1.345 0.098 1.211 0.224 10/20 NA NA 1.040 0.334 1.094 0.258 0.850 0.663 1.311 0.160 10/27 NA NA 1.235 0.086 1.405* 0.024 0.608* 0.999 0.886 0.546 11/3 NA NA 0.992 0.449 1.356* 0.049 1.139 0.246 0.832 0.648 11/10 0.719 0.674 1.187 0.121 1.200 0.136 NA NA 0.893 0.533 11/17 NA NA 1.080 0.264 1.567* 0.010 0.949 0.557 NA NA 11/24 NA NA NA NA 1.433* 0.015 0.812 0.757 2.2 22* 0.005 Note: a Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern (< 1). b P value for null hypothesis of spatial randomness. values indicate significant aggregation indices ( P < 0.05) or signif icant uniform indices at ( P > 0.95).

PAGE 90

90 Table 3 1 2 Weekly SADIE aggregation indices of TYLCV incidence, Fall 2008 Date Farm A Farm C Farm E Farm F Farm L I a a P b I a a P b I a a P b I a a P b I a a P b 8/25 NA NA 1.043 0.451 NA NA NA NA NA NA 9/1 NA NA 0.827 0.885 NA NA NA NA 1.003 0.390 9/8 NA NA 0.958 0.535 NA NA NA NA 0.931 0.429 9/15 NA NA 0.790 0.949 1.182 0.139 NA NA 1.453 0.094 9/22 NA NA 0.993 0.443 1.403* 0.038 0.65* 0.987 1.119 0.291 9/29 NA NA 0.841 0.897 1.4* 0.038 0.723 0.919 1.086 0.328 10/6 NA NA 1.039 0.356 1.4* 0.038 0.711 0.935 1.043 0.374 10/13 NA NA 0.76* 0.969 1.105 0.249 0.884 0.591 1.046 0.360 10/20 NA NA 0.831 0.859 1.108 0.279 1.166 0.213 0.918 0.516 10/27 NA NA 0.814 0.897 1.244 0.101 1.137 0.239 0.877 0.579 11/3 NA NA 0.846 0 .836 1.330 0.058 1.597* 0.041 0.695 0.850 11/10 NA NA 0.814 0.899 1.253 0.090 1.597* 0.041 0.647 0.916 11/17 NA NA 0.851 0.823 1.253 0.090 1.706* 0.017 0.647 0.916 11/24 NA NA 0.851 0.823 1.315 0.063 1.826* 0.012 0.773 0.727 12/1 NA NA 0.864 0.800 1.29 2 0.067 1.71* 0.024 0.777 0.722 12/8 NA NA NA NA 1.329 0.057 1.380 0.091 NA NA Note: a Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern (< 1). b P value for null hypothesis of spatial randomness. values indicate significant aggregation indices (P < 0.05) or significant uniform indices at (P > 0.95).

PAGE 91

91 Table 3 1 3 Weekly SADIE aggregation indices of adult B. tabaci counts, Spring 2009 Date Farm B Farm H Farm I Farm J I a a P b I a a P b I a a P b I a a P b 2/16 1.095 0.287 NA NA NA NA NA NA 2/23 1.000 0.412 NA NA NA NA NA NA 3/2 1.269 0.111 1.077 0.366 NA NA NA NA 3/9 0.819 0.818 0.879 0.775 1.507* 0.016 NA NA 3/16 0.777 0.856 1.002 0.429 1.342 0.057 NA NA 3/23 1.239 0.129 1.269 0.074 NA NA 1.706 0.009 3/30 1.006 0.368 1.102 0.233 0.834 0.775 1.132 0.217 4/6 1.270 0.122 0.766 0.947 1.057 0.314 0.752* 0.953 4/13 0.924 0.563 1.166 0.166 1.044 0.334 0.791 0.892 4/20 1.187 0.158 1.000 0.431 0.942 0.551 NA NA 4/27 1.081 0.283 0.928 0.610 NA NA NA NA 5/4 1.258 0.130 0.942 0.576 1.060 0.326 0.809 0.847 5/11 1.346 0.055 1.307* 0.046 0.556* 0.990 NA NA 5/18 1.099 0.250 1.120 0.192 0.723* > 0.999 NA NA Note: a Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern ( < 1). b P value for null hypothesis of spatial randomness. values indicate significant aggregation indices ( P < 0.05) or significant uniform indices at ( P > 0.95).

PAGE 92

92 Table 3 1 4 Weekly SADIE aggregation indices of TYLCV incidence, Spri ng 2009 Date Farm B Farm H Farm I Farm J I a a P b I a a P b I a a P b I a a P b 2/16 NA NA NA NA NA NA NA NA 2/23 NA NA NA NA NA NA NA NA 3/2 NA NA NA NA NA NA NA NA 3/9 NA NA NA NA NA NA NA NA 3/16 NA NA NA NA NA NA 1.087 0.285 3/23 NA NA NA NA NA NA 1.0 87 0.285 3/30 NA NA 0.911 0.681 NA NA 1.087 0.285 4/6 0.918 0.558 1.107 0.256 NA NA 1.087 0.285 4/13 1.110 0.242 1.107 0.256 NA NA 1.087 0.285 4/20 1.081 0.283 1.107 0.256 NA NA 1.063 0.293 4/27 1.081 0.283 1.107 0.256 NA NA 1.063 0.293 5/4 1.065 0.3 03 1.107 0.256 NA NA 1.063 0.293 5/11 1.048 0.335 1.107 0.256 NA NA 1.063 0.293 5/18 1.048 0.335 1.107 0.256 NA NA 1.063 0.293 Note: a Overall index of dispersion ( I a ) indicates either an aggregated (> 1), random (= 1), or uniform pattern ( < 1). b P value for null hypothesis of spatial randomness. values indicate significant aggregation indices (P < 0.05) or significant uniform indices at (P > 0.95).

PAGE 93

93 Figure 3 1. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm A and Farm B, Fall 2007.

PAGE 94

94 Figure 3 2. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm C and Farm D, Fall 2007

PAGE 95

95 Figure 3 3. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm E and Farm F Fall 2007.

PAGE 96

96 Figure 3 4. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm G, Fall 2007.

PAGE 97

97 Figure 3 5. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm H and Farm I, Spring 2008.

PAGE 98

98 Figure 3 6. Spatial interpo lation of adult populations of B. tabaci and TYLCV from Farm J and Farm K, Spri ng 2008.

PAGE 99

99 Figure 3 7. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm A and Farm C, Fall 2008.

PAGE 100

100 Figure 3 8. Spatial interpolation of adult popu lations of B. tabaci and TYLCV from Farm L and Farm E Fall 2008.

PAGE 101

101 Figure 3 9. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm F Fall 2008.

PAGE 102

102 Figure 3 10. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm H and Farm B Spring 2009.

PAGE 103

103 Figure 3 11. Spatial interpolation of adult populations of B. tabaci and TYLCV from Farm I and Farm J Spring 2009.

PAGE 104

104 CHAPTER 4 RELATIONSHIP OF ABUNDANCE OF BEMISIA TABACI TO INCIDENCE OF TYLC V IN THE FIELD AND ITS I MPLICATIONS TO MANAG EMENT AND EPIDEMIOLO GY Purpose The sweetpotato whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) biotype B also known as the silverleaf whitefly, B. argentifolii Bellows and Perring is the key insect pest of tomatoes, Solanu m lycopersicum (L.) in south Florida (Schuster et al. 1996a) In tomatoes, Biotype B can cause direct damage including irregular ripening disorder of fruit, inhibition of fruit softening and general reduction of plant vigor (Schuster 2001, McCollum et al. 2004) In Florida, B. tabaci has become a limiting pest species due to its ab ility to vector plant viruses such as Tomato yellow leaf curl virus (TYLCV) (family Geminiviridae, genus Begomovirus ) (Polston et al. 1999) TYLCV causes one of the most de vastating diseases of cultivated tomato world wide and is transmitted in a persistent circulative manner by B. tabaci Symptoms of TYLCV in tomato are usually expressed 2 3 weeks after infection (Rom et al. 1993) and include leaf curling, chlorosis of leaf margins, reduction of leaf size, mottling, abscission of flowers, plant stunting and yield reduction (Polston et al. 1999, Mohamed 2010) First observati ons of a disease with TYLCV like symptoms was reported in Israel in 1939 1940 and was associated with outbreaks of B. tabaci (Pico et al. 1996) Later work indicated this disease was caused by a virus transmitted by B. tabaci (Cohen and Harpaz 1964) TYLCV can cause losses of 100% in tropical and subtropical regions and can be the limiting factor in commercial tomato production (Czosnek and Laterrot 1997) The effect of field populations of B. tabaci on incidence of TYLCV is of great importance to researchers and crop managers. Some south Florida tomato growers and IPM scouts have indicated confusion of the relationship between B. tabaci and

PAGE 105

105 TYLCV, i.e. some populations of B. tabaci appear to result in a higher incidence of TYLCV in tomato than others. For TYLCV management, early season adult B. tabaci populations are more important than late season populations. Plants infected early can have a greater negative impact on yield (Saikia and Muniyappa 1989) and serve as inoculum for future epidemics. Previous observations in the present study area concluded from area wide maps created from bi weekly monitoring of B. tabaci in tomato that populations appeared to originate close (< 2 km) to tomato fields (Taylor, unpublished). There were no indications of mass migrations of B. tabaci as seen in other drier areas of the world (Cohen et al. 1988, Byrne 1999) Because there was no indication of area wide migration of B. tabaci areas surrounding tomato farms were indicated as important to B. tabaci populations and TYLCV epidemiology. These p opul ations of B. tabaci were evaluated for their effect on subsequent TYLCV incidence i n south ern Florida tomato I mplications for management of both B. tabaci and TYLCV are discussed. Methods and Materials Study Sites Populations of B. tabaci and TYLCV inc idence were monitored on commercial tomato farms for four seasons in central Florida. Farm sizes ranged from 23.6 to 273.0 ha and were located in Manatee Co., Florida. The farms were selected because they were spatially isolated by distances over 10 km f rom other commercial tomato production. B. tabaci is capable of traveling 7 km, with most migration considered trivial and under 2.7 km (Cohen et al. 1988, Byrne 1999) Farms were managed by c ommercial growers so pesticide sprays and cultural practices were based on standard grower practices. There were cultivar differences between and within farms and all

PAGE 106

106 samples were taken on plastic culture staked tomatoes. After transplanting, sampling in cluded adult whitefly counts (total number of adults on 6 contiguous plants) and TYLCV incidence (visual inspection of 50 contiguous plants). Before first tie ( 3 4 weeks after transplant), adult whitefly counts were taken by scouts on whole plant samples. After first tie, counts were taken on the abaxial surface from two leaves of the third node from the top of two stems per plant, using a leaf turn technique (Naranjo et al. 1995, Palumbo et al. 1995) Scouts were advised to only record tomato plants with obvious sympto ms of TYLCV infection, which included upward curling of leaves, reduction of leaflet area and yellowing of young leaves (Polston et al. 1999) The growers farms were divid ed into blocks and sampling points remained constant throughout the season (i.e. the same 6 plants were used for B. tabaci counts and the same 50 plants were evaluated for TYLCV incidence). There was approximately one sample point for 1.4 ha within each block. Scouting methodology, sample distribution and sample size were based on results collected from previous work by Schuster et al. (2007b) Sampling va ried throughout the season as certain blocks were planted at different times, pesticide applications kept scouts from entering blocks, or rain events postponed scouting efforts. At the beginning of each season, geographical positioning system (GPS) coordi nates were collected using a GeoExplorer 3000 Series (Trimble ). Data coordinates were converted from decimal degrees to Universal Transverse Mercator (UTM) coordinate system using ArcGIS 9.2 (ESRI 2006) UTM is an adaptation of the Mercator projection and is based on distance in meters. Data Analyse s Bi weekly counts of B. tabaci adults were averaged by week over all four seasons. Incidence of plants with sympto ms of TYLCV infection was collected bi weekly and

PAGE 107

107 weekly virus percentages of infection were taken from the last recording per week. To estimate the effect of B. tabaci populations on incidence of TYLCV, the percentage of incidence of TYLCV infection from the observed week of B. tabaci counts were subtracted from the percentage of TYLCV infection three weeks after counts of B. tabaci This lag is based on the length of time to symptom expression at a conservative 3 week period (Rom et al. 1993) Other authors have used a lag period to express relationships of field populations of insect vectors and their viruses (Korie et al. 2000) B. tabaci weekly means and their subsequent lag virus were analyzed using correlation ( PROC CORR ) (SAS Institute 2002) Further analysis of the relationship between B. tabaci and TYLCV was examined by linear regression analysis using the PROC REG function of SAS (SAS Institute 2002). The slopes and intercepts of regression equations of individual farms were compared within the same week using the PROC GLM : Generalized Linear Model function of SAS (SAS Institute 2002). From this analysis, populations of B. tabaci on farms could be compared against other populations from the same week and inferences to their origin could be discussed. Using B. tabaci counts and lag incidence of TYLCV, Spatial Analyses by Distance IndicEs ( SADIE ) (version 1.22) was used to establish spatial associations between two data sets that share the s ame spatial locations (Perry and Dixon 2002) SA DIE uses spatial patterns to ass ess correlation and differs from correlation analysis because of the inclusion of the spatial relationship between B. tabaci counts and subsequent TYLCV symptom expression This spatial association is expressed as the cor relation coefficient, X with a positive association for X > 0 ( P < 0.025) and a negative association for X < 0 ( P > 0.975). The significance of X was tested against values X rand

PAGE 108

108 from a randomization test that included a Dutilleul (1993) adjustment procedure to provide a probability value. Positive values arise when two data sets have either patches or gaps coinciding spatially. Negative values indicate that gaps or patches between two data sets are spatially disassociated with each other. Other authors have determined spatial associations of different species taken at different times, two different specie s sampled together, and the same species sampled at different times (Ferguson et al. 2000, Thomas et al. 2001, Holland et al. 2005, Tillman et al. 2009, Reay Jones et al. 2010) Weekly maps of adult B. tabaci and incidence of TYLCV were interpolated using inverse distance weighted (IDW), a statistical method in GIS software ArcView 9.2 (ESRI 2006) IDW is based on the assum ption that points near by are mor e closely related than points fa rther apart and estimated predictions are based on values at sampled locations. Values are determined using a linear weighted combination of observed values and those weights are functions of the distance between locations. The most commonly used weighed power of two was used. Unlike other interpolation methods, IDW does not require a variogram model and is appropriate for small data sets (Kravchenko 2003) Cross validation was used to estimate the fit of the IDW model. Cross validation removes one sample point at a time and compares observed and predicted values for that point (Isaaks and Srivastava 1989) The root mean square prediction error (RMSE) produced by cross validation is presented as the summary statistic to check the accuracy of the model. IDW maps were created using weekly means of B. tabaci adults and final weekly lag of TYLCV incidence. Und erlying digital

PAGE 109

109 images in the form of digital orthographic photos were downloaded from the Land Boundary Information System (LABINS 2011) Results Fall 2007 In the fall of 2007, significant positive correlations (P < 0 .05) between the numbers of B. tabaci and incidence of TYLCV were found in 17.3% of comparisons of sampled farms per weeks sampled (Table 4 1). On some weeks (weeks starting on 8 August 2007 3 September 2007 10 September 2007 17 September 2007 and 2 2 October 2007 ) there were multiple farms with significant correlations between the numbers of adult whiteflies and later TYLCV incidence (Table 4 1). Of those weeks with significant correlations, some farms had significant linear regressions of B. tabaci t o TYLCV (Table 4 2). On week 20 August 2007 Farms A, B and D had significant regressions, although the slopes and intercepts were not significantly different from each other [(F = 3.03; df = 2, 101; P = 0.0529) (F = 1.2; df = 2, 103; P = 0.304), respecti vely] (Figure 4 1). On week 3 September 2007 Farms C and E had significant regressions; however, the slopes were not significantly different but the intercepts were [(F = 0.66; df = 1, 39; P = 0.422) (F = 11.13; df = 1, 40; P = 0.0018), respectively] (Fi gure 4 2). On week 1 0 September 2007 Farms E and F had significant regressions and the slopes and intercepts were significantly different [(F = 23.78; df = 1, 53; P < 0.0001) (F = 21.56; df = 1, 54; P < 0.0001), respectively] (Figure 4 3). On week 1 7 Se ptember 2007 Farms E and F had significant regressions; however, the slopes were not significantly different but the intercepts were [ (F = 2.86; df = 1, 53; P = 0.0965) (F = 24.71; df = 1, 54; P < 0.0001), respectively (Figure 4 4) ] On week 22 October 2 007 Farms D and F had significant regressions and the slopes were not significantly different but the intercepts

PAGE 110

110 were [(F = 2.16; df = 1, 57; P = 0.147) (F = 7.09; df = 1, 58; P = 0.0023), respectively] (Figure 4 5). Significant positive spatial associat ions from SADIE analysis were shown in 11.1% of available comparisons between B. tabaci numbers and TYLCV incidence (Table 4 3). Spring 2008 In the spring of 2008, significant positive correlations of B. tabaci numbers and incidence of TYLCV were found i n 10.5% and negative correlations were found in 2.6% in comparisons of sampled farms per weeks sampled (Table 4 4). On some weeks and some farms there were significant regressions of adult B. tabaci to TYLCV; however, none of the analyses were for the sam e week of sampling (Table 4 2). Significant positive spatial associations from SADIE analysis were shown in 2.6% and negative spatial associations were shown in 2.6% of available comparisons of B. tabaci numbers and TYLCV incidence (Table 4 5). Fall 2008 In the fall of 2008, significant positive correlations of the numbers of B. tabaci and incidence of TYLCV were found in 15.2% of sampled farms per weeks sampled (Table 4 6). On some weeks there were significant regressions of adult whiteflies to TYLCV (Ta ble 4 2). On week 1 September 2007 Farms C and E had significant regressions, but the slopes and intercepts were not significantly different from each other [(F = 0.06; df = 1, 66; P = 0.80) (F = 1.83; df = 1, 67; P = 0.180), respectively] (Figure 4 6). Significant positive spatial associations from SADIE analysis were shown in 6.5% of available pairs of B. tabaci and TYLCV (Table 4 7).

PAGE 111

111 Spring 2009 In the spring of 2009, significant positive correlations between numbers of B. tabaci and incidence of TY LCV were found in 8.3% of sampled farms per weeks sampled (Table 4 8). On one week there was one farm with significant regression of adult whiteflies to TYLCV (Table 4 2). Significant negative spatial associations from SADIE analysis were shown in 8.3% o f available pairs of B. tabaci and TYLCV (Table 4 9). IDW Interpolation IDW maps were created to visually express populations of B. tabaci and TYLCV for each week in which there was more than one farm with significant regression coefficients (Figures 4 7 t o 4 12) IDW interpolation was conducted on individual farms to create IDW maps of multiple farms on weekly summary maps. For interpolation methods such as IDW, mean error and root mean square error (R MS E) from cross validation tests can be used to evalu ate how precise ly the method is producing interpolation maps. The value of the mean square error depends on the scale of the data so overall, adult whitefly R MS E were similar to TYLCV incidence R MS E (Table 4 10). Smaller R MS E indicate a better fit of the model t o the observed data. Similar R MS E were presented for other IDW interpolation maps of insect counts (Tillman et al. 2009, Reay Jones et al. 2010) Discussion Tomato growers, scouts, and consultants in southern Florida report inconsistent relations hips between field populations of B. tabaci and subsequent incidence of plants with symptoms of TYLCV infection. Some populations of B. tabaci appear more viruliferous than others, which causes crop managers to treat every whitefly as

PAGE 112

112 viruliferous. We ca n conclude in our study area that not all populations of B. tabaci were equally viruliferous; however, determining the underlying reasons behind these differences is much more difficult. In this study, there were instances of B. tabaci populations being h ighly correlated with symptomatic TYLCV infected plants and other instances of B. tabaci populations having no relationship to symptomatic TYLCV i nfected plants. Depending upon the origin of the whiteflies, populations of B. tabaci could vary in their abi lity to cause TYLCV epidemics. Low correlation s of B. tabaci to TYLCV infected plants could have been due to our sampling method, because not all plants examined for symptoms of TYLCV infection were examined for adult B. tabaci Insecticide applications were also not recorded due to privacy issues with commercial growers. Due to the dynamic population fluctuations of B. tabaci it is possible that bi weekly sampling w as not precise enough to account for all populations. Since early infection of TYLCV in tomato can cause significant losses, early season immigrating adult B. tabaci populations are more important than late season populations. Early fall season populations of adult B. tabaci had stronger and more frequent correlations to incidence of TYLCV infected than late season populations in fall 2007 a nd fall 2008 (Tables 4 1 and 4 6 ). In the spring seasons, populations later in the season had stronger correlations to symptomatic TYLCV infected plants than earlier populations (Tables 4 4 and 4 8 ). Th is suggests that temperature or other variables had an effect on adult B. tabaci and incidence of TYLCV between these seasons. Also, our method of determining new symptomatic TYLCV infected plants was a destructive sample; that is once a plant was indica ted as symptomatic it would remain that way throughout the season, thus removing it from future new virus counts. Using fixed, geo

PAGE 113

113 referenced sites for evaluating incidence of TYLCV infected plants was based upon preliminary research which used geo refere nced sites that changed weekly and gave no indication o f the degree of virus progression. Also, linear regression models can be a concern for this type of analysis. The underlying principle assumes that the relationship is linear and that the distributio n is normal. These analysis tools were chosen to give a better understanding and evaluate B. tabaci populations based on the severity of subsequent virus at each farm. Because samples were taken at fixed geo referenced points, data collected was not cons idered random as the data suggested aggregated spatial patterns associated with both abundance of B. tabaci and symptomatic TYLCV infected plants. A suggested Markov Chain model could be used to model random processes between time steps as indicated by ou r disease development and fluctuations in B. tabaci counts. SADIE analyses suggested similar findings to correlations but gave an indication of the spatial association of B. tabaci and subsequent incidence of plants with symptoms of TYLCV infection. Ear ly season populations of B. tabaci in the fall of 2007 and 2008 were more likely to be positively spatially associated with symptomatic TYLCV infected plants than dates later in the season (Tables 4 3 and 4 7 ). In the spring seasons of 2008 and 2009, two of the three weeks with significant spatial associations between B. tabaci and symptomatic TYLCV infected plants were negative (Table 4 5 and 4 9) These data highlight the difficulties in analyzing such a dynamic pest. Positive spatial associations of p opulations of B. tabaci to symptomatic TYLCV infected plants are the most critical because they reveal the origin of those immigrating populations. Because of the two tailed nature of the spatial association test,

PAGE 114

114 significance is lowered to P < 0.025 and P > 0.975, reducing the number of significant associations relative to the number of significant associations indicated in the correlation analyses. Farms with positive spatial associat ions in the early season (weeks of 20 August 2007 3 September 2007 1 0 September 2007 and 17 September 2007 ) and late season (week of 22 October 2007 ) of fall 2007 provide simila r results to correlation analyse s but give some indication to the positive spatial relationships between numbers of B. tabaci and symptomatic TYLCV infected plants (Table 4 3 ). Another week, 1 September 2008 had significantly associated populations of B. tabaci to TYLCV on multiple farms (Table 4 7 ). Early season populations of whiteflies would not be originating from within the newly planted tom ato because of the short time available for whitefly reproduction. Therefore, these populations can be assumed to have originated outside tomato fields On the week of 20 August 2007, the slopes of the regression lines of the numbers of B. tabaci adults and symptomatic TYLCV infected plants from Farms A, B and D were not significantly different from each other at P < 0.05 but the slopes were significantly different at P < 0.1 (Figure 4 1). After a visual inspection of IDW maps, Farm s A and D have very di fferent densities and in farm distributions of B. tabaci populations as compared to Farm B (Figure 4 7), although the incidence of symptomatic TYLCV infected plants wa s very low on all three farms. There is no indication as to the origin of these populati ons other than mining operations along the northwest corner of Farm A and a cultural manipulation of a field (field discing) to the east of Farm D. On the weeks of 3 September 2007 17 September 2007 and 22 October 2007, only the intercepts of the regres sion lines between farms were significantly different (Figures 4 2, 4 4 and 4

PAGE 115

115 5). IDW maps of week 3 September 2007 (notice scale change), indicated different pest pressure and TYLCV incidence between Farms C and E (Figures 4 8) ; however, the regression a nalysis indicated similar rates of TYLCV infected plants (Figure 4 2) Similar findings occurred on 17 September 2007 and 22 October 2007 (Figures 4 10 and 4 11). These data suggest that B. tabaci counts were much higher on some farms, but the rate of ne wly infected tomato did not change significantly. On the week of 10 September 2007 Farms E and F had significantly different slopes indicating B. tabaci populations arose from different sources (Figure 4 3). These populations were also much higher than counts at other farms which prompted a more thorough investigation. Previous to the 10 September 2007 scouting date weeds were removed from a water ditch along the north side of Farm F adjacent to those sample sites with high B. tabaci populations (Fig ure 4 9). Although the B. tabaci populations of Farm E were located in the southeast corner closest to the ditch clean ern border, the B. tabaci populations of Farm E and F were significantly different in terms of the amount of new incidence of TYLCV infected plants three weeks later. Other weeks throughout the study had individual farms with significant regressions of B. tabaci counts to subsequent TYLCV infected plants but no comparisons to their B. tabaci populations could be ma de between weeks (Table 4 2). The present study area was separated from other tomato production by a distance greater than the suggested migratory patterns of B. tabaci (Cohen et al 1988) This site was selected to reduce the amount of tomato area that needed to be scouted and to reduce the influence of non sampled tomato farms influencing populations. Cultivated tomato has been indicated as a main influence of B. tabaci and TYLC V outbreaks, but

PAGE 116

116 as shown in some areas of the world, weeds can influence the system as well (Cohen et al. 1988, Polston and Lapidot 2007) Studies in south Florida indicated that weeds, especially in the o ver summering period, are poor intermediate hosts of B. tabaci (Stansly 1995) but populations within weeds paralleled those found in neighboring tomato fields, suggesting that weeds can bridge the gap between crops (Schuster et al. 1992) In India, populations of immature lifestages of B. tabaci found on weeds outnumbered those found on tomato suggesting the importance of weeds (Ramappa et al. 1998) As indicated by an epidemiological model, a B. tabaci vectored Indian tomato leaf c url geminivirus (TLCV), was primarily influenced by the immigration of vectors from alternative hosts (Holt et al. 1997) The authors demo nstrated that disease incidence was sensitive to vector mortality only when vector numbers were low. Viruliferous vectors may migrate into tomato in numbers greater than those needed for disease (Holt et al. 1997) Similar results were presented in this study. Seasons had varying levels of pest pressure and in those seasons which had very low B. tabaci counts subsequent viru s incidence was also low. In the seasons where B. tabaci counts were high, incidence of TYLCV was also high even though management tactics were similar across all seasons. These results indicate that the system is much more complex than previously thoug ht and confirms the theory that areas closely surrounding tomato fields are very important to B. tabaci populations and TYLCV epidemiology (Turechek 2010) These results indicate the need for furth er research into the influence of weeds and other hosts including cultivated crops. Area wide management is important, as B.

PAGE 117

117 tabaci has over 600 hosts and could be considered a mobile pest (Mound and Halsey 1978, G reathead 1986, Secker et al. 1998, Byrne 1999) The origins of viruliferous populations in the present study are unclear. There were indications that B. tabaci may have migrated from areas in which possible whitefly hosts were destroyed or disturbed, in cluding a mining operation, a fallow field and a large drainage ditch. Some cucurbit species have been shown to host TYLCV Mld (Anfoka et al. 2009) a nd some cultivars of pepper can host TYLCV (Polston et al. 2006) Cucu rbits were in production with in the study area but unfortunately limited resources precluded scouting them Also no weeds were indicated as hosts of TYLCV in a recent survey in west central Florida (Polston et al. 2009) F allow fields can harbor hosts of both B. tabaci and TYLCV and it has been suggested that fallow fields be planted with a known non host of both B. tabaci and TYLC V. Sorghum sudangrass, Sorghum bicolor (L.) has been suggested as one such option because, not only is it a non host of either pest, it also suppresses growth of broadleaf plants through allelopathy (Putnam et al. 1983 ) T o manage B. tabaci and TYLCV, it is recommended that field preparation for future plantings should be spatially and temporally separated from the early season plantings when tomato is most vulnerable t o B. tabaci and TYLCV.

PAGE 118

118 Table 4 1. Correlations of adult B. tabaci weekly means to tomato plants with new incidence of TYLCV in fection three weeks later Fall 2007 Date Farm A Farm B Farm C Farm D Farm E Farm F Farm G R a P b R a P b R a P b R a P b R a P b R a P b R a P b 8/13 0.06 0.740 NA NA NA NA NA NA 0.03 0.902 NA NA NA NA 8/20 0.43* 0.007 0.57* 0.001 0.13 0.635 0.51* 0.003 0.10 0.617 0.16 0.523 NA NA 8/27 0.07 0.688 0.26 0.057 0.35 0.206 0.61* 0.001 0.10 0.618 0.09 0.651 NA NA 9/3 0.22 0.181 0.11 0. 428 0.56* 0.031 0.04 0.825 0.58* 0.001 0.13 0.514 NA NA 9/10 0.12 0.460 0.22 0.096 0.32 0.244 0.15 0.496 0.63* 0.001 0.37* 0.049 0.03 0.881 9/17 0.05 0.747 0.11 0.427 0.01 0.979 0.12 0.567 0.38* 0.047 0.40* 0.033 0.12 0.517 9/24 0.27 0.096 0.07 0.608 0.01 0.973 0.21 0.257 0.19 0.343 0.02 0.918 0.01 0.950 10/1 0.01 0.938 0.09 0.519 0.16 0.562 0.13 0.466 0.00 0.999 0.17 0.388 0.25 0.166 10/8 0.25 0.137 0.12 0.386 0.02 0.938 0.08 0.648 0.00 0.994 0.50* 0.006 0.26 0.162 10/15 0.15 0.37 1 0.08 0.546 0.21 0.459 0.15 0.408 0.15 0.447 0.50* 0.007 0.07 0.693 10/22 0.09 0.590 0.18 0.196 0.34 0.217 0.36* 0.046 0.19 0.344 0.54* 0.003 0.24 0.170 10/29 0.10 0.545 0.03 0.829 0.31 0.261 0.08 0.653 0.32 0.102 0.07 0.724 0.23 0.195 11/5 N A NA 0.05 0.691 NA NA 0.08 0.681 0.20 0.297 0.24 0.240 NA NA Note: a Correlation of adult B. tabaci to new TYLCV three weeks later. b P value of < 0.05 indicates significant values for = 0.1. values indicate significance. NA indicates one or more variables with no value.

PAGE 119

119 Table 4 2. Regression of B. tabaci adults to incidence of tomato plants with symptoms of TYLCV infection three weeks later 2007 2009 Season Date Farm Slope Int ercept R 2 P value Fall 2007 8/20 A 0.178 0.612 0.19 0.007 B 1.202 0.0859 0.33 0.0002 D 0.346 0.631 0.26 0.0027 8/27 D 0.147 0.917 0.38 0.0002 9/3 C 0.336 0.0531 0.31 0.0307 E 0.86 4.311 0.34 0.0012 9/10 E 2.122 7.49 0.4 0.0003 F 0.143 4.015 0.14 0.0489 9/17 E 4.937 18.067 0.14 0.0472 F 1.181 4.251 0.16 0.0325 10/8 F 6.155 8.789 0.25 0.006 10/15 F 6.249 10.553 0.24 0.0069 10/22 D 7.691 9.909 0.13 0.0458 F 2.291 4.51 0.29 0.0028 Spring 2008 3/3 I 1 0 0.49 0.0001 3/31 J 0 .551 0.984 0.09 0.0357 4/7 H 0.332 0.113 0.03 0.0269 4/14 H 0.609 0.058 0.11 0.0001 Fall 2008 9/1 C 0.608 0.52 0.15 0.0138 E 0.521 0.0034 0.14 0.0383 9/8 C 1.25 0.475 0.21 0.003 9/22 E 2.925 0.165 0.29 0.0026 10/13 E 2.26 0.0179 0.38 0.0003 10/20 L 0.552 0.0154 0.06 0.0285 11/17 F 1.028 0.287 0.22 0.003 Spring 2009 3/30 B 0.4 0.0001 0.18 0.0093

PAGE 120

120 Table 4 3. Indices of spatial association ( X ) for adult B. tabaci and tomato plants with new symptoms of TYLCV infection three weeks later Fall 2007 Date Farm A Farm B Farm C Farm D Farm E Farm F Farm G X a P b X a P b X a P b X a P b X a P b X a P b X a P b 8/13 0.210 0.080 NA NA NA NA NA NA 0.030 0.445 NA NA NA NA 8/20 0.433 0.028 0.57* 0.005 0.134 0.197 0.51* 0.007 0.099 0.310 0.161 0.61 8 NA NA 8/27 0.043 0.378 0.256 0.040 0.347 0.104 0.61* 0.003 0.099 0.293 0.088 0.591 NA NA 9/3 0.222 0.090 0.108 0.222 0.558 0.045 0.041 0.597 0.58* 0.002 0.126 0.739 NA NA 9/10 0.123 0.221 0.222 0.956 0.321 0.835 0.146 0.254 0.63* 0.001 0.369 0.05 3 0.033 0.420 9/17 0.054 0.340 0.108 0.792 0.007 0.504 0.123 0.285 0.378 0.029 0.398 0.035 0.117 0.706 9/24 0.271 0.061 0.070 0.652 0.010 0.456 0.206 0.135 0.51* 0.007 0.035 0.557 0.113 0.458 10/1 0.013 0.490 0.091 0.710 0.163 0.602 0.134 0.730 0.000 0.435 0.117 0.192 0.247 0.910 10/8 0.246 0.948 0.122 0.787 0.022 0.392 0.084 0.624 0.011 0.452 0.50* 0.007 0.266 0.080 10/15 0.149 0.819 0.089 0.704 0.207 0.278 0.152 0.783 0.150 0.699 0.49* 0.009 0.073 0.344 10/22 0.090 0.260 0.175 0. 937 0.339 0.099 0.356 0.033 0.188 0.788 0.54* 0.015 0.245 0.102 10/29 0.101 0.699 0.029 0.393 0.310 0.732 0.083 0.641 0.316 0.974 0.069 0.291 0.245 0.102 11/5 NA NA 0.058 0.339 NA NA 0.076 0.435 0.204 0.841 0.285 0.115 NA NA Note: a Overall inde x of association ( X ) between adult B. tabaci and incidence of TYLCV three weeks later. b P value for positive association for X > 0 ( P < 0.025) and a negative association for X < 0 ( P > 0.0975). values indicate significant values. NA indicates one or more variables with no value.

PAGE 121

121 Table 4 4. Correlations of adult B. tabaci weekly means to tomato plants with new incidence of TYLCV in fection three weeks later, Spring 2008 Date Farm H Farm I Farm J Farm K R a P b R a P b R a P b R a P b 1/21 NA NA NA NA NA NA NA NA 1/28 NA NA NA NA NA NA NA NA 2/4 0.136 0.095 NA NA 0.301 0.240 NA NA 2/11 0.032 0.701 NA NA 0.106 0.498 NA NA 2/18 0.030 0.744 NA NA 0.113 0.454 NA NA 2/25 0.079 0.337 0.032 0.792 0.200 0.143 NA NA 3/3 NA NA 0.703* 0.001 0.137 0.318 N A NA 3/10 0.046 0.641 NA NA 0.088 0.059 NA NA 3/17 0.033 0.690 NA NA 0.045 0.748 NA NA 3/24 0.049 0.557 NA NA 0.202 0.138 0.091 0.779 3/31 0.059 0.472 0.099 0.358 0.303* 0.027 0.041 0.898 4/7 0.180* 0.027 0.042 0.655 0.127 0.356 0.135 0.676 4/14 0.327* 0.001 0.067 0.474 0.058 0.674 0.104 0.747 4/21 0.041 0.613 0.064 0.494 0.181* 0.027 0.155 0.630 4/28 NA NA 0.007 0.944 0.245 0.072 0.404 0.193 5/5 NA NA 0.111 0.235 NA NA NA NA Note: a Correlation of adult B. tabaci to new TYLCV three w eeks later. b P value of < 0.05 indicates significant values for = 0.1. values indicate significance. NA indicates one or more variables with no value.

PAGE 122

122 Table 4 5. Indices of spatial association ( X ) for adult B. tabaci and tomato plants with new s ymptoms of TYLCV infection three weeks later Spring 2008 Date Farm H Farm I Farm J Farm K X a P b X a P b X a P b X a P b 1/21 NA NA NA NA NA NA NA NA 1/28 NA NA NA NA NA NA NA NA 2/4 0.136 0.044 NA NA 0.301 0.798 NA NA 2/11 0.032 0.148 NA NA 0.125 0. 499 NA NA 2/18 0.030 0.101 NA NA 0.113 0.539 NA NA 2/25 0.079 0.373 0.032 0.082 0.200 0.105 NA NA 3/3 NA NA 0.703 0.027 0.137 0.181 NA NA 3/10 0.046 0.571 NA NA 0.088 0.233 NA NA 3/17 0.033 0.491 NA NA 0.045 0.255 NA NA 3/24 0.049 0.331 NA NA 0 .203 0.969 0.091 0.083 3/31 0.059 0.886 0.099 0.547 0.300* 0.989 0.041 0.249 4/7 0.192 0.048 0.043 0.173 0.127 0.218 0.135 0.173 4/14 0.327* 0.006 0.067 0.312 0.058 0.378 0.104 0.253 4/21 0.041 0.970 0.013 0.410 0.181 0.160 0.155 0.445 4/28 N A NA 0.016 0.617 0.245 0.070 0.404 0.250 5/5 NA NA 0.111 0.745 NA NA NA NA Note: a Overall index of association ( X ) between adult B. tabaci and incidence of TYLCV three weeks later. b P value for positive association for X > 0 ( P < 0.025) and a negat ive association for X < 0 ( P > 0.0975). values indicate significant values. NA indicates one or more variables with no value.

PAGE 123

123 Table 4 6. Correlations of adult B. tabaci weekly means to tomato plants with new incidence of TYLCV in fection three weeks later Fall 2008 Date Farm A Farm C Farm E Farm F Farm L R a P b R a P b R a P b R a P b R a P b 8/25 NA NA 0.087 0.596 NA NA NA NA 0.131 0.294 9/1 NA NA 0.386* 0.014 0.380* 0.038 0.091 0.694 0.142 0.255 9/8 NA NA 0.457* 0.003 NA NA 0.033 0.813 0.069 0.537 9/15 NA NA 0.258 0.109 0.000 1.000 0.003 0.986 0.013 0.909 9/22 NA NA 0.095 0.594 0.538* 0.003 0.063 0.654 0.045 0.687 9/29 NA NA 0.009 0.955 0.000 1.000 0.065 0.642 0.072 0.496 10/6 NA NA 0.065 0.690 NA NA 0.024 0.865 0.188 0.131 10/13 NA NA 0.103 0.563 0.617* 0.001 0.235 0.091 0.016 0.888 10/20 NA NA 0.026 0.873 0.251 0.180 0.025 0.860 0.242* 0.029 10/27 NA NA 0.060 0.714 0.040 0.836 0.109 0.436 0.057 0.656 11/3 NA NA 0.002 0.989 0.219 0.245 0.098 0.486 0.135 0.228 11/10 NA NA 0.081 0.620 0.258 0.169 NA NA 0.004 0.975 11/17 NA NA 0.090 0.583 0.146 0.451 0.469* 0.003 NA NA Note: a Correlation of adult B. tabaci to new TYLCV three weeks later. b P value of < 0.05 indicates significant values for = 0.1. values indica te significance. NA indicates one or more variables with no value.

PAGE 124

124 Table 4 7. Indices of spatial association ( X ) for adult B. tabaci and tomato plants with new symptoms of TYLCV infection three weeks later Fall 2008 Date Farm A Farm C Farm E Farm F Fa rm L X a P b X a P b X a P b X a P b X a P b 8/25 NA NA 0.061 0.126 NA NA NA NA 0.131 0.816 9/1 NA NA 0.219 0.123 0.380* 0.023 0.091 0.145 0.142 0.127 9/8 NA NA 0.000 0.483 NA NA 0.033 0.264 0.069 0.703 9/15 NA NA 0.273 0.050 0.000 0.377 0.003 0.391 0 .013 0.448 9/22 NA NA 0.360 0.030 0.538* 0.002 0.063 0.605 0.045 0.494 9/29 NA NA 0.151 0.746 0.000 0.370 0.065 0.386 0.076 0.368 10/6 NA NA 0.014 0.427 NA NA 0.024 0.352 0.188 0.105 10/13 NA NA 0.059 0.323 0.617* 0.001 0.235 0.082 0.016 0.355 1 0/20 NA NA 0.067 0.465 0.251 0.120 0.025 0.552 0.242 0.040 10/27 NA NA 0.145 0.253 0.040 0.396 0.109 0.464 0.057 0.382 11/3 NA NA 0.181 0.151 0.219 0.887 0.098 0.808 0.135 0.084 11/10 NA NA 0.075 0.321 0.258 0.137 NA NA 0.004 0.534 11/17 NA NA 0 .191 0.116 0.146 0.220 0.469 0.030 NA NA Note: a Overall index of association ( X ) between adult B. tabaci and incidence of TYLCV three weeks later. b P value for positive association for X > 0 ( P < 0.025) and a negative association for X < 0 ( P > 0.0975 ). values indicate significant values. NA indicates one or more variables with no value.

PAGE 125

125 Table 4 8. Correlations of adult B. tabaci weekly means to tomato plants with new in fection of TYLCV incidence three weeks later Spring 2009 Date Farm B Farm H Farm I Farm J R a P b R a P b R a P b R a P b 2/16 NA NA NA NA NA NA NA NA 2/23 NA NA NA NA NA NA NA NA 3/2 NA NA NA NA NA NA NA NA 3/9 0.107 0.429 NA NA NA NA NA NA 3/16 0.066 0.626 0.126 0.137 NA NA NA NA 3/23 0.025 0.085 0.046 0.546 NA NA 0.102 0 .678 3/30 0.422* 0.009 NA NA NA NA 0.165 0.500 4/6 NA NA 0.028 0.725 NA NA NA NA 4/13 NA NA 0.040 0.605 NA NA NA NA 4/20 NA NA 0.172 0.057 NA NA NA NA 4/27 NA NA 0.032 0.790 NA NA NA NA Note: a Correlation of adult B. tabaci to new TYLCV three w eeks later. b P value of < 0.05 indicates significant values for = 0.1. values indicate significance. NA indicates one or more variables with no value.

PAGE 126

126 Table 4 9. Indices of spatial association ( X ) for adult B. tabaci and tomato plants with new s ymptoms of TYLCV infection three weeks later Spring 2009 Date Farm B Farm H Farm I Farm J X a P b X a P b X a P b X a P b 2/16 NA NA NA NA NA NA NA NA 2/23 NA NA NA NA NA NA NA NA 3/2 NA NA NA NA NA NA NA NA 3/9 0.182 0.296 NA NA NA NA NA NA 3/16 0.059 0 .466 0.127 0.163 NA NA NA NA 3/23 0.026 0.034 0.047 0.331 NA NA 0.102 0.160 3/30 0.422 0.138 NA NA NA NA 0.165 0.393 4/6 NA NA 0.029 0.128 NA NA NA NA 4/13 NA NA 0.041 0.239 NA NA NA NA 4/20 NA NA 0.171 0.186 NA NA NA NA 4/27 NA NA 0.032* >.99 9 NA NA NA NA Note: a Overall index of association ( X ) between adult B. tabaci and incidence of TYLCV three weeks later. b P value for positive association for X > 0 ( P < 0.025) and a negative association for X < 0 ( P > 0.0975). values indicate signi ficant values. NA indicates one or more variables with no value.

PAGE 127

127 Table 4 10. Cross validation results of IDW interpolation analysis for B. tabaci weekly means and tomato plants with new symptoms of TYLCV incidence three weeks later Fall 2007 2008 Seas on Date Farm Adult B. tabaci Mean Error Adult B. tabaci RMSE TYLCV Incidence Mean Error TYLCV Incidence RMSE Fall 2007 8/20 A 0.0441 4.165 0.021 1.849 B 0.00958 0.573 0.036 1.259 D 0.348 3.312 0.103 2.105 9/3 C 0.185 2.239 0.1 1.672 E 0. 369 5.497 0.774 6.331 9/10 E 0.458 4.51 3.477 12.57 F 0.0136 13.15 0.172 6.106 9/17 E 0.141 1.401 3.923 16.2 F 0.0337 2.195 0.225 6.489 10/22 D 0.00734 0.465 0.535 8.924 F 0.23 2.64 0.231 12 Fall 2008 9/1 C 0.0449 0.822 0.0224 1 .54 E 0.0602 1.542 0.15 1.886 Note: Adult B. tabaci counts were summarized by weekly averages and incidence of TYLCV was taken from the new virus three weeks after the B. tabaci counts. Smaller root mean square error (RSME) indicates a better fit of the model.

PAGE 128

128 Figure 4 1. Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later Week of 20 August 2007 y = 0.178x + 0.612 R = 0.19 y = 0.346x + 0.631 R = 0.26 y = 1.202x + 0.0859 R = 0.33 0 1 2 3 4 5 6 7 8 9 0 5 10 15 20 25 TYLCV Incidence Adult B. tabaci per 6 plants Farm A Farm D Farm B

PAGE 129

129 Figure 4 2. Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later Week of 3 September 2007 y = 0.336x 0.0531 R = 0.31 y = 0.86x + 4.311 R = 0.34 0 5 10 15 20 25 30 0 5 10 15 20 25 30 TYLCV Incidence Adult B. tabaci per 6 plants Farm C Farm E

PAGE 130

130 Figure 4 3. Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later Week of 10 September 2007 y = 2.122x + 7.49 R = 0.40 y = 0.143x + 4.015 R = 0.14 0 10 20 30 40 50 60 0 20 40 60 80 TYLCV Incidence Adult B. tabaci per 6 plants Farm E Farm F

PAGE 131

13 1 Figure 4 4. Regression of adult B. tabaci counts on tomato pl ants with new symptoms of TYLCV three weeks later Week of 17 September 2007 y = 4.937x + 18.067 R = 0.14 y = 0.181x + 4.251 R = 0.16 0 10 20 30 40 50 60 70 0 2 4 6 8 10 TYLCV Incidence Adult B. tabaci per 6 plants Farm E Farm F

PAGE 132

132 Figure 4 5. Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later Week of 22 October 2007 y = 7.691x + 9.909 R = 0.13 y = 2.291x + 4.51 R = 0.29 0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 TYLCV Incidence Adult B. tabaci per 6 plants Farm D Farm F

PAGE 133

133 Figure 4 6. Regression of adult B. tabaci counts on tomato plants with new symptoms of TYLCV three weeks later Week of 1 September 2008 y = 0.608x + 0.52 R = 0.15 y = 0.521x + 0.0034 R = 0.14 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 TYLCV Incidence Adult B. tabaci per 6 plants Farm C Farm E

PAGE 134

134 Figure 4 7. Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three w eeks later from Farms A, B, and D from the w eek of 20 August 2007

PAGE 135

135 Figure 4 8. Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms C and E from the w ee k of 3 September 2007

PAGE 136

136 Figure 4 9. Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms E and F from the w eek of 10 September 2007

PAGE 137

137 Figure 4 10. Spat ial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms E and F from the w eek of 17 September 2007

PAGE 138

138 Figure 4 11. Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants with new symptoms of TYLCV infection three weeks later from Farms D and F from the w eek of 22 October 2007

PAGE 139

139 Figure 4 12. Spatial interpolation of adult populations of B. tabaci and incidence of tomato plants wit h new symptoms of TYLCV infection three weeks later from Farms C and E from the w eek of 1 September 2008

PAGE 140

140 CHAPTER 5 FACTORS INFLUENCING ABUNDANCE AND SEVERI TY OF BEMISIA TABACI AND TYLCV IN TOMATO Purpose Biotype B of the sweetpotato whitefly, Bemisia taba ci (Gennadius) (Hemiptera: Aleyrodidae), also known as the silverleaf whitefly, B. argentifolii Bellows and Perring, is a serious pest of many agricultural crops around the world (Perring et al. 1993) Biotype B has become the key insect pest of tomatoes, Solanum lycopersicum (L.), in s outh ern Florida (Schuster et al. 1996a) displacing the native non B biotypes (McKenzie et al. 2004) Biotype B of B. tabaci can cause direct damage to tomatoes including an irregular r ipening disorder of fruit, inhibition of fruit softening and general reduction of plant vigor (Schuster et al. 1996b, Schuster 2001, McCollum et al. 2004) In southern Florida, B. tabaci has become a limiting pest species due to its ability to vector plant viruses such as Tomato yellow leaf curl virus (TYLCV) (family Geminiviridae, genus Begomovirus ) (Polston et al. 1999) TYLCV cau ses one of the most devastating diseases of cultivated tomato world wide. Infection by TYLCV can result in losses of up to 100% in tropical and subtropical regions and can be the limiting factor in commercial tomato production (Czosnek and Laterrot 1997) Many envi ronmental and geographical factors can influence distribution of B. tabaci and TYLCV. One of the most important environmental variables for all insects is temperature, which can a ffect both life history and flight of B. tabaci (Butler et al. 1983, Blackmer and Byrne 1993b) Relative humidity a ffects B. tabaci developm ent with extreme humidit y being unfavorable for development (Gerling et al. 1986) B. tabaci also exhibited a greater phototacti c orientation between relative humidit y of 40 60% in glasshouse studies (Blackmer and Byrne 1993a) Bec ause B. tabaci are capable of

PAGE 141

141 sustaining flights longer than 2 hours and into head winds up to 30 cm/s, wind speed and direction could influence long distance and trivial migrations (Byrne 1999, Isaacs et al. 1999) Adult populations of B. tabaci decline after a rain event (Zalom et al. 1985, Henneberry et al. 1995) and in cotton and cantaloupe, ove rhead watering has been shown to reduce numbers of whitefly eggs and nymphs (Castle et al. 1996, Castle 2001) B. tabaci also shows preference for landing, feeding and oviposition on its host range of over 600 plant species (Mound and Halsey 1978, Greathead 1986, Secker et al. 1998) Along with other insects, B. tabaci has exhibited edge effects in tomato (Garcia 2006, Turechek 2010, Taylor unpublished). Cultural controls which include UV reflective soil mulches, have b een shown to reduce settling of B. tabaci adults on tomato plants (Csizinszky et al. 1997, Csizinszky et al. 1999) TYLCV distribution is directly related to the movement of B. tabaci but environmental factors can in fluence symptom expression in the plant (Lapidot et al. 2000, Lapidot et al. 2006) Individual cultivars within species can vary in their expression of TYLCV symptoms and some hosts can be symptomless (Lapidot et al. 2006, Polston et al. 2006) D ate of infection can have a great impact on the severity of TYLCV epidemics and early infection of TYLCV in tomato causes the greatest yield loss (Polston and Lapidot 2007) Classification and regression trees (CART) are useful for analyzing complex ecological data, including evaluating the relationships between biotic and/or abiotic factors in the system (Grunwald et al. 2009) CART is suited to analyz e datasets that are nonlinear, have missing data values and are complex (Breiman et al. 198 4) Classical regression methods rely on assumptions of the distribution and variance of data whereas CART based models do not require the data to be linear and can handle

PAGE 142

142 of target variables by repeatedly splitting the data. Each split is defined by a simple rule based on a single explanatory variable to which the overall goal of the split is to make each partition as homogen e ous as possible (De'ath and Fabricius 2000) Parent nodes are always split into exactly two child nodes, which may then be further split leading to a tree. Optimized splitting rules are identified at each level of the tree. CART has been used to identify factors in fluencing insect populations including southern pine beetles, sawflies, and honey bees (MacQuarrie et al. 2010, vanEngelsdorp et al. 2010, Duehl et al. 2011) In this study, CART analysis was used to evaluate the environmental, geographical and cropping variables that influenced the distribution of B. tabaci adult populations and plants with symptoms of TYLCV infection in southern Florida tomato. Methods and Materials Study Sites Populations of B. tabaci and incide nce of TYLCV infected plants were monitored on commercial tomato farms for four seasons in west central Florida. Farm sizes ranged from 23.6 to 273.0 ha and were located in a study area of 53.8 km 2 in Manatee Co., Florida. The farms were selected becau se they were spatially isolated by distances over 10 km from other commercial tomato production. Farms were managed by commercial growers so pesticide sprays and cultural practices were based on standard grower practices. There were cultivar differences between and within farms and all samples were taken on plastic cultured, staked and tied tomatoes. Twice weekly sampling by scouts was initiated as each field was transplanted and included adult whitefly counts (total number of adults on 6 contiguous pla nts) and incidence of TYLCV infection (visual inspection of 50 contiguous plants). Before first tie

PAGE 143

143 ( 3 4 weeks after transplanting), whitefly counts were taken on whole plant samples. After first tie, counts were taken on the abaxial surface from two l eaves at the third node from the top of two branches on each of the six plants, using a leaf turn technique (Naran jo et al. 1995, Palumbo et al. 1995) Incidence of TYLCV infected plants was based on a conservative estimate of visual symptom expression and was considered cumulative throughout the growing season. Scouts were trained to only record tomato plants with obvious symptoms of TYLCV infection, which included upward curling of leaves, reduction of leaflet area and yellowing of young leaves (Polston et al. 1999) arms were divided into blocks and sampling points remained constant throughout the season (i.e. the same 6 plants were used for B. tabaci counts and the same 50 plants were evaluated for incidence of TYLCV infection). Scouting methodology was determined by previous work by Schuster et al. (2007b) Explanatory Variables Predictor variables were environmental, geographical, or cropping variables (Table 5 1). Environmental data were taken from the nearest Florida Automated Weather Network (FAWN) located in Balm Florida which was approximately 10 miles from the study area. Environmental data from the previous day was chosen as the most important for whitefly movement because sampling occurred from early morning until mid afternoon of each sample day. Along with daily averages, the hours between 6:00 and 10:00 AM were included because those times have been indicated as peak hours for B. tabaci flight (Blackmer and Byrne 1993b) Rel ative humidity and rainfall have been indicated as factors in both life history and flight propensity in B. tabaci For predicting TYLCV incidence both wind speed and direction were excluded because

PAGE 144

144 wind speed from the day of virus counts would not have any effect on TYLCV incidence due to lag of symptom expression. Geographical variables such as proximity to a buffer area and area of block were included in this analysis. Buffer was indicated as the distance (m) to nearest non tomato. Non tomato buffer includes non tomato crops, old tomato fields, woods, weed banks, ditches and other similar areas. This excludes roads between fields and any non tomato area under 5 m in width. Area of block was indicated by the area (ha) of contiguous tomato surrounding a sample site. Tomato blocks were considered contiguous if separation between tomato fields was less than 100 m. Mulch types (white, UV reflective, black) row orientation (north/south, northeast/southwest, east/west, northwest/southeast) and grass alon g in field, shallow drainage ditches (grass planted for wind protection of newly set tomato transplants ), varied throughout seasons and within farms. Tomato type (round, roma, grape and cherry) has been indicated as an important variable in management of both B. tabaci and TYLCV in southern Florida (Polston and Lapidot 2007) CART Analysis The data analysis used CART methodology developed by Breiman et al. (1984) implemented in CART 6.0 software (Salford Systems, San Diego, CA). In the single regression tree analysis a dult B. tabaci and TYLCV incidence were targe t variables (dependent) and other environmental, geographical and cropping factors were predictor variables (independent) (Table 5 1). The CART analysis repetitively splits the data into homogeneous subsets. Splitting is conducted until all splits are pu re as compared to their parental node and the least squares splitting rule was used.

PAGE 145

145 CART can test the predictive capacity of the obtained trees and a 10 fold cross validation method was used. Cross validation allows CART to build trees based on data su bsets then calculates the error rate based on the unused portions of the data set. The data set is first broken into 10 randomly selected partitions. At each step, nine parts of the data are used to construct the largest possible tree and the remaining p art is used to obtain estimates of the error rate of the selected sub tree. This procedure is repeated for each partition and the error rate is averaged over all ten partitions to generate the cross validated relative error. The resubstitution relative e rror (RRE) is a measure of the goodness of the fit of the model for the regression tree predictions and 1 REE is an approximation of the R 2 statistic (Breiman et al. 1984, Steinberg and Colla 1995) lim it on the number of nodes one can express. CART also allows for the calculation of the relative importance of each explanatory variable. In this study, multiple trees are presented as possible tree sequences for adult B. tabaci and TYLCV incidence (Table 5 2). Our study consisted of four seasons so data from f all of 2007 and fall 2008 were combined and data from s pring of 2008 and spring 2009 were combined for analysis (Table 5 2). Results B. tabaci CART analysis of B. tabaci in the combined fall seas ons indicated average temperature (100) was the most important predictor variable, followed by wind direction (79.66), wind speed 6:00 10:00 AM (73.52), wind speed average (65.97), buffer (62.07), wind direction 6:00 10:00 AM (60.26), relative humidity (55 .47), area of block (42.29), 3). To express the importance of these predictor

PAGE 146

146 variables, a regression tree of 18 terminal nodes is presented in Figure 5 1. In the first split, farm is indicated as the most important variable, with farms A, B, C, F, G and L having fewer whiteflies [Average (Avg) = 0.543, standard deviation (STD) = 2.42] than farms D and E (Avg = 2.179, STD = 5.945). In the fall seasons, environmental variables such as temperature and wind speed/direct ion play a large role in predicting adult B. tabaci populations (Table 5 3 and Figure 5 1). The split leading to terminal node 1 (TNode), node 3, node 7 and node 14 was determined by average temperature, the most important predictor variable. Cooler temp eratures (< 26.13 C in the left branch and < 28.22 C in the right branch) indicated lower adult B. tabaci populations (Figure 5 1) Average temperature continued to split the left branch until temperature was < 26.20 C at node 5. At higher temperatures, buffer distance < 18.5 m resulted in higher whitefly counts. On the right tree branch, when temperature was lower, farms D and E also had lower counts of B. tabaci Buffer distance was important in multiple splits within the right tree, with shorter dis tances to non tomato increasing counts of B. tabaci Tomato type, tomato age and mulch type had relatively low importance to whitefly counts in the fall seasons. In contrast, CART analysis of B. tabaci in the combined spring seasons indicated buffer dist ance (99.99) was the most important predictor variable, followed by relative humidity (87.21), row orientation (62.6), age (57.01), average temperature (56.75) tomato type (44.44), rainfall (34.85), wind direction 6:00 block (3 .97) (Table 5 4). From the regression tree of 19 terminal nodes, environmental variables are important in the higher parts of the tree but later splits indicate the importance of buffer distance, row orientation and age (Figure 5 2). Wind direction from

PAGE 147

147 the southeast to south (< 328.67 deg) caused a small first split followed by more influential second splits on the left and right branches of average temperature and tomato type. Of those data with south west to south winds, roma tomatoes had higher whitef ly counts. On the left branch, warmer average temperature (> 23.52 C) increased whitefly counts. Farms B, H and I had lower B. tabaci populations than farms J and K. Higher relative humidit y favored an increase in whitefly counts. Buffer distances that influenced B. tabaci counts in the spring season were not as short as those found in the fall season. Like the fall seasons, spring adult B. tabaci populations were less influenced by mulch type but were more influenced by age of tomato and tomato type ( Figure 5 2) TYLCV CART analysis of the incidence of TYLCV infected plants in the combined fall seasons indicated buffer distance (100) was the most important predictor variable, followed by average temperature (99.99), area of block (76.14), farm (73.1 6), age (46.5), tomato type (44.96), row orientation (32.11), rainfall (21.62) and mulch (19.35) (Table 5 3). In the TYLCV analysis, wind data was not included, but geographical variables such as buffer distance and area of block were very important along with the environmental variable, average temperature. From the regression tree of 12 terminal nodes, the variable farm was the first split and farms B, C, F, G and L had less TYLCV incidence (Avg = 3.556, STD = 8.541) compared to farms A, D and E (Avg = 12.683, STD = 19.413). The first split was followed by average temperature and further down the tree there were splits by geographical and cropping factor predictor variables (Figure 5 3). In the fall seasons, both grape and cherry tomato types had more incidence of TYLCV than round and roma type tomatoes (Figure 5 3).

PAGE 148

148 In the combined spring seasons, CART analysis of incidence of TYLCV infected plants indicated buffer distance (100) was the most important predictor variable, followed by row orientation (80.12), farm (72.46), mulch (67.84), average temperature (34.96), relative humidity (22.71), area of block (22.66), age (20.1), tomato type (9.51) and rainfall (0.68) (Table 5 4). From the regression tree of 13 terminal nodes, farm was the first split fo llowed by average temperature. On the first split, farms B, H, I and K had very low incidence of TYLCV (Avg = 0.058, STD = 0.378) where farm J had higher incidence of TYLCV (Avg = 2.457, STD = 1.33). Similarly to the fall seasons, geographical and croppi ng factor variables were important further down the tree but average temperature played a major role in determining splits (Figure 5 4) Although buffer distance was considered very influential, differences in distance were very subtle during the spring s easons because virus incidence was very low. Discussion Overall, r esults from CART analysis suggest that temperature was very influential in predicting B. tabaci abundance and the incidence of TYLCV infected plants B. tabaci populations were highest d uring warmer conditions in the early fall seasons and the end of the spring seasons. The increase of TYLCV incidence regardless of temperature change suggests that average temperature is only an indicator of the length of season. This analysis suggests that, towards the end of the growing season, there were greater incidences of TYLCV infected plants This result was to be expected. Widespread and substantial freeze events before both the spring 2008 and spring 2009 crops were not included in the envi ronmental data set because they occurred before scouting commenced (Turechek 2010) These events had the potential to cause widespread destruction of hosts of whiteflies, including crops and weeds.

PAGE 149

149 Temperature was also low enough to cause mortality of whiteflies in unprotected areas. Relative humidity was influential for predicting B. tabaci in the fall and spring seasons and marginally influential for predicting TYLCV incidence in the spring seaso n. Relative humidit y > 8 3 % increased whitefly counts and lower humidit y < 7 6 % increased TYLCV incidence. Environmental variables such as wind speed and direction were more Higher wind speed was s hown to increase whitefly numbers in at least one terminal node split in fall 2007 (Figure 5 1) Wind speed has been positively correlated with whitefly density in similar work by Turechek (2010) Rainfall has been observed to reduce adult whitefly counts but was only marginally influential in the spring seasons. This indicated winter and spring rains influenced whitefly populations greater than in the fall. Buffer distance was im portant in predicting both adult B. tabaci and symptomatic TYLCV infected plants. In most cases, smaller buffer distances resulted in increased B. tabaci abundance and TYLCV incidence. The average buffer distance over a ll data sets most influential to ad ult whitefly counts was 38 m and for the incidence of TYLCV infected plants was 72 m. The a rea of block was marginally influential in predicting B. tabaci abundance and TYLCV incidence. Larger contiguous areas have been indicated as having fewer whitefli es and TYLCV incidence (Turechek, personal communication). In the present study, area of block was influential in the prediction of B. tabaci in two farms at the end of the fall season s with the larger block s (> 33 ha) having more whiteflies. This split of farm size was associated with a difference between farms as this split indicated the differences in populations associated with farms D and E. Both block sizes were under 35 ha though t he larger block of farm D (35 ha) was indicated to

PAGE 150

150 have higher pop ulations of B. tabaci than farm E (31 ha). Similarly, s maller blocks of tomato (< 33 ha) had a higher TYLCV incidence indicating the interior of smaller blocks were closer to buffer s and to sources of viruliferous hosts. In the spring season, the regress ion tree indicated larger field size had more TYLCV incidence After reviewing the data set, it was determined that in reality the split was caused by the differences in incidence of TYLCV infected plants between the 2008 and 2009 seasons F arm J was sma ller in size in 2008 than in 2009 seasons thus, influencing the split between TNode 12 and TNode 13 (Figure 5 4). Rows in farm F oriented east/west or northwest/southeast had higher incidences of TYLCV infection compared to rows oriented north/south or northeast/southwest. The influence of tomato type varied between seasons and was greater in the fall seasons. Age of tomato was only a predictor in one split in any tree. Middle aged tomatoes at the end of the spring seasons had more whiteflies than the earliest and latest plantings. Mulch types were more influential in predicting TYLCV incidence than predicting B. tabaci populations. Environmental, geographical, and cropping factor variables can impact both adult populations of B. tabaci and the inci dence of TYLCV infected plants CART analysis confirmed assumptions that environmental variables such as temperature, wind speed and wind direction influence populations of B. tabaci Geographical variables such as buffer distance are important to both B tabaci populations and TYLCV incidence. As indicated in a previous study smaller block size resulted in higher B. tabaci abundance and incidence of TYLCV infected plants (Turechek, personal communication). Other factors such as rainfall, mulch type an d tomato type were thought to have more

PAGE 151

151 influence on B. tabaci counts and /or incidence of TYLCV infected plants but this analysis suggests that on a large scale those factors do not have as much influence as previously thought. Future work could be desig ned to evaluate on a smaller scale some of th o se factors that were indicated as important in the present study Other variables could only be assessed on a large scale study which, as the present study points out is very dynamic with many factors influen cing B. tabaci and TYLCV.

PAGE 152

152 Table 5 1 Description of variables in the CART analysis used to study the influence of environmental, geographical and cropping variables on the abundance of B. tabaci adults and the incidence of tomato plants with symptoms of TYLCV infection Variable Character a Type b Values Farm farms C Farms A K Buffer buffer N distance to nearest non tomato (m) Tomato type tomtype C 1 (round), 2 (roma), 3 (grape), 4 (cherry) Row orientation row C 1 (north/south), 2 (northeast/southwest ), 3 (east/west), 4 (northwest/southeast) Plastic mulch mulch C 1 (white), 2 (UV reflective), 3 (black) Area of Block areaofblock N area of contiguous tomato (ha) Age of tomato age C age of tomato based on planting date by months, 1 (1st), 2 (2nd), 3 (3 rd) Grass row middles rowmid C 1 (grass), 2 (no grass) Average daily temperature tempavdb N temperature ( C) from day before sample Relative humidity rhdb N % relative humidity from day before sample Wind speed average windspavdb N wind speed average from 2 meters above ground (mph) from day before sample Wind direction average winddiravgdb N wind direction average from 2 meters above ground from day before sample Wind speed average 6:00 10:00 AM wind sp 6 10 db N wind speed average from 2 meters ab ove ground (mph) 6:00 11:00 AM from day before sample Wind direction average 6:00 10:00 AM wind dir 6 10 db N wind direction average from 2 meters above ground 6:00 11:00 AM from day before sample Sum of rainfall rainfalldb N sum of rainfall (inches) fr om day before sample Note: a Character corresponds to variable in CART analysis. b C = categorical variable and N = numeric variable.

PAGE 153

153 Table 5 2 Summary of 10 fold cross validation results to predict adult B. tabaci and TYLCV incidence Season Dependen t variable Terminal Nodes Cross validated relative error Resubstitution relative error R 2 Fall Seasons B. tabaci 139 0.72 0.36 0.64 18 0.80 0.56 0.44 TYLCV 673 0.22 0.09 0.91 12 0.55 0.53 0.47 Spring Seasons B. tabaci 24 0.81 0.46 0.54 19 0.83 0.5 0 0.5 0 TYLCV 42 0.51 0.41 0.59 13 0.66 0.61 0.39

PAGE 154

154 Table 5 3 Ranking of predictor variables on B. tabaci abundance and TYLCV incidence in Fall 2007/2008 Target variables Predictor variables Relative Importance B. tabaci tempavdb 100 winddira vdb 79.66 wind sp 6 10 db 73.52 windspavdb 65.97 buffer 62.07 wind dir 6 10 db 60.26 rhdb 55.47 areaofblock 42.29 farm 19.34 rainfalldb 6.23 tomtype 5.53 age 1.89 row 1.55 mulch 1.21 TYLCV buffer 100 tempavdb 99.99 areaofblock 76.14 farm 73.16 age 46.5 tomtype 44.96 row 32.11 rainfalldb 21.62 mulch 19.35

PAGE 155

155 Table 5 4 Ranking of predictor variables on B. tabaci abundance and TYLCV incidence in Spring 2008/2009 Target variables Predictor variables Relative Importan ce B. tabaci buffer 99.99 rhdb 87.21 row 62.6 age 57.01 tempavdb 56.75 tomtype 44.44 rainfalldb 34.85 winddir 6 10 db 25.55 windsp 6 10 db 8.57 farm 7.85 mulch 7.55 winddiravdb 7.07 windspavdb 6.73 areaofblock 3.97 TYLCV buffer 100 row 80.12 farm 72.46 mulch 67.84 tempavdb 34.96 rhdb 22.71 areaofblock 22.66 age 20.1 tomtype 9.51 rainfalldb 0.68 rowmid 0

PAGE 156

156 Figure 5 1. Regression tree of the variables influencing populations of B. tabaci in Fall 2007/2008

PAGE 157

157 Figure 5 2. Regression tree of the variables influencing populations of B. tabaci in Spring 2008/2009

PAGE 158

158 Figure 5 3. Regression tree of the variables influencing TYLCV incidence in Fall 2007/2008

PAGE 159

159 Figure 5 4. Regression tree of the variables in fluencing TYLCV incidence in Spring 2008/2009 .

PAGE 160

160 CHAPTER 6 CONCLUSIONS Biotype B of the sweetpotato whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae), also known as the silverleaf whitefly, B. argentifolii Bellows and Perring, is a serious pest of many agricultural crops around the world (Perring et al. 1993) The nomenclature of Bemisia spp. has been widely discussed and current research suggest there are multiple unique species worldwide (Dinsdale et al. 2010) The polyphagous nature of the sweetpotato whitefly leads to management problem s and could be associated with its high value pest s tatus in many commodities (Naranjo and Ellsworth 2001) Biotype B has become the key insect pest of tomatoes, Solanum lycopersicum (L.), in south ern Florida (Sch uster et al. 1996a) displacing the native non B biotypes (McKenzie et al. 2004) In tomatoes, Biotype B can cause direct damage including irregular ripening disorder of fruit, inhibition of fruit softening and general reduction of plant vigor (Schuster 2001, McCollum et al. 2004) In southern Florida, B. tabaci has become a limiting pest species in tomato due to its ability to vector Tomato yellow leaf curl virus (TYLCV) (fa mily Geminiviridae genus Begomovirus ) (Polston et al. 1999) TYLCV is vectored in a persistent circulative manner by B. tabaci and symptoms of infection in tomato include upward curling of leaflet margins, reduction of leaflet area, yellowing of young leaves, stunting of plants and abscission of flowers (Polston et al. 1999) With these symp toms there is considerable loss in plant vigor and significant yield loss particularly if infection occurs during early growth. Unfortunately, growers often rely heavily on the use of insecticides targeting the whitefly vector to c ontrol TYLCV. As a res ult, insecticide resistance is

PAGE 161

161 widespread, including many different insecticide classes (Palumbo et al. 2001, Horowitz et al. 20 07) A greater understanding of the distribution of B. tabaci and TYLCV infected plants the relationship between them, and the variables influencing populations may lead to the development of new management recommendations. The specific objectives for research were: to evaluate seasonal abundance of B. tabaci and incidence of TYLCV in Florida tomatoes, to investigate the spatial and temporal distribution of B. tabaci adults and TYLCV infected plants in Florida tomatoes, to investigate the relationship between the abundance of B. tabaci and incidence of TYLCV in Florida tomatoes, and to investigate the environmental, geographical, and cropping factor variables influencing the abundance of B. tabaci and incidence of TYLCV infected tomatoes in Florida. To accomplish these objectives, the abundance of B. tabaci and incidence of plants with symptoms of TYLCV infect ion were monitored twice weekly on commercial tomato farms from the fall 2007 season through the spring 2009 season in c entral Florida. Farm sizes ranged from 23.63 to 272.99 ha and were located in a study area of 53.8 km 2 in Manatee Co., Florida. T he analyses of the data using Geographical Information Systems (GIS), Spatial Analysis by Distance IndicEs (SADIE) and classification and regression tree (CART) analysis ha ve produced much more detailed explanation s of in field distribution, vector/ disease relationship and influencing factors than previously reported. B. tabaci is a mobile pest and has been shown in the present study to have varying aggregation in both space and time. Over the entire study area, dist ributions of whiteflies were significantly aggregated in every season but spring 2009 when the

PAGE 162

162 population of the whitefly was very low Weekly fluctuations throughout the study area suggest that within the earlier sampling dates of each season, whitefli es were more likely to be aggregated. In some seasons and on some farms, brief periods of significant re aggregation 6 10 weeks later were indicated. Aggregation of adult B. tabaci could have been influenced by migration, in farm reproduction and managem ent tactics such as pesticide applications. TYLCV distribution was more static than B. tabaci counts and TYLCV was shown to follow similar spatio temporal patterns associated with its vector. Strong spatial dependence in insect data indicates that estim ating populations at non sampled locations is possible (Liebhold et al. 1993) With the use of inverse distance weighted (IDW) maps created by a GIS program, populations of both B. tabaci and TYLCV were indicated to be associate d more closely with the edges of tomato fields This edge effect was apparen t in farms regardless of scale and continued throughout the study area. Tomato growers, scouts, and consultants in southern Florida report inconsistent relationships between field populations of B. tabaci and subsequent incidence of plants with symptoms of TYLCV infection. In this study, there were instances of B. tabaci populations being highly correlated with subsequent incidence of plants with symptoms of TYLCV infect ion and other instances of B. tabaci populations having no relationship to later inci dence of plants with symptom s of TYLCV infect ion Early fall season populations of adult B. tabaci had stronger and more frequent correlations to incidence of TYL CV than late season populations In the spring seasons, populations later in the season had stronger co rrelations to symptomatic TYLCV infected plants than earlier

PAGE 163

163 populations. Using SADIE spatial association tests, early season populations of B. tabaci in the fall of 2007 and 2008 were more likely to be positively spatially as sociated with symp tomatic TYLCV infected plants than dates later in the season. Early season populations of whiteflies would not be originating from within the newly planted tomato because of the short time available for whitefly reproduction. Therefore, these populations can be assumed to have originated outside tomato. The origins of viruliferous populations in the present study are unclear. There were indications that B. tabaci may have migrated from areas in which possible whitefly hosts were destroyed or disturbed, including a mining operation, a fallow field and a large drainage ditch. Environmental, geographical, and cropping factor variables can impact both adult populations of B. tabaci and TYLCV incidence. CART analysis confirmed assumptions that environmenta l variables such as temperature, wind speed and wind direction influence populations of B. tabaci Results from CART analysis suggest temperature was very influential in predicting B. tabaci abundance and TYLCV incidence. B. tabaci populations were highe st during warmer conditions in the early fall seasons and the end of the spring seasons. The increase of TYLCV incidence throughout the season regardless of temperature change suggests that average temperature is only an indicator of the length of season Geographical variables such as buffer distance are important to both B. tabaci abundance and TYLCV incidence. In most cases, smaller buffer distances increased B. tabaci abundance and TYLCV incidence. The average buffer distance most influential to ad ult whitefly counts was about 38 m and for TYLCV incidence was 72 m. In the present study area of block was influential in the prediction of B. tabaci in two farms at the end of the fall season, with the smaller block (< 33 ha)

PAGE 164

164 having more whiteflies. T his split of block size was associated with a difference between farms though as both block sizes were less than 35 ha S maller block size (< 33 ha) increased counts TYLCV incidence indicating the interiors of smaller blocks had less buffer distance to sources of viruliferous hosts. Other factors such as rainfall, mulch type and tomato type were thought to have more influence on B. tabaci counts and TYLCV incidence, but this analysis suggests that on a large scale those factors nce as previously thought. Future work could encompass a larger study area or increase the intensity of sampling. Future work could also include geostatistical analysis of this data set and use semivariograms along with kriging. Semivariograms express the variance of sample pairs against the distance between sample points and provide important ecological information on the spatial patterns of organisms. These spatial patterns could be used to indicate the distance between sampling locations to de velop sampling plans which require independent samples. Highly dynamic populations such as those seen in insect count data increase the level of uncertainty in geostatistical analysis and much work will have to be conducted to create validated results. These r esults also indicate the need for further research into the influence of weeds and other hosts including cultivated crops. Area wide management is important, as B. tabaci has over 600 hosts and could be considered a mobile pest (Mound and Halsey 1978, Greathead 1986, Secker et al. 1998, Byrne 1999) Future work could also be designed to address some of the factors that were indicated as important to the distribution of B. tabaci and TYLCV on a smaller plot scale. Due to the dynamic nature of insect movement some variables would only be able to be assessed on a larger scale study.

PAGE 165

165 The d istribution s of B. tabaci adults and TYLCV infected plants were aggregated on commercial tomato throughout the study area. P opul ations were shown to have not migrated from distant areas and were located along field edges. N on tomato hosts and environmental and geographical variables were shown to influence populations of B. tabaci and incidence of TYLCV infected plants

PAGE 166

166 LIST OF REFERENCES Abhary, M. K., G. H. Anfoka, M. K. Nakhla, and D. P. Maxwell. 2006. Post transcriptional gene silencing in controlling viruses of the Tomato yellow leaf curl virus complex. Arch. Virol. 151: 2349 2363. Ahmed, A. H. M., E A. Elhag, and N. H. H. Bashir. 1987. Insecticide resistance in the cotton whitefly ( Bemisia tabaci Genn.) in the Sudan Gezira. Tro. Pest. Manag. 33: 67 72. Ahmed, N. E., H. O. Kanan, Y. Sugimoto, Y. Q. Ma, and S. Inanaga. 2001. Effect of imidacloprid on incidence of tomato yellow leaf curl virus. Plant Dis. 85: 84 87. Ahohuendo, B. C. and S. Sarkar. 1995. Partial control of the spread of African cassava mosaic virus in Benin by intercropping. J. Plant Dis. Protec. 102: 249 256. Akad, F., A. Eybishtz, D Edelbaum, R. Gorovits, O. Dar Issa, N. Iraki, and H. Czosnek. 2007. Making a friend from foe: expressing a GroEL gene from the whitefly Bemisia tabaci in the phloem of tomato plants confers resistance to tomato yellow leaf curl virus. Arch. Virol. 152: 1323 1339. Al Musa, A. 1982. Incidence, economic importance, and control of tomato yellow leaf curl in Jordan. Plant Dis. 66: 561 563. Alvarez, P. A. and A. J. Abud Antun. 1995. Reporte de Republica Dominicana. CEIBA. 36: 39 47. Anfoka, G., F. Haj Ahma d, M. Abhary, and A. Hussein. 2009. Detection and molecular characterization of viruses associated with tomato yellow leaf curl disease in cucurbit crops in Jordan. Plant Pathol. 58: 754 762. Antignus, Y., N. Mor, R. Ben Joseph, M. Lapidot, and S. Cohen. 1996. Ultraviolet absorbing plastic sheets protect crops from insect pests and from virus diseases vectored by insects. Environ. Entomol. 25: 919 924. Antignus, Y., M. Lapidot, D. Hadar, Y. Messika, and S. Cohen. 1998. Ultraviolet adsorbing screens serve as optical barriers to protect crops from virus and insect pests. J. Econ. Entomol. 91: 1401 1405. Antignus, Y., D. Nestel, S. Cohen, and M. Lapidot. 2001. Ultraviolet deficient greenhouse environment affects whitefly attraction and flight behavior. Envir on. Entomol. 30: 394 399.

PAGE 167

167 Antignus, Y., R. Vunsh, O. Lachman, M. Pearlsman, L. Maslenin, U. Hananya, and A. Rosner. 2004. Truncated Rep gene originated from Tomato yellow leaf curl virus Israel [Mild] confers strain specific resistance in transgenic toma to. Ann. Appl. Biol. 144: 39 44. Arsenio, A. F., E. Neto, N. Ramos, S. Mangerico, E. Fortunato, L. Stigter, J. E. Fernandes, A. M. P. Lavadinho, and D. Jouro. 2002. Control of the Bemisia tabaci / Tomato yellow leaf curl virus complex on protected tomato crops in Algarve (Portugal). EPPO Bulletin. 32: 31 35. Arx, R. V. von, J. Baumgartner, and V. Delucchi. 1984. Sampling of Bemisia tabaci (Genn.) (Sternorrhyncha: Aleyrodidae) in Sudanese cotton fields. J. Econ. Entomol. 77: 1130 1136. Ascencio Ibanes, J T., R. Diaz Plaza, J. Mendez Lozano, Z. Monslave Fonnerga, G. R. Arguello Astroga, and R. F. Rivera Bustamante. 1999. First report of tomato yellow leaf curl geminivirus in Yucatan Mexico. Plant Dis. 83: 1178. Attard, D. 2002. Methods of controlling To mato yellow leaf curl virus (TYLCV) and its vector Bemisia tabaci in the Maltese Islands. EPPO Bulletin. 32: 39 40. Ausher, R. 1997. Implementation of integrated pest management in Israel. Phytoparasitica. 25: 119 141. Avidov, Z. 1956. Bionomics of the t obacco whitefly ( Bemisia tabaci Gennad.) in Israel. Ktavim. 7: 25 41. Avilla, C. J. L. Collar, M. Duque, P. Perez, and A. Fereres. 1997. Impact of floating rowcovers on bell pepper yield and virus incidence. HortScience. 32: 882 883. Azeb, A. K., M. M. Megahed, and H. D. El Mirsawi. 1971. On the range of host plants of Bemisia tabaci (Genn.) (Homoptera: Aleyrodidae). Bulletin de la Societe Entomologique d'Egypte. 54. Azeb, A. K., M. M. Megahed, and H. D. El Mirsawi. 1972. On the biology of Bemisia tab aci (Genn.) Bulletin de la Societe Entomologique d'Egypte. 55: 305 315. Bacci, L., A. LB. Crespo, T. L. Galvan, E. Pereira, M. C. Picanco, G. A. Silva, and M. Chediak. 2007. Toxicity of insecticides to the sweetpotato whitefly (Hemiptera: Aleyrodidae) an d its natural enemies. Pest Manage. Sci. 63: 699 706. Bai, D., S. C. R. Lummis, W. Leicht, H. Breer, and D. B. Satelle. 1991. Actions of imidacloprid and related nitromethylene on cholinergic receptors of an identified insect motor neurone. Pestic. Sci. 33: 197 204.

PAGE 168

168 Baldin, E. L. L. and R. A. Beneduzzi. 2010. Characterization of antibiosis and antixenosis to the whitefly silverleaf Bemisia tabaci B biotype (Hemiptera: Aleyrodidae) in several squash varieties. J. Pest Sci. 83: 223 229. Barinaga, M. 1993 Is devastating whitefly invader really a new species? Science. 259: 30. Barnes, J. M. R. Trinidad Correa, T. V. Orum, R. Felix Gastelum, and M. R. Nelson. 1999. Landscape ecology as the new infrastructure for improved management of plant viruses and their insect vectors in agroecosystems. Ecosystem Health. 5: 26 35. Barrigossi, J. A. F., L. J. Young, C. A. Gotway Crawford, G. L. Hein, and L. G. Higley. 2001. Spatial and probability distribution of Mexican bean beetle (Coleoptera: Coccinellidae) egg mass populations in dry bean. Environ. Entomol. 30: 244 253. Bayhan, E., M. R. Ulusoy, and J. K. Brown. 2006. Effects of different cucurbit species and temperature on selected life history traits of the 'B' Biotype of Bemisia tabaci Phytoparasitica. 34: 235 242. Beckler, A. A., B. W. French, and L. D. Chandler. 2005. Using GIS in areawide pest management: A case study in South Dakota. Transactions in GIS. 9: 109 127. Bedford, I. D., R. W. Briddon, J. K. Brown, R. C. Rosell, and P. G. Markham. 1994. Gemi nivirus transmission and biological characterization of Bemisia tabaci (Gennadius) biotypes from different geographic regions. Ann. Appl. Biol. 125: 311 325. Bellows, T. S., T. M. Perring, K. Arakawa, and C. A. Farrar. 1988. Patterns in diel flight activi ty on Bemisia tabaci (Homoptera: Aleyrodidae) in cropping systems in southern California. Environ. Entomol. 17: 225 228. Bellows, T. S. Jr., T. M. Perring, R. J. Gill, and D. H. Headrick. 1994. Description of a species of Bemisia (Homoptera: Aleyrodidae). Ann. Entomol. Soc. Am. 87: 195 206. Bentz, J. A., J. Reeves III, P. Barbosa, and B. Francis. 1995a. Effect of nitrogen fertilizer source and level on ovipositional choice of poinsettia by Bemisia argentifolii (Homoptera: Aleyrodidae). J. Econ. Entomol. 8 8: 1388 1392. Bentz, J. A., J. Reeves III, P. Barbosa, and B. Francis. 1995b. Nitrogen fertilizer effect on selection, acceptance, and suitability of Euphorbia pulcherrima (Euphorbiaceae) as a host plant to Bemisia tabaci (Homoptera: Aleyrodidae). Environ Entomol. 24: 40 45.

PAGE 169

169 Bentz, J. A., J. Reeves, P. Barbosa, and B. Francis. 1995c. Within plant variation in nitrogen and sugar content of poinsettia and its effects on the oviposition pattern, survival, and development of Bemisia argentifolii (Homoptera: Aleyrodidae). Environ. Entomol. 24: 271 277. Berlinger, M. J. 1986. Host plant resistance to Bemisia tabaci Agric. Eco. Environ. 17: 69 82. Berlinger, M. J. and S. Lebiush Mordechi. 1996. Physical methods for the control Bemisia pp. 617 634. In D. Gerl ing and R. T. Mayer [eds.], Bemisia 1995: Taxonomy, Biology, Damage, Control and Management. Intercept, Andover. Berlinger, M. J., Z. Magal, and A. Benzioni. 1983. The importance of pH in food selection by the tobacco whitefly, Bemisia tabaci Phytoparasi tica. 11: 151 160. Berlinger, M. J., R. A. J. Taylor, S. Lebiush Mordechi, S. Shalhevet, and I. Spharim. 2002. Efficiency of insect exclusion screens for preventing whitefly transmission of tomato yellow leaf curl virus of tomatoes in Israel. Bull. Entomo l. Res. 92: 367 373. Bernays, E. A. 1999. When host choice is a problem for a generalist herbivore: experiments with the whitefly, Bemisia tabaci Ecol. Entomol. 24: 260 267. Bethke, J. A. and R. A. Redak. 1997. Effect of imidacloprid on the silverleaf w hitefly, Bemisia argentifolii Bellows and Perring (Homoptera: Aleyrodidae), and whitefly parasitism. Ann. Appl. Biol. 130: 397 407. Bethke, J. A. R. A. Redak, and T. D. Paine. 1994. Screens deny specific pests entry to greenhouses. Calif. Agric. 48: 37 40. Bezerra, Mary Ann S., Maria R. V. De Oliveira, and Simo D. Vasconcelos. 2004. Does the presence of weeds affect Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) infestation on tomato plants in a semi arid agro ecosystem? Neotrop. Entomol. 33: 769 775. Bi, J. L. G. R. Ballmer, D. L. Hendrix, T. J. Henneberry, and N. C. Toscano. 2001. Effect of cotton nitrogen fertilization on Bemisia argentifolii populations and honeydew production. Entomol. Exp. Appl. 99: 25 36. Bird, T.L. and K. Kruger. 2006 Response of the polyphagous whitefly Bemisia tabaci B biotype (Hemiptera: Aleyrodidae) to crop diversification influence of multiple sensory stimuli on activity and fecundity. Bull. Entomol. Res. 96: 15 23.

PAGE 170

170 Birdsall, S. L. D. Ritter, and P. L. Carso n. 1995. Economic impact of the silverleaf whitefly in Imperial Valley, California, from 1991 to 1995, pp. 176. In T. J. Henneberry, L. C. Toscano, R. M. Faust and J. R. Coppedge [eds.], Silverleaf whitefly: 1996 supplement to the five year national resear ch and action plan. U. S. Department of Agriculture, Agricultural Research Server, Washington, DC. Blackmer, J. L. and D. N. Byrne. 1993a. Environmental and physiological factors influencing phototactic flight of Bemisia tabaci Physiol. Entomol. 18: 336 342. Blackmer, J. L. and D. N. Byrne. 1993b. Flight behavior of Bemisia tabaci in a vertical flight chamber: effect of time of day, sex, age and host quality. Physiol. Entomol. 18: 223 232. Blackmer, J. L. and D. N. Byrne. 1999. Changes in amino acid in Cucumis melo in relation to life history traits and flight propensity of Bemisia tabaci Entomol. Exp. Appl. 93: 29 40. Blackmer, J. L., V. A. Lindley, and D. N. Byrne. 1995a. Histological examination of flight muscle development and breakdown in Bemisia tabaci (Homoptera: Aleyrodidae): Relationship to age and flight behavior. J. Morphol. 226: 213 221. Blackmer, J. L., D. N. Byrne, and Z. Tu. 1995b. Behavioral, morphological, and physiological traits associated with migratory Bemisia tabaci J. Insect Be hav. 8: 251 267. Bleeker, P. M., P. J. Diergaarde, K. Ament, J. Guerra, M. Weidner, S. Schutz, M. T. J. de Both, M. A. Haring, and R. C. Schuurink. 2009. The role of specific tomato volatiles in tomato whitefly interaction. Plant Physiol. 151: 925 935. B lua, M. J. and N. C. Toscano. 1994. Bemisia argentifolii (Homoptera: Aleyrodidae) development and honeydew production as a function of cotton nitrogen status. Environ. Entomol. 23: 316 321. Boica Junior, A. L. Z. R. Campos, A. L. Lourencao, and A. L. Ca mpos. 2007. Adult attractiveness and oviposition preference of Bemisia tabaci (Genn.) (Homoptera: Aleyrodidae) B biotype in cotton genotypes. Scientia Agricola. 64: 147 151. Bolstad, P. 2005. GIS Fundamentals. Eider Press, White Bear Lake. Bonsignore, C. P., F. Manti, and V. Vacante. 2008. Field and tree distribution of Capnodis tenebrionis (Linnaeus, 1767) (Col., Buprestidae) adults in an apricot orchard in Italy. J. Appl. Entomol. 132: 216 224. Borah, R. K. 1994. Influence of planting dates on the inci dence of insect pests of brinjal ( Solanum melongena l.) in a hilly area of Assam. J. Agric. Sci. Soc. North East India. 7: 209 211.

PAGE 171

171 Borth, P.W. and R.T. Huber. 1987. Modeling pink Bollworm establishment and dispersion in cotton with the Kriging technique, pp. 267 274, Proceedings of the Beltwide Cotton Production Research Conference. National Cotton Council of America, Memphis, TN., Dallas, TX. Bosco, D., G. Mason, and G. P. Accotto. 2004. TYLCSV DNA, but not infectivity, can be transovarially inherited b y the progeny of the whitefly vector Bemisia tabaci (Gennadius). Virology. 323: 276 283. Breiman, L., J. H. Friedman, R. A. Olshen, and C. J. Stone. 1984. Classification and regression trees. Wadsworth International Group, Belmont Brown, J. K., D. R. F rohlich, and R. C. Rosell. 1995. The sweetpotato or silverleaf whiteflies: biotypes of Bemisia tabaci or a species complex. Annu. Rev. Entomol. 40: 511 534. Brown, J. K., T. M. Perring, A. D. Cooper, I. D. Bedford, and P. G. Markham. 2000. Genetic analysi s of Bemisia (Homoptera: Aleyrodidae) populations by isoelectric focusing electrophoresis. Biochem. Genet. 38: 12 25. Brunetti, A., M. Tavazza, E. Noris, R. Tavazza, P. Caciagli, G. Ancora, S. Crespi, and G. P. Accotto. 1997. High expression of truncated viral rep protein confers resistance to tomato yellow leaf curl virus in transgenic tomato plants. Mol. Plant Microbe Inter. 10: 571 579. Brunetti, A., R. Tavazza, E. Noris, A. Lucioli, G. P. Accotto, and M. Tavazza. 2001. Transgenically expressed T Rep of tomato yellow leaf curl Sardinia virus acts as a trans dominant negative mutant, inhibiting viral transcription and replication. J. Virol. 75: 10573 10581. Bryceson, K. P. 1991. Likely locust infestation areas of western New South Wales, Australia, l ocated by satellite. Geocarto International. 6: 21 37. Burban, C., L. D. C. Fishpool, C. Fauquet, D. Fargette, and J. C. Thouvenel. 1992. Host associated biotypes within West African populations of the whitefly Bemisia tabaci (Genn.), Homoptera: Aleyrodid ae. J. Appl. Entomol. 113: 416 423. Butler, D. N., T. J. Henneberry, and T. E. Clayton. 1983. Bemisia tabaci (Homoptera: Aleyrodidae): development, oviposition and longevity in relation to temperature. Ann. Entomol. Soc. Am. 76: 310 313. Butler, Jr., G. D. and T. J. Henneberry. 1984. Bemisia tabaci : effect of cotton leaf pubescence on abundance. Southwest. Entomol. 9: 91 94.

PAGE 172

172 Butler, Jr., G. D., T. J. Henneberry, P.A. Stansly, and D. J. Schuster. 1993. Insecticidal effects of selected soaps, oils and dete rgents on the sweetpotato whitefly: (Homoptera: Aleyrodidae). Flor. Ento. 76: 161 167. Byamukama, E., R. W. Gibson, V. Aritua, and E. Adipala. 2004. Within crop spread of sweet potato virus disease and the population dynamics of its whitefly and aphid ve ctors. Crop Protect. 2004: 109 116. Byrne, D. N. 1999. Migration and dispersal by the sweet potato whitefly, Bemisia tabaci Agric. Fore. Metero. 97: 309 316. Byrne, D. N. and T. S. Jr. Bellows. 1991. Whitefly Biology. Annu. Rev. Entomol. 36: 431 457. B yrne, D. N. and M. A. Houck. 1990. Morphometric identification of wing polymorphism in Bemisia tabaci (Homoptera: Aleyrodidae). Ann. Entomol. Soc. Am. 83: 487 493. Byrne, D. N. and P. K. von Bretzel. 1987. Similarity in flight activity rhythms in coexisti ng species of Aleyrodidae, Bemisia tabaci and Trialeurodes abutilonea Entomol. Exp. Appl. 43: 215 219. Byrne, D. N., R. J. Rathman, T. V. Orum, and J. C. Palumbo. 1996. Localized migration and dispersal by the sweet potato whitefly, Bemisia tabaci Oecol ogia. 105: 320 328. Byrne, F. J. and A. L. Devonshire. 1993. Insensitive acetylcholinesterase and esterase polymorphism insusceptible and resistance populations of the tobacco whitefly, Bemisia tabaci Pestic. Biochem. Physiol. 45: 34 42. Byrne, F. J., S J. Castle, N. Prabhaker, and L. C. Toscano. 2003. Biochemical study of resistance to imidacloprid in B biotype Bemisia tabaci from Guatemala. Pest Manage. Sci. 59: 347 352. Caciagli, P., D. Bosco, and L. Al Bitar. 1995. Relationships of the Sardinian is olate of tomato yellow leaf curl geminivirus with its whitefly vector Bemisia tabaci Gennadius. Eur. J. Plant Pathol. 101: 163 170. Caciagli, P., V. M. Piles, D. Marian, M. Vecchiati, V. Masenga, G. Mason, T. Falcioni, and E. Noris. 2009. Virion stability is important for the circulative transmission of Tomato yellow leaf curl Sardinia virus by Bemisia tabaci but virion access to the salivary glands does not guarantee transmissibility. J. Virol. 83: 5784 5795.

PAGE 173

173 Cahill, M. K. Gorman, S. Day, I. Denholm, A. Elbert, and R. Nauen. 1996. Baseline determination and detection of resistance to imidacloprid in Bemisia tabaci (Homoptera: Aleyrodidae). Bull. Entomol. Res. 86: 343 349. Calvitti, M. and P. C. Remotti. 1998. Host preference and performance of Bemisia argentifolii (Homoptera: Aleyrodidae) on weeds in Central Italy. Environ. Entomol. 27: 1350 1356. Candau, J. R. A. Fleming, and A. Hopkin. 1998. Spatiotemporal patterns of large scale defoliation caused by the spruce budworm in Ontario since 1941. Cana J. Fores. Res. 28: 1733 1741. Cardoza, Y. J. H. L. McAuslane, and S. E. Webb. 2000. Effect of leaf age and silverleaf symptoms on oviposition site selection and development of Bemisia argentifolii (Homoptera: Aleyrodidae) on zucchini. Environ. Entomol 29: 220 225. Carriere, Y., P.C. Ellsworth, P. Dutilleul, C. Ellers Kirk, V. Barkley, and L. Antilla. 2006. A GIS based approach for areawide pest management: the scales of Lygus hesperus movement to cotton from alfalfa, weeds, and cotton. Entomol. Exp. Appl. 118: 203 210. Castle, S. J. 2001. Differences between cotton and melon in host acceptance by Bemisia tabaci pp. 1056 1059, Proceedings of the Beltwide Cotton Conference. National Cotton Council, Memphis, TN. Castle, S. J., T. J. Henneberry, and N. C. Toscano. 1996. Suppression of Bemisia tabaci (Homoptera: Aleyrodidae) infestations in cantaloupe and cotton with sprinkler irrigation Crop Protect. 15: 657 663. Chellemi, D. O., K. G. Rohrbach, R. S. Yost, and R. M. Sonoda. 1988. Analysis of the spati al pattern of plant pathogens and diseased plants using geostatistics. Phytopathology. 78: 221 226. Chu, C. C., T. J. Henneberry, and A. Cohen. 1995. Bemisia argentifolii (Homoptera: Aleyrodidae): host preference and factors affecting oviposition and feed ing site preference. Environ. Entomol. 24: 354 360. Cigliano, M. M., W. P. Kemp, and T. M. Kalaris. 1995. Spatiotemporal characteristics of rangeland grasshopper (Orthoptera: Acrididae) regional outbreaks in Montana. J. Orthoptera Res. 4: 111 126. Cocco, A., A. Q. Cossu, P. Erre, G. Nieddu, and P. Luciano. 2010. Spatial analysis of gypsy moth populations in Sardinia using geostatistical and climate models. Agric. For. Entomol. 12: 417 426.

PAGE 174

174 Cohen, S. 1982. Control of whitefly vectors of viruses by color m ulches, pp. 45 56. In K. F. Harris and K. Maramorosch [eds.], Pathogens, Vectors, and Plant Diseases: Approaches to Control. Academic Press, New York. Cohen, S. and Y. Antignus. 1994. Tomato yellow leaf curl virus, a whitefly borne geminivirus of tomatoes pp. 259 288. In K. F. Harris [ed.], Advances in Disease Vector Research. Springer, New York. Cohen, S. and M. J. Berlinger. 1986. Transmission and cultural control of whitefly borne viruses. Agriculture, Ecosystems, & Environment. 17: 89 97. Cohen, J. and A. Gera. 1995. Lisianthus leaf curl a new diseases of lisianthus caused by tomato yellow leaf curl virus. Plant Dis. 79: 416 420. Cohen, S. and I. Harpaz. 1964. Periodic, rather than continual acquisition of new tomato virus by its vector, the tobacco whitefly ( Bemisia tabaci Gennadius). Entomol. Exp. Appl. 7: 155 166. Cohen, S. and F. E. Nitzany. 1966. Transmission and host range of the tomato yellow leaf curl virus. Phytopathology. 56: 1127 1131. Cohen, S., J. Kern, I. Harpaz, and R. Ben Joseph. 19 88. Epidemiological studies of the tomato yellow leaf curl virus (TYLCV) in the Jordan Valley, Israel. Phytoparasitica. 16: 259 270. Colvin, J., L. D. C. Fishpool, D. Fargette, J. Sherington, and C. Fargette. 1998. Bemisia tabaci (Hemiptera: Aleyrodidae) trap catches in a cassava field in Cote d'Ivoire in relation to environmental factors and the distribution of African cassava mosaic disease. Bull. Entomol. Res. 1998: 369 378. Costa, H. S. and J. K. Brown. 1991. Variation in biological characteristics a nd esterase patterns among populations of Bemisia tabaci (Genn.) and the association of one population with silverleaf symptom induction. Entomol. Exp. Appl. 61: 211 219. Costa, H. S. and K. L. Robb. 1999. Effects of ultraviolet absorbing greenhouse plast ic films on flight behavior of Bemisia argentifolii (Homoptera: Aleyrodidae) and Frankliniella occidentalis (Thysanoptera: Thripidae). J. Econ. Entomol. 92: 557 562. Costa, H. S., J. K. Brown, and D. N. Byrne. 1991. Host plant selection by the whitefly, B emisia tabaci (Gennadius) (Hom., Aleyrodidae) under greenhouse conditions. J. Appl. Entomol. 112: 146 152. Costa, H. S., D. E. Ullman, M. W. Johnson, and B. E. Tabashnik. 1994. Row covers effect on sweetpotato whitefly (Homoptera: Aleyrodidae) densities, incidence of silverleaf, and crop yield in zucchini. J. Econ. Entomol. 87: 1616 1621.

PAGE 175

175 Coudriet, D. L., N. Prabhaker, A. N. Kishara, and D. E. Meyerdirk. 1985. Variation in development rate on different hosts and overwintering of the sweet potato whitefly Bemisia tabaci (Homoptera: Aleyrodidae). Environ. Entomol. 14: 516 519. Coudriet, D. L., D. E. Meyerdirk, N. Prabhaker, and A. N. Kishaba. 1986. Bionomics of sweetpotato whitefly (Homoptera: Aleyrodidae) on weed hosts in the Imperial Valley, California. E nviron. Entomol. 15: 1179 1183. Csizinszky, A. A., D. J. Schuster, and J. B. Kring. 1995. Color mulches influence yield and insect pest populations in tomato. J. Amer. Soc. Hort. Sci. 120: 778 784. Csizinszky, A. A., D. J. Schuster, and J. B. Kring. 1997 Evaluation of color mulches and oil sprays for yield and for the control of silverleaf whitefly, Bemisia argentifolii (Bellows and Perring) on tomatoes. Crop Protect. 16: 475 481. Csizinszky, A. A., D. J. Schuster, and J. E. Polston. 1999. Effects of ul traviolet reflective mulches on tomato yields and on the silverleaf whitefly. HortScience. 34: 911 914. Czosnek, H. 2007. Interactions of Tomato Yellow Leaf Curl Virus with its whitefly vector, pp. 157 170. In H. Czosnek [ed.], Tomato Yellow Leaf Curl Vir us Disease. Springer. Czosnek, H. and H. Laterrot. 1997. A worldwide survey of tomato yellow leaf curl viruses. Arch. Virol. 142: 1391 1406. Czosnek, H., N. Navot, and H. Laterrot. 1990. Geographical distribution of tomato yellow leaf curl virus. A firs t survey using a specific DNA probe. Phytophathology Mediterranean. 29: 1 6. Czosnek, H., M. Ghanim, G. Rubinstein, S. Morin, V. Fridman, and M. Zeidan. 2001. Whiteflies: vectors or victims? of geminiviruses. Advances in Virus Research. 57: 291 322. Czosnek, H., M. Ghanim, and M. Ghanim. 2002. The circulative pathway of begomoviruses in the whitefly vector Bemisia tabaci -insights from studies with Tomato yellow leaf curl virus Ann. Appl. Biol. 140: 215 231. D'Hondt, M. and M. Russo. 1985. Tomato yellow leaf curl in Senegal. J. Phytopathol. 112: 153 160.

PAGE 176

176 Dallot, S., T. Gottwald, G. Labonne, and J. B. Quiot. 2003. Spatial pattern analysis of Sharka disease ( Plum pox virus Strain M) in peach orchards in southern France. Phytopathology. 93: 1543 155 2. Darnell, S. J., L. J. Meinke, L. J. Young, and C. Gotway. 1999. Geostatistical investigation of the small scale spatial variation of western corn rootworm (Coleoptera: Chrysomelidae) adults. Environ. Entomol. 28: 266 274. De'ath, G. and K. E. Fabriciu s. 2000. Classification and regression trees: A powerful yet simple technique for ecological data analysis. Ecology. 81: 3178 3192. De Barro, P. J. 1995. Bemisia tabaci biotype B: a review of its biology distribution and control, pp. 57. Technical Paper, Division of Entomology, SCIRO, Canberra, Australia. De Barro, P. J. and F. Driver. 1997. Use of RAPD PCR to distinguish the B biotype from other biotypes of B. tabaci (Gennadius) (Hemiptera: Aleyrodidae). Aust. J. Entomol. 36: 149 152. De Barro, P. J., F. Driver, J. W. H. Trueman, and J. Curran. 2000. Phylogenetic relationship of world populations of Bemisia tabaci (Gennadius) using ribosomal ITS1. Molecular Phylogeny and Evolution. 16: 29 36. Dean, D. E. and D. J. Schuster. 1995. Bemisia argentifolii (Homoptera: Aleyrodidae) and Macrosiphum euphorbiae (Homoptera: Aphididae) as prey for two species of Chrysopidae. Environ. Entomol. 24: 1562 1568. DeBano, S. J., P. B. Hamm, A. Jensen, S. I. Rondon, and P. J. Landolt. 2010. Spatial and temporal dynamics of potato tuberworm (Lepidoptera: Gelechiidae) in the Columbia Basin of the Pacific Northwest. Environ. Entomol. 39: 1 14. Denholm, I., M. Cahill, F. J. Byrne, and A. L. Devonshire. 1996. Progress with documenting and combating insecticide resistance in Bemisia pp. 577 603. In D. Gerling and R. T. Mayer [eds.], Bemisia : 1995 Taxonomy, Biology, Damage, Control and Management. Intercept Andover. Denholm, I., M. Cahill, T. J. Dennehy, and A. R. Horowitz. 1998. Challenges with managing insecticide resistanc e in agricultural pests, exemplified by the whitefly Bemisia tabaci Philosophical Transactions of the Royal Society. 353: 1757 1767. Dennehy, T. J., L. Williams III, J. S. Russell, X. Li, and M. Wigert. 1996. Monitoring and management of whitefly resist ance to insecticides in Arizona, pp. 743 748. In P. Dugger and D. Richter [eds.], Proceedings Beltwide Cotton Conference. National Cotton Council, Memphis, TN.

PAGE 177

177 Dinsdale, A., L. Cook, C. Riginos, Y. M. Buckley, and P De Barro. 2010. Refined global analysis of Bemisia tabaci (Hemiptera: Sternorrhyncha: Aleyrodoidea: Aleyrodidae) mitochondrial cytochrome oxidase 1 to identify species level genetic boundaries. Ann. Entomol. Soc. Am. 103: 196 208. Dittrich, V., S. Uk, and G. H. Ernst. 1990a. Chemical control a nd insecticide resistance of whiteflies, pp. 263 286. In D. Gerling [ed.], Whiteflies: their Bionomics, Pest Status and Management. Intercept, Andover, UK. Dittrich, V., G. H. Ernst, O. Ruesch, and S. Uk. 1990b. Resistance mechanisms in sweetpotato whitef ly (Homoptera: Aleyrodidae) populations from Sudan, Turkey, Guatemala, and Nicaragua. J. Econ. Entomol. 83: 1665 1670. Dowell, R. V. 1990. Integrating biological control of whiteflies into crop management systems. In D. Gerling [ed.], Whiteflies: their Bionomics, Pest Status and Management. Intercept, Andover, UK. Duehl, A. J., F. H. Koch, and F. P. Hain. 2011. Southern pine beetle regional outbreaks modeled on landscape, climate and infestation history. For. Ecol. Manage. 261: 473 479. Dutilleul, P. 1 993. Modifying the t test for assessing the correlation between two spatial processes. Biometrics. 49: 305 314. Ekbom, B. S. and X. Rumei. 1990. Sampling and spatial patterns of whiteflies., pp. 255 263. In D. Gerling [ed.], Whiteflies: their Bionomics, P est Status and Management. Intercept, Andover, UK. El Gendi, S. S., K. M. Adam, and M. A. Bachatly. 1997. Effect of the planting date of tomato on the population density of Bemisia tabaci (Genn.) and Heliothis armigera (HB), viral infection and yield. Ar ab. J. Agric. Sci. 5: 135 144. Elbert, A., H. Overbeck, H. Iwaya, and S. Tsuboi. 1990. Imidacloprid, a novel systemic nitromethylene analogue insecticide for crop protection, pp. 21 28, Proceedings of the Brighton Crop Protection Conference, Pest and Dise ases. British Crop Protection Council, Farnham, UK. Ellsbury, M. M., W. D. Woodson, S. A. Clay, D. Malo, J. Schumacher, D. E. Clay, and C. G. Carlson. 1998. Geostatistical characterization of the spatial distribution of adult corn rootworm (Coleoptera : C hrysomelidae) emergence. Environ. Entomol. 27: 910 917. Ellsworth, P. C. and J. L. Martinez Carrillo. 2001. IPM for Bemisia tabaci : a case study from North America. Crop Protect. 20: 853 869.

PAGE 178

178 Ellsworth, P. C. and D. L. Meade. 1994. Action thresholds for whiteflies in Arizona, pp. 878 881. In D. J. Herber and D. A. Richter [eds.], Proceedings Beltwide Cotton Conferences. National Cotton Council, Memphis, TN. Ellsworth, P. C., L. Moore, T. F. Watson, and T. J. Dennehy. 1994. Insect pest management for cott on. The University of Arizona, Cooperative Extension. 27 pp Ellsworth, P.C., R. Tronstad, J. Leser, P. B. Goodell, L. D. Godfrey, T. J. Henneberry, D. Hendrix, D. Brushwood, S.E. Naranjo, S. J. Castle, and R. L. Nichols. 1999. Sticky Cotton Sources & Solutions. IPM Series No. 13 .The University of Arizona Cooperative Extension.Tucson, AZ. http://ag.arizona.edu/crops/cotton/insects/wf/stickycss.pdf.4/16/2007 ESRI. 2006. ArcView ArcInfo. Version 9.2. Environmental Systems Research Institute. Redlands, CA. Evans, G. A. 2007. Host plant list of the whiteflies (Aleyrodidae) of the world. http://www.sel.barc.usda.gov:8080/1WF/WhiteflyHost.pdf January 18th 2011. Fargette, D. and C. Fauquet. 1988. A preliminary study on the influence of intercropping maize and cassava on the spread of African cassava mosaic virus by whiteflies. Aspects Appl. Biol. 17: 195 202. Fargette, D., C. Fauquet, and J. C. Thouvenel 1985. Field studies on the spread of African cassava mosaic. Ann. Appl. Biol. 106: 285 294. Fargette, D., C. Fauquet, E. Grenier, and J. M. Thresh. 1990. The spread of African cassava mosaic virus into the within cassava fields. J. Phytopathol. 130: 28 9 302. Faria, M. and S. P. Wraight. 2001. Biological control of Bemisia tabaci with fungi. Crop Protect. 20: 767 778. Farias Larios, J., M. Orozco Santos, and S. Guzman Gonzalez. 1995. Yield of three cultivars of muskmelon growth on transparent mulch and floating rowcover in a tropical region. HortScience. 30: 890 891. Farias Larios, J., M. Orozco Santos, S. Guzman Gonzalez, and J. Perez. 1996. Effect of plastic mulch, floating rowcovers and microtunnels on insect populations and yield of muskmelon. Hort Science. 31: 677. Farias, P. R. S., S. R. Roberto, J. R. S. Lopes, and D. Perecin. 2004. Geostatistical characterization of the spatial distribution of Xylella fastidiosa sharpshooter vectors on citrus. Neotrop. Entomol. 33: 13 20.

PAGE 179

179 Farias, P. R. S., J. C Barbosa, A. C. Busoli, W. L. Overal, V. S. Miranda, and S. M. Ribeiro. 2008. Spatial analysis of the distribution of Spodoptera frugiperda (J. E. Smith) (Lepidoptera: Noctuidae) and losses in maize crop productivity using geostatistics. Neotrop. Entomol. 37: 321 327. Fenigstein, A., M. Eliyahu, S. Gan Mor, and D. Veierov. 2001. Effects of five vegetable oils on the sweetpotato whitefly Bemisia tabaci Phytoparasitica. 29: 197 203. Ferguson, A. W., Z. Klukowski, B. Walczak, J. N. Perry, M. A. Mugglestone S. J. Clark, and I. H. Williams. 2000. The spatio temporal distribution of adult Ceutorhynchus assimilis in a crop of winter oilseed rape in relation to the distribution of their larvae and that of the parasitoid Trichomalus perfectus Entomol. Exp. Appl 95: 161 171. Fitzgerald, J. W., R. N. Coulson, P. E. Pulley, R. O. Flamm, F. L. Oliveria, K. M. Swain, and D. B. Drummond. 1994. Suppression tactics for Dendroctonus frontalis Zimmerman (Coleoptera: Scolytidae): An examination of the occurrence of infe stations adjacent to the treatment sites. J. Econ. Entomol. 87: 417 425. Fleischer, S. J., P. E. Blom, and R. Weisz. 1999. Sampling in precision IPM: When the objective is a map. Phytopathology. 89: 1112 1118. Flint, H. M., F. D. Wilson, D. Hendrix, J. E Leggett, S. E. Naranjo, T. J. Henneberry, and J. W. Radin. 1994. The effect of plant water stress on beneficial and pest insects including the pink bollworm and the sweetpotato whitefly in two short season cultivars of cotton. Southwest. Entomol. 19: 11 22. Flint, H. M., J. W. Radin, N. J. Parks, and L. L. Reaves. 1995. The effects of drip or furrow irrigation of cotton on Bemisia argentifolii (Homoptera: Aleyrodidae). J. Agric. Ento. 12: 25 32. Flint, H. M., S. E. Naranjo, J. E. Leggett, and T. J. Henn eberry. 1996. Cotton water stress, arthropod dynamics, and management of Bemisia tabaci (Homoptera: Aleyrodidae). J. Econ. Entomol. 89: 1288 1300. Fluckiger, C. R., H. Kristinsson, R. Senn, A. Rindlisbacher, H. Buholzer, and G. Voss. 1992. A novel agent to control aphids and whiteflies, Brighton Crop Protection Conference: Pests and Diseases. The British Crop Protection Council, Farnham, UK. Font, M. I., C. Cordoba, A. Garcia, R. Santiago, and C. Jorda. 2005. First report of tobacco as a natural host of Tomato yellow leaf curl virus in Spain. Plant Dis. 89: 910.

PAGE 180

180 Friedmann, M., M. Lapidot, S. Cohen, and M. Pilowsky. 1998. A novel source of resistance to tomato yellow leaf curl virus exhibiting a symptomless reaction to viral infection. J. Amer. Soc. Hort. Sci. 123: 1004 1007. Frohlich, D. R., I. Torres Jerez, I. D. Bedford, P. G. Markham, and J. K. Brown. 1999. A phylogeographical analysis of the Bemisia tabaci species complex based on mitochondrial DNA markers. Mol. Ecol. 8: 1683 1691. Gachoka, K. K., D Obeng Ofori, and E. Y. Danquah. 2005. Host suitability of two Ghanaian biotypes of Bemisia tabaci (Homoptera: Aleyrodidae) on five common tropical weeds. Int. J. Trop. Insect Sci. 25: 236 244. Gamez Jimenez, C., J. L. Romero Romero, M. E. Santos Cervant es, N. E. Leyva Lopez, and Mendez Lozano. 2009. Tomatillo ( Physalis ixocarpa ) as a natural new host for Tomato yellow leaf curl virus in Sinaloa, Mexico. Plant Dis. 93: 545. Garcia Andres, S., F. Monci, J. Navas Castillo, and E. Moriones. 2006. Begomoviru s genetic diversity in the native plant reservoir Solanum nigrum : Evidence for the presence of a new virus species of recombinant nature. Virology. 350: 433 442. Garcia, F. J. 2006. Analysis of the spatio temporal distribution of Heliocoverpa armigera Hb in a tomato field using a stochastic approach. Biosys. Eng. 93: 253 259. Ge, Shaw Kui, R. I. Carruthers, Zu Fei Ma, Guang Xue Zhang, and Dian Mo Li. 2005. Spatial heterogeneity and population risk analysis of cotton bollworm, Helicoverpa armigera in Ch ina. Insect Sci. 12: 255 262. Gennadius, P. 1889. Disease of tobacco plantations in the Trikonia. The aleyrodid of tobacco. Ellenike Georgia. 5: 1 3. Gerling, D. 1967. Bionomics of the whitefly parasite complex associated with cotton in southern Californ ia (Homoptera: Aleyrodidae; Hymenoptera: Aphelinidae). Ann. Entomol. Soc. Am. 60: 1306 1321. Gerling, D. 1984. The overwintering mode of Bemisia tabaci in Israel. Phytoparasitica. 12: 109 118. Gerling, D. 1986. Natural enemies of Bemisia tabaci biologic al characteristics and potential as biological control agents: a review. Agric., Ecosyst. Environ. 17: 99 110. Gerling, D. and A. R. Horowitz. 1984. Yellow traps for evaluating the population levels and dispersal patterns of Bemisia tabaci (Gennadius) (Ho moptera: Aleyrodidae). Ann. Entomol. Soc. Am. 77: 753 759.

PAGE 181

181 Gerling, D. and V. Kravechenko. 1996. Pest Management of Bemisia Out of Doors. In D. Gerling and R. T. Mayer [eds.], Bemisia : 1995 Taxonomy, Biology, Damage, Control and Management. Intercept Andov er, UK. Gerling, D. and R. T. Mayer [eds.]. 1996. Bemisia : 1995 Taxonomy, Biology, Damage, Control and Management. Intercept, Andover, UK Gerling, D., A. R. Horowitz, and J. Baumgaertner. 1986. Autecology of Bemisia tabaci Agric. Eco. Environ. 17: 5 19. Gerling, D., O. Alomar, and J. Arno. 2001. Biological control of Bemisia tabaci using predators and parasitoids. Crop Protect. 20: 779 799. Ghanim, M. and V. Medina. 2007. Localization of tomato yellow leaf curl virus in its whitefly vector Bemisia taba ci pp. 171 183. In H. Czosnek [ed.], Tomato Yellow Leaf Curl Virus Disease. Springer, Dordrecht. Ghanim, M., S. Morin, M. Zeidan, and H. Czosnek. 1998. Evidence for transovarial transmission of tomato yellow leaf curl virus by its vector, the whitefly Bem isia tabaci Virology. 240: 295 303. Ghanim, M., S. Morin, and H. Czosnek. 2001. Rate of Tomato yellow leaf curl virus translocation in the circulative transmission pathway of its vector, the whitefly Bemisia tabaci Phytopathology. 91: 188 196. Ghanim, M., I. Sobol, M. Ghanim, and H. Czosnek. 2007. Horizontal transmission of begomoviruses between Bemisia tabaci biotypes. Arthropod Plant Interactions. 1: 195 204. Gilbertson, R. L., M. R. Rojas, T. Kon, and J. Jaquez. 2007. Introduction of Tomato yellow l eaf curl virus into the Dominican Republic: The development of a successful integrated pest management strategy, pp. 279 303. In H. Czosnek [ed.], Tomato Yellow Leaf Curl Virus Disease. Springer, Dordrecht. Gill, R. J. 1992. A review of the sweetpotato wh itefly in southern California. Pan Pac. Entomol. 68: 144 152. Goldman, V. and H. Czosnek. 2002. Whiteflies ( Bemisia tabaci ) issued from eggs bombarded with infectious DNA clones of Tomato yellow leaf curl virus from Israel (TYLCV) are able to infect tomat o plants. Arch. Virol. 147: 787 801. Goovaerts, P. 1997. Geostatistics for Natural Resource Evaluation. Oxford University Press, New York.

PAGE 182

182 Gottwald, T. R., M. Cambra, P. Moreno, E. Camarasa, and J. Piquer. 1996. Spatial and temporal analyses of citrus tristeza virus in eastern Spain. Phytopathology. 86: 45 55. Greathead, A. H. 1986. Host plants, pp. 17 26. In M. J. W. Cock [ed.], Bemisia tabaci A Literature survey on the cotton whitefly with an annotated bibliography. CAB International Institutes, Bi ological Control, Silwood Park. Gribko, L. S., A. M. Liebhold, and M. E. Hohn. 1995. Model to predict Gypsy moth (Lepidoptera, Lymantriidae) defoliation using Kriging and logistic regression. Environ. Entomol. 24: 529 537. Gronenborn, B. 2007. The tomato yellow leaf curl virus genome and function of its proteins, pp. 67 84. In H. Czosnek [ed.], Tomato Yellow Leaf Curl Virus Disease Springer, Dordrecht. Groves, R. L., J. Chen, E. L. Civerolo, M. W. Freeman, and M. A. Viveros. 2005. Spatial analysis of al mond leaf scorch disease in the San Joaquin Valley of California: Factors affecting pathogen distribution and spread. Plant Dis. 89: 581 589. Gruenhagen, N. M. and T. M. Perring. 2001. Plant influences on silverleaf whitefly oviposition and development an d the potential for enemy free space. Entomol. Exp. Appl. 99: 387 391. Grunwald, S., S. H. Daroub, and T. A. Lang. 2009. Tree based modeling of complex interactions of phosphorus loadings and environmental factors. Sci. Total Environ. 407: 3772 3783. Hag ler, J. R. and S. E. Naranjo. 1994. Determining the frequency of heteropteran predation on sweetpotato whitefly and pink bollworm using multiple ELISAs. Entomol. Exp. Appl. 72: 59 66. Hanif Khan, S., J. K. Brecht, C. A. Powell, and P. J. Stofella. 1999 Ethylene levels and fruit quality of silverleaf whitefly infested dwarf cherry tomatoes, pp. 134 138, Proceedings Florida State Horticulture Society. Harrawijn, P. and H. Kayser. 1997. Pymetrozine, a fast acting and selective inhibitor of aphid feeding: in situ studies with electronic feeding monitoring of feeding behavior. Pestic. Sci. 49: 130 140. Harrison, B. D. 1985. Advances in geminivirus research. Annu. Rev. Phytopathol. 23: 55 82.

PAGE 183

183 Hebertson, E. G. and M. J. Jenkins. 2008. Climate factors associ ated with historic spruce beetle (Coleoptera: Cuculionidae) outbreaks in Utah and Colorado. Environ. Entomol. 37: 281 292. Henneberry, T. J., D. H. Hendrix, H. H. Perkins, S. E. Naranjo, H. M. Flint, D. Akey, L. F. Jech, and R. A. Burke. 1995. Bemisia ar gentifolii (Homoptera: Aleyrodidae) populations and relationships to sticky cotton and cotton yields. Southwest. Entomol. 20: 255 271. Hernandez, A. and J. J. Pacheco. 1998. Chancing a planting date: a silverleaf whitely case, pp. 574 577. In P. Dugger an d D. Richter [eds.], Proceedings Beltwide Cotton Conferences. National Cotton Council, Memphis, TN. Hilje, L., H. S. Costa, and P. A. Stansly. 2001. Cultural practices for managing Bemisia tabaci and associated viral diseases. Crop Protect. 20: 801 812. Hoelmer, K. A. 1996. Whitefly parasitoids: Can they control field populations of Bemisia ? In D. Gerling and R. T. Mayer [eds.], Bemisia 1995: Taxonomy, Biology, Damage, Control and Management. Intercept, Andover, UK. Holland, J. M., C. F. G. Thomas, T. Bi rkett, S. Southway, and H. Oaten. 2005. Farm scale spatiotemporal dynamics of predatory beetles in arable crops. J. Appl. Ecol. 42: 1140 1152. Holmstrom, K. E., M. G. Hughes, S. D. Walker, W. L. Kline, and J. Ingerson Mahar. 2001. Spatial mapping of adult corn earworm and European corn borer populations in New Jersey. HortTechnology. 11: 103 109. Holt, J., M. J. Jeger, J. M. Thresh, and G. W. Otim Nape. 1997. An epidemiological model incorporating vector population dynamic applied to African Cassava Mosai c Virus Disease. J. Appl. Ecol. 34: 793 806. Hooks, C. R. R. J. R. Valenzuela, and J. Defrank. 1998. Incidence of pests and arthropod natural enemies in zucchini grown with living mulches. Agriculture Ecosystems & Environment. 69: 217 231. Horowitz, A. R. and I. Ishaaya. 1994. Managing resistance to insect growth regulators in the sweetpotato whitefly (Homoptera: Aleyrodidae). J. Econ. Entomol. 87: 866 879. Horowitz, A. R. and I. Ishaaya. 1996. Chemical control of Bemisia management and application, pp. 537 556. In D. Gerling and R. T. Mayer [eds.], Bemisia : 1995 Taxonomy, Biology, Damage, Control and Management Intercept, Andover.

PAGE 184

184 Horowitz, A. R., S. Kontsedalov, and I. Ishaaya. 2004. Dynamics of resistance to the neonicotinoids acetamiprid and thia methoxam in Bemisia tabaci (Homoptera: Aleyrodidae). J. Econ. Entomol. 97: 2051 2056. Horowitz, A. R., I. Denholm, and S. Morin. 2007. Resistance to insecticides in the TYLCV vector, Bemisia tabaci pp. 305 325. In H. Czosnek [ed.], Tomato Yellow Leaf Cur l Virus Disease. Springer, Dordrecht. Husain, M. and K. N. Trehan. 1933. Observations on the life history, bionomics and control of the white fly of cotton ( Bemisia gossypiperda M. EL.). Indian J. Agric. Sci. 3: 701 753. Husain, M. and K. N. Trehan. 1940 Final report on the scheme of investigations on the whitefly on cotton in the Punjab. Indian J. Agric. Sci. 10: 101 109. Ifoulis, A. A. and M. Savopoulou Soultani. 2006. Use of geostatistical analysis to characterize the spatial distribution of Lobesia botrana (Lepidoptera : Tortricidae) larvae in northern Greece. Environ. Entomol. 35: 497 506. Ioannou, N. 1985. Yellow leaf curl and other diseases of tomato in Cyprus. Plant Pathol. 345: 428 434. Ioannou, N. 1987. Cultural management of tomato yellow le af curl disease in Cyprus. Plant Pathol. 36: 367 373. Isaacs, R. and D. N. Byrne. 1998. Aerial distribution, flight behaviour and eggload: their inter relationship during dispersal by the sweetpotato whitefly. J. Agric. Ento. 67: 741 750. Isaacs, R., M. A. Willis, and D. N. Byrne. 1999. Modulation of whitefly take off and flight orientation by wind speed and visual cues. Physiol. Entomol. 24: 311 318. Isaaks, E. H. and R. M. Srivastava. 1989. An Introduction to Applied Geostatistics. Oxford University Pr ess, New York. Ishaaya, I., Z. Mendelson, K. R. Simon Ascher, and J. E. Casida. 1987. Cypermethrin synergism by pyrethroid esterase inhibitors in adults of the whitefly Bemisia tabaci Pestic. Biochem. Physiol. 28: 155 162. Ishaaya, I., Z. Mendelson, and A. R. Horowitz. 1993. Toxicity and growth suppression exerted by diafenthiuron in the sweetpotato whitefly Bemisia tabaci Phytoparasitica. 21: 199 204. Ishaaya, I., A. Barazani, S. Kontsedalov, and A. R. Horowitz. 2007. Insecticides with novel modes of action: Mechanism, selectivity and cross resistance. Entomological Research. 37: 148 152.

PAGE 185

185 Iwao, S. 1968. A new regression method for analyzing the aggregation pattern of animal populations. Res. Popul. Ecol. 10: 1 20. Jiang, Y. X., H. Lei, J. L. Collar, B. Martin, M. Muniz, and A. Fereres. 1999. Probing and feeding behavior of two distinct biotypes of Bemisia tabaci (Homoptera: Aleyrodidae) on tomato plants. J. Econ. Entomol. 92: 357 366. Jiang, Y. X., C. De Blas, L. Barrios, and A. Fereres. 2000. Corre lation between whitefly (Homoptera: Aleyrodidae) feeding behavior and transmission of tomato yellow leaf curl virus. Ann. Entomol. Soc. Am. 93: 573 579. Jindal, V., G. S. Dhaliwal, and A. K. Dhawan. 2008. Mechanisms of resistance in cotton to whitefly Bem isia tabaci (Homoptera: Aleyrodidae): antibiosis. Int. J. Trop. Insect Sci. 27: 216 222. Jing, Y., J. Huang, Rui yan Ma, and Ju cai Han. 2003. Host plant preferences of Bemisia tabaci Gennadius. Entomologia Sinica. 10: 109 114. Johnson, C. G. 1969. Migra tion & Dispersal of Insects by Flight. Methuen & Company Limited, London. Jones, W. A. and G. L. Snodgrass. 1998. Development and fecundity of Deraeocoris nebulosus (Heteroptera: Miridae) on Bemisia argentifolii (Homoptera: Aleyrodidae). Flor. Ento. 81: 3 45 350. Kadir, H. A. and C. O. Knowles. 1991. Toxicological studies of the thiourea diafenthiuron in diamondback moth (Lepidoptera: Yponomeutidae), two spotted spider mites (Acari: Tetranychidae) and bulb mite (Acari: Acaridae). J. Econ. Entomol. 84: 780 784. Kayser, H., L. Kaufmann, F. Schurmann, and P. Harrewijn. 1994. Pymetrozine (GGA 215'944): a novel compound for aphid and whitefly control: an overview of its mode of action, pp. 737 742, Brighton Crop Protection Conference: Pests and Diseases. Briti sh Crop Protection Council, Farnham, UK. Kemp, W. P., T. M. Kalaris, and W. F. Quimby. 1989. Rangeland grasshopper (Orthoptera: Acrididae) spatial variability: macroscale population assessment. J. Econ. Entomol. 82: 1270 1276. Kemp, W. P., K. M. O'Neill, M. M. Cigliano, and S. Torrusio. 2002. Field scale variations in plant and grasshopper communities: A GIS based assessment. Transactions in GIS. 6: 115 133.

PAGE 186

186 Kennedy, J. S. 1985. Migration, behavioral and ecological, pp. 7 26. In M. A. Rankin [ed.], Migr ation: Mechanisms and Adaptive Significance. Contributions in Marine Science Supplement. Kennedy, M. 2000. Understanding Map Projections.Environmental Systems Research Institute, Inc.Redlands. http://www.duc.auburn.edu/academic/classes/fory/7470/lab08/unde rstanding%20map%20projections.pdf.Feb 14, 2011. Kim, H., S .T. Kim, and M. P. Jung. 2007. Spatio temporal dynamics of Scotinophara lurida (Hemiptera: Pentatomidae) in rice fields. Ecol. Res. 22: 204 213. Kim, J K., J J. Park, H. Park, and K. Cho. 2001. Unbiased estimation of greenhouse whitefly, Trialeurodes vaporariorum mean density using yellow sticky trap in cherry tomato greenhouses Entomol. Exp. Appl. 100: 235 243. Klowden, M. J. 2002. Physiological Systems in Insects. Academic Press, San Diego, California. Korie, S., J. N. Perry, M. A. Mugglestone, S. J. Clark, C. F. G. Thomas, and M. N. M. Roff. 2000. Spatiotemporal association s in beetle and virus count data. J. Agric. Biol. Envion. Stat. 5: 214 239. Kravchenko, A. N. 2003. Influence of spatial structure on accuracy of interpolation methods. Soil Sci. Soc. Am. J. 67: 1564 1571. Kunik, T., R. Salomon, D. Zamir, N. Navot, M. Ze idan, I. Michelson, Y. Gafni, and H. Czosnek. 1994. Transgenic tomato plants expressing the tomato yellow leaf curl capsid protein are resistant to the virus. Biotechnology. 12: 500 504. Kunik, T., K. Palanaichelvam, H. Czosnek, V. Citovsky, and Y. Gafni. 1998. Nuclear import of the capsid protein of tomato yellow leaf curl virus (TYLCV) in plant and insect cells. The Plant Journal. 13: 393 399. LABINS. 2011. FDEP, Bureau of Survey and Mapping. http://d ata.labins.org/2003/index.cfm Lapidot, M. and M. Friedmann. 2002. Breeding for resistance to whitefly transmitted geminiviruses Ann. Appl. Biol. 140: 109 127. Lapidot, M., O. Goldray, R. Ben Joseph, S. Cohen, M. Friedmann, S. Nahon, L. Chen, and M. Pi lowsky. 2000. Breeding tomatoes for resistance to tomato yellow leaf curl begomovirus. Bull. OEPP. 30: 317 321.

PAGE 187

187 Lapidot, M., M. Friedmann, M. Pilowsky, R. Ben Joseph, and S. Cohen. 2001. Effect of host plant resistance to Tomato yellow leaf curl virus (TY LCV) on virus acquisition and transmission by its whitefly vector. Phytopathology. 91: 1209 1213. Lapidot, M., R. Ben Joseph, L. Cohen, Z. Machbash, and D. Levy. 2006. Development of a scale for evaluation of Tomato yellow leaf curl resistance level in to mato plants. Phytopathology. 96: 1404 1408. Larew, H. G. and J. C. Locke. 1990. Repellency and toxicity of a horticultural oil against whiteflies on Chrysanthemum. HortScience. 25: 1406 1407. Laufs, J., W. Traut, F. Heyraud, V. Matzeit, S. G. Rodgers, J. Schell, and B. Gronenborn. 1995. In vitro cleavage and joining at the viral origin of replication by the replication initiator protein of tomato yellow leaf curl virus. Proceedings of the National Academy of Sciences. 92: 3879 3883. Lecoustre, R., D. Far gette, C. Fauquet, and P. de Reffye. 1989. Analysis and mapping of the spatial spread of African Cassava Mosaic Virus using geostatistics and the kriging technique. Phytopathology. 79: 913 920. Legaspi Jr., B. C., J. C. Allen, C. C. Brewster, J. A. Morale s Ramos, and E. G. King. 1998. Areawide management of the cotton boll weevil: use of a spatio temporal model in augmentative biological control. Ecol. Model. 110: 151 164. Leggett, J. E. 1993. Comparison of arthropods sampled from cultivars of upland and pima cotton with drip and furrow irrigation. Southwest. Entomol. 18: 37 43. Levy, D. and M. Lapidot. 2008. Effect of plant age at inoculation on expression of genetic resistance to tomato yellow leaf curl virus. Arch. Virol. 153: 171 179. Liebhold, A. M. Z. Xu, M. E. Hohn, J. S. Elkinton, M. Ticehurst, G. L. Benzon, and R. W. Campbell. 1991. Geostatistical analysis of Gypsy moth (Lepidoptera, Lymantriidae) egg mass populations. Environ. Entomol. 20: 1407 1417. Liebhold, A. M., R. E. Rossi, and W. P. Kem p. 1993. Geostatistics and geographic information systems in applied insect ecology. Annu. Rev. Entomol. 38: 303 327. Liebhold, A., E. Luzander, R. Reardon, A. Bullard, A. Roberts, W. Ravlin, S. Delost, and B. Spears. 1996. Use of a geographic informatio n system to evaluate regional treatment effects in a gypsy moth (Lepidoptera: Lymantriidae) management program. J. Econ. Entomol. 89: 1192 1203.

PAGE 188

188 Liquido, N. J. and M. E. Irwin. 1986. Longevity, fecundity, change in degree of gravidity, and lipid content with adult age, and lipid utilization during tethered flight of the corn leaf aphid, Rhopalosiphum maidis (Fitch). Ann. Appl. Biol. 108: 449 459. Liu, T. X. and P.A. Stansly. 1995a. Toxicity and repellency of some biorational insecticides to Bemisia argen tifolii on tomato leaves. Entomol. Exp. Appl. 74: 137 143. Liu, T. X. and P.A. Stansly. 1995b. Oviposition by Bemisia argentifolii (Homoptera: Aleyrodidae) on tomato: effects of leaf factors and insecticide residues. J. Econ. Entomol. 88: 992 997. Liu, T X. and P. A. Stansly. 1998. Life history of Bemisia argentifolii (Homoptera: Aleyrodidae) on Hibiscus rosa sinensis (Malvaceae). Flor. Ento. 81: 437 445. Liu, T. X. and P.A. Stansly. 2000. Insecticidal activity of surfactants and oils against silverleaf whitefly ( Bemisia argentifolii ) nymphs (Homoptera: Aleyrodidae) on collards and tomato. Pest Manage. Sci. 56: 861 866. Liu, T. X., R. D. Oetting, and G. D. Buntin. 1993a. Distribution of Trialeurodes vaporariorum and Bemisia tabaci (Homoptera, Aleyrodida e) on some greehouse grown ornamental plants. J. Entomol. Sci. 28: 102 112. Liu, T. X., R. D. Oetting, and G. D. Buntin. 1993b. Population dynamics and distribution of Trialeurodes vaporariorum and Bemisia tabaci (Homoptera: Aleyrodidae) on Poinsettia fo llowing applications of three chemical insecticides. J. Entomol. Sci. 28: 126 135. Lopez Avila, A. 1986. Taxonomy and biology, pp. 3 11. In M. J. W. Cock [ed.], Bemisia tabaci A Literature Survey on the cotton whitefly with an annotated bibliography. CA B International Institutes, Biological Control, Silwood Park. Lyons, D. B., C. J. Sanders, and G. C. Jones. 2002. The use of geostatistics and GIS as tools for analyzing pheromone trap data at a landscape level: an update. IOBC WPRS Bulletin. 25: 1 14. MacQuarrie, C. J. K., J. R. Spence, and D. W. Langor. 2010. Using classification tree analysis to reveal causes of mortality in an insect population. Agric. For. Entomol. 12: 143 149. Makkouk, K. M. 1978. A study on tomato viruses in the Jordan Valley wit h special emphasis on Tomato yellow leaf curl. Plant Dis. 62: 259 268.

PAGE 189

189 Mansaray, A. and A. J. Sundufu. 2009. Oviposition, development and survivorship of the sweetpotato whitefly Bemisia tabaci on soybean, Glycine max and the garden bean, Paseolus vulga ris J. Insect Sci. 9: 1 6. Mansour, A. and A. Al Musa. 1992. Tomato yellow leaf curl virus: host range and virus vector relationships. Plant Pathol. 41: 122 125. Markham, P. G., I. D. Bedford, S. Liu, D. F. Frolich, R. Rosell, and J. K. Brown. 1996. The transmission of geminiviruses by biotypes of Bemisia tabaci (Gennadius), pp. 69 75. In D. Gerling and T. Meyer [eds.], Bemisia 1995: Taxonomy, Biology, Damage, Control, and Management. Intercept, Andover, U.K. Martin, J. H., D. Mifsud, and C. Rapisar da. 2000. The whiteflies (Hemiptera: Aleyrodidae) of Europe and the Mediterranean basin. Bull. Entomol. Res. 90: 407 448. Martinez Zubiaur, Y., D. Fonseca, M. Quinones, and I. Palenzuela. 2004. Presence of Tomato yellow leaf curl virus infecting squash ( Curcubita pepo ) in Cuba. Plant Dis. 88: 572. Mason, G., M. Rancati, and D. Bosco. 2000. The effect of thiamethoxam, a second generation neonicotinoid insecticide, in preventing transmission of tomato yellow leaf curl geminivirus (TYLCV) by the whitefly Be misia tabaci (Gennadius). Crop Protect. 19: 473 479. Matsuura, S. and S. Hoshino. 2008. Comparative spatial dispersal of Tomato yellow leaf curl virus vectored by B and Q biotypes of Bemisia tabaci in tomato glasshouses. Phytoparasitica. 36: 42 51. Mayn ard, D. N. and D. J. Cantliffe. 1989. Squash silverleaf and tomato irregular ripening: new vegetable disorders in Florida VC 37: 4. McAuslane, H. L. 1996. Influence of leaf pubescence on ovipositional preference of Bemisia argentifolii (Homoptera: Aleyro didae) on soybean. Environ. Entomol. 25: 834 841. McCollum, T. G., P. J. Stofella, C. A. Powell, D. J. Cantliffe, and S. Hanif Khan. 2004. Effects of silverleaf whitefly feeding on tomato fruit ripening. Postharvest Biol. Technol. 31: 183 190. McKenzie, C. L., P. K. Anderson, and N. Villarreal. 2004. An extensive survey of Bemisia tabaci (Homoptera: Aleyrodidae) in agricultural ecosystems in Florida. Fla. Entomol. 87: 403 407.

PAGE 190

190 Mehta, P. A. J. W. Wyman, M. K. Nakhla, and D. P. Maxwell. 1994. Transmissio n of tomato yellow leaf curl geminivirus by Bemisia tabaci (Homoptera: Aleyrodidae). J. Econ. Entomol. 87: 1291 1297. Mewally, S. A. G. 1999. Effect of planting date and certain weather factors on the population fluctuations of three insect pests infestin g kidney beans in Qualyobia governorate. Egyptian Journal of Agricultural Research. 77: 139 149. Michelson, I., D. Zamir, and H. Czosnek. 1994. Accumulation and translocation of tomato yellow leaf curl virus (TYLCV) in a Lycopersicon esculentum breeding l ine containing the L. chilense TYLCV tolerance gene Ty 1 Phytopathology. 84: 928 933. Midgarden, D. G., R. R. Youngman, and S. J. Fleischer. 1993. Spatial analysis of counts of western corn rootworm (Coleoptera: Chrysomelidae) adults on yellow sticky tra ps in corn: geostatistics and dispersion indices. Environ. Entomol. 22: 1124 1133. Miller, G. L., S. Nakahara, R. W. Carlson, D. R. Miller, and M. B. Stoetzel. 2001. Systematic Entomology Laboratory Whitefly Web Page. http://www.sel.barc.usda.gov/whitefly/wfframe.htm Feb 1st, 2011. Mohamed, E. F. 2010. Interaction between some viruses which attack tomato ( Lycopersicon esculentum Mill.) plants and their effect on growth and yield of tomato plants. J. Amer. Sci. 6: 311 320. Momol, M. T., G. W. Simone, W. Dankers, R. K. Sprenkel, S. M. Olson, E. A. Momol, J. E. Polston, and E. Heibert. 1999. First report of tomato yellow leaf curl virus in tomato in Georgia. Plant Dis. 83: 487. Momol, T ., S. M. Olson, J. Funderburk, and R. K. Sprenkel. 2001. Management of Tomato Yellow Leaf Curl Virus (TYLCV) in tomato in North Florida. http://edis.ifas.ufl.edu/NFREC1 8/1/2007, Mor, U. 1987. Bemisia ta baci and cotton physiology: a 5 year summary of the influence of water stressed plants on the pest population. Phytoparasitica. 15: 261. Morilla, G., D. Janssen, S. Garcia Andres, E. Moriones, M. Cuadrado, and E. R. Bejarano. 2005. Pepper ( Capsicum annuu m ) is a dead end host for Tomato yellow leaf curl virus Phytopathology. 95: 1089 1097. Morin, S., M. Ghanim, M. Zeidan, H. Czosnek, M. Verbeek, and F. J. M. van den Heuvelt. 1999. A GroEL homologue from endosymbiotic bacteria of the whitefly Bemisia tab aci in implicated in the circulative transmission of tomato yellow leaf curl virus. Virology. 256: 75 84.

PAGE 191

191 Moriones, E. and J. Navas Castillo. 2000. Tomato yellow leaf curl virus, an emerging virus complex causing epidemics worldwide. Virus Res. 71: 123 13 4. Moriones, E. J. Arno, G. P. Accotto, E. Noris, and Cavallarin. 1993. First report of Tomato yellow leaf curl virus in Spain. Plant Dis. 77: 953. Mound, L. A. 1965. Effects of leaf hair on cotton whitefly population in the Sudan Gezira. Empire Cotton G rowing Review. 42: 33 40. Mound, L. A. and S. H. Halsey. 1978. Whitefly of the world: A systematic catalogue of the Aleyrodidae (Homoptera) with host plant and natural enemy data. Wiley, New York. Muniyappa, V., H. M. Venkatesh, H. K. Ramappa, R. S. Kul karni, M. Zeidan, C Y Tarba, M. Ghanim, and H. Czosnek. 2000. Tomato leaf curl virus from Bangalore (ToLCV Ban4): sequence comparison with Indian ToLCV isolates, detection in plants and insects, and vector relationships. Arch. Virol. 145: 1583 1598. Muniz M. 2000. Host suitability of two biotypes of Bemisia tabaci on some common weeds. Entomol. Exp. Appl. 95: 63 70. Nakhla, M. K., D. P. Maxwell, R. T. Martinez, M. G. Carvalho, and R. L. Gibertson. 1994. Widespread occurrence of the Eastern Mediterranea n strains of tomato yellow leaf curl geminivirus in tomatoes in the Dominican Republic. Plant Dis. 78: 926. Naranjo, S. E. 1996. Sampling Bemisia for research and pest management applications, pp. 209 224. In D. Gerling and R. T. Mayer [eds.], Bemisia : 19 95 Taxonomy, Biology, Damage, Control and Management. Intercept, Andover, UK. Naranjo, S. E. 2001. Conservation and evaluation of natural enemies in IPM systems for Bemisia tabaci Crop Protect. 20: 835 852. Naranjo, S.E. and P.C. Ellsworth. 2001. Introd uction Special Issue: Challenges and opportunities for pest management of Bemisia tabaci in the new century. Crop Protect. 20: 707. Naranjo, S. E. and H. M. Flint. 1994. Spatial distribution of preimaginal Bemisia tabaci (Homoptera: Aleyrodidae) in cotto n and development of fixed precision, sequential sampling plans. Environ. Entomol. 23: 254 266.

PAGE 192

192 Naranjo, S. E. and H. M. Flint. 1995. Spatial distribution of adult Bemisia tabaci (Homoptera: Aleyrodidae) in cotton, and development and validation of fixed precision sampling plans for estimating population density. Environ. Entomol. 24: 261 270. Naranjo, S. E., H. M. Flint, and T. J. Henneberry. 1995. Comparative analysis of selected sampling methods for adult Bemisia tabaci (Homoptera: Aleyrodidae) in cott on. J. Econ. Entomol. 88: 1666 1678. Naranjo, S. E., P. C. Ellsworth, P. C. Chu, T. J. Henneberry, D. G. Riley, T. F. Watson, and R. L. Nichols. 1998. Action thresholds for the management of Bemisia tabaci (Homoptera: Aleyrodidae) in cotton J. Econ. Entom ol. 91: 1415 1426. Naranjo, S. E., S. J. Castle, P. J. De Barro, and S. S. Liu. 2010. Population Dynamics, Demography, Dispersal and Spread of Bemisia tabaci In P. A. Stansly and S. E. Naranjo [eds.], Bemisia : Bionomics and Management of a Global Pest. S pringer, London. Natwick, E. T. and A. Durazo III. 1985. Polyester covers protect vegetable from whiteflies and virus disease. Calif. Agric. 39: 21 22. Navas Castillo, J., S. Sanchez Campos, J. A. Diaz, E. Saez Alonso, and E. Moriones. 1999. Tomato yell ow leaf curl virus Is causes a novel disease of common bean and severe epidemics in tomato in Spain. Plant Dis. 83: 29 32. Naveed, M. A Salam, and M. A. Saleem. 2007. Contribution of cultivated crops, vegetables, weeds and ornamental plant in harboring of Bemisia tabaci (Homoptera: Aleyrodidae) and associated parasitoids (Hymenoptera: Aphelinidae) in cotton agroecosystems in Pakistan. J. Pest Sci. 80: 191 197. Navot, N., R. Ber, and H. Czosnek. 1989. Rapid detection of tomato yellow leaf curl virus in s quashes of plant and insect vectors. Phytopathology. 79: 562 568. Nelson, M. R., R. Felix Gastelum, T. V. Orum, L. J. Stowell, and D. E. Myers. 1994. Geographic information systems and geostatistics in the design and validation of regional plant virus man agement programs. Phytopathology. 84: 898 905. Nelson, M. R., T. V. Orum, R. Jaime Garcia, and A. Nadeem. 1999. Applications of geographic information systems and geostatistics in plant disease epidemiology and management. Plant Dis. 83: 308 319. Nichols on, W. F., R. Senn, C. R. Fluckiger, and D. Fuog. 1996. Pymetrozine a novel compound for control of whiteflies, pp. 635 639. In D. Gerling and R. T. Mayer [eds.], Bemisia : 1995 Taxonomy, Biology, Damage, Control and Management. Intercept Andover, UK.

PAGE 193

193 No mikou, M., A. Janssen, R. Schraag, and M. W. Sabelis. 2001. Phytoseiid predators as potential biological control agents for Bemisia tabaci Exp. Appl. Acarol. 25: 271 291. Noonan, G. R. 2003. GIS Technology. A powerful tool for entomologists. Insight. A Milwaukee Public Museum Series in Natural History http://www.mpm.edu/downloads/collections/pubs/insight/NoonanGISrevie w.pdf.Feb 15, 2011. Nuessly, G. S., D. E. Meyerdirk, D. L. Coudriet, and T. J. Henneberry. 1994. The effect of short season cotton production schedules on Bemisia tabaci (Gennadius). Southwest. Entomol. 19: 209 217. Obrycki, J. J. and T. J. Kring. 1998. Predaceous coccinellidae in biological con trol. Annu. Rev. Entomol. 43: 295 321. Ohnishi, J., T. Kitamura, F. Terami, and K. Honda. 2009. A selective barrier in the midgut epithelial cell membrane of the nonvector whitefly (Trialeurodes vaporariorum ) to Tomato yellow leaf curl virus uptake. J. Ge n. Plant Path. 75: 131 139. Omondi, A. B., D. Obeng Ofori, R. A. Kyerematen, and E. Y. Danquah. 2005. Host preference and suitability of some selected crops for two biotypes of Bemisia tabaci in Ghana. Entomol. Exp. Appl. 115: 393 400. Orozco Santos, M., M. Lopez Arriaga, O. Perez Zamora, and S. F. Delgadillo. 1994. Effect of transparent mulch, floating row covers and oil sprays on insect populations, virus diseases and yield of cantaloupe. Biol. Agric. Hort. 10: 229 234. Orozco Santos, M., O. Perez Za mora, and M. Lopez Arriaga. 1995. Floating row cover and transparent mulch to reduce insect population, virus diseases and increase yield in cantaloupe. Fla. Entomol. 78: 493 501. Palmer, J. O. 1985. Ecological genetics of wing length, flight propensity and early fecundity in a migratory insect. In M. A. Rankin [ed.], Migration: Mechanisms and Adaptive Significance. Contributions in Marine Science Supplement. Palumbo, J. C., A. Jr. Tonhasca, and D. N. Byrne. 1995. Evaluation of three sampling methods for estimating adult sweetpotato whitefly (Homoptera: Aleyrodidae) abundance on cantaloupes. J. Econ. Entomol. 88: 1393 1400. Palumbo, J. C., A. R. Horowitz, and N. Prabhaker. 2001. Insecticidal control and resistance management for Bemisia tabaci. Crop Prot ect. 20: 739 765.

PAGE 194

194 Park, J J., D. H. Lee, K I. Shin, J H. Lee, and K. Cho. 2009. Analysis of spatial and temporal associations of adult and immature Frankliniella occidentalis Pergande (Thysanoptera: Thripidae) in cucumber greenhouses. Appl. Entomol. Zool. 44: 569 577. Park, Y. L. and J. J. Obrycki. 2004. Spatio temporal distribution of corn leaf Aphids (Homoptera: Aphididae) and lady beetles (Coleoptera: Coccinellidae) in Iowa cornfields. Biol. Control. 31: 210 217. Park, Y L. and J. J. Tollefson. 2005. Characterization of the spatial dispersion of corn root injury by corn rootworms (Coleoptera: Chrysomelidae). Field and Forage Crops. 98: 378 383. Park, Y L. and J. J. Tollefson. 2006. Development and economic evaluation of spatial sampling plans for corn rootworm Diabrotica spp. (Col., Chrysomelidae) adults. J. Appl. Entomol. 130: 337 342. Park, Y L., T. M. Perring, C. A. Farrar, and C. Gispert. 2006. Spatial and temporal distributions of two sympatric Homalodisca spp. (Hemiptera: Cicadellidae): Implica tions for areawide pest management. Agric., Ecosyst. Environ. 113: 168 174. Pasek, J. E. 1988. Influence of wind and windbreaks on local dispersal of insects. Agric., Ecosyst. Environ. 22/23: 539 554. Patel, V. C. and H. K. Patel. 1966. Inter relationshi p between whitefly ( Bemisia tabaci Genn.) population and the incidence of leaf curl in bidi tobacco ( Nicotiana tabacum L.) in relation to different planting dates. Indian J. Entomol. 28: 339 344. Patz, J. A. and U. E. C. Confalonieri. 2005. Human health: ecosystem regulation of infectious diseases, pp. 391 415, The Millenium Ecosystem Assessment Series, Findings of the Conditional and Trends Working Group. Island Press, Washington D. C. Perez, C. J., P. Alvarado, C. Narvaez, F. Miranda, L. Hernandez, H. Vanegas, A. Hruska, and A. M. Shelton. 2000. Assessment of insecticide resistance in five insect pests attacking field and vegetable crops in Nicaragua. J. Econ. Entomol. 93: 1779 1787. Perring, T. M. 2001. The Bemisia tabaci species complex. Crop Protect 20: 725 737. Perring, T. M. R. N. Royalty, and C. A. Farrar. 1989. Floating row covers for the exclusion of virus vectors and the effect of disease incidence and yield of cantaloupe. J. Econ. Entomol. 82: 1709 1715.

PAGE 195

195 Perring, T. M., A. D. Cooper, R. J Rodriguez, C. A. Farrar, and T. S. Bellows, Jr. 1993. Identification of a whitefly species by genomic and behavioral studies. Science. 259: 74 77. Perry, A. S. 1985. The relative susceptibility to several insecticides of adult whiteflies (Bemisia tabaci ) from various cotton growing areas in Israel. Phytoparasitica. 13: 77 78. Perry, J. N. 1995. Spatial analysis by distance indices. J. Anim. Ecol. 64: 303 314. Perry, J. N. 1998. Measures of spatial pattern for counts. Ecology. 79: 1008 1017. Perry, J. N. and P. M. Dixon. 2002. A new method to measure spatial association for ecological count data. Ecoscience. 9: 133 141. Perry, J. N., L. Winder, J. M. Holland, and R. D. Alston. 1999. Red blue plots for detecting clusters in count data. Ecol. Lett. 2: 10 6 113. Pico, B., M. J. Diez, and F. Nuez. 1996. Viral diseases causing the greatest economic losses to the tomato crop. II. The Tomato yellow leaf curl virus a review. Scientia Horticulturae. 67: 151 196. Pico, B., M. Ferriol, M. J. Diez, and F. Nuez 1999. Developing tomato breeding lines resistant to tomato yellow leaf curl virus. Plant Breeding. 118: 537 542. Pollard, D. G. 1955. Feeding habits of the cotton whitefly, Bemisia tabaci Genn. (Homoptera: Aleyrodidae). Ann. Appl. Biol. 43: 664 671. Po lston, J. E. and P. K. Anderson. 1997. The emergence of whitefly transmitted geminiviruses in tomato in the Western Hemisphere. Plant Dis. 81: 1358 1369. Polston, J. E. and M. Lapidot. 2007. Management of tomato yellow leaf curl virus: US and Israel persp ectives, pp. 251 262. In H. Czosnek [ed.], Tomato Yellow Leaf Curl Virus Disease. Springer, Dordrecht. Polston, J. E. and T. Sherwood. 2003. Pymetrozine interferes with transmission of Tomato yellow leaf curl virus by the whitefly Bemisia tabaci Phytopar asitica. 31: 490 498. Polston, J. E., D. Bois, C. A. Serra, and C. Concepcion. 1994. First report of a Tomato Yellow Leaf Curl like Geminivirus in the Western Hemisphere. Plant Dis. 78: 831. Polston, J. E., D. O. Chellemi, D. J. Schuster, R. J. McGovern, and P.A. Stansly. 1996. Spatial and temporal dynamics of tomato mottle geminivirus and Bemisia tabaci (Genn.) in Florida tomato fields. Plant Dis. 80: 1022 1028.

PAGE 196

196 Polston, J. E., R. J. McGovern, and L. G. Brown. 1999. Introduction of tomato yellow leaf cu rl virus in Florida and implications for the spread of this and other geminiviruses of tomato. Plant Dis. 81: 1358 1369. Polston, J. E., A. Cohen, T. A. Sherwood, R. Ben Joseph, and M. Lapidot. 2006. Capsicum species: Symptomless hosts and reservoirs of T omato yellow leaf curl virus Phytopathology. 96: 447 452. Polston, J. E., D. J. Schuster, and J. E. Taylor. 2009. Identification of weed reservoirs of Tomato yellow leaf curl virus in Florida, pp. 32 33. In E. Simonne, C. Snodgrass and M. Ozores Hampton [eds.], Florida Tomato Institute. University of Florida, Naples, Florida. Powell, D. A. and T. S. Bellows. 1992. Preimaginal development and survival of Bemisia tabaci on cotton and cucumber. Environ. Entomol. 21: 359 363. Prabhaker, N., D. L. Coudriet, and N. C. Toscano. 1988. Effect of synergists on organophosphate and permethrin resistance in sweetpotato whitefly (Homoptera: Aleyrodidae). J. Econ. Entomol. 81: 34 39. Prabhaker, N., N. C. Toscano, and D. L. Coudriet. 1989. Susceptibility of the immatur e and adult stages of the sweetpotato whitefly (Homoptera: Aleyrodidae) to selected insecticides. J. Econ. Entomol. 82: 983 988. Prabhaker, N., N. C. Toscano, T. M. Perring, G. Nuessley, K. Kido, and R. R. Youngman. 1992. Resistance monitoring of the Swee tpotato whitefly (Homoptera: Aleyrodidae) in the Imperial Valley of California. J. Econ. Entomol. 85: 1063 1068. Prabhaker, N., N. C. Toscano, S. J. Castle, and T. J. Henneberry. 1997. Selection for resistance to imidacloprid in silverleaf whiteflies from the Imperial Valley and development of a hydroponic bioassay for resistance monitoring. Pestic. Sci. 51: 419 428. Prabhaker, N., N. C. Toscano, and T. J. Henneberry. 1998. Evaluation of insecticide rotations and mixtures as resistance management strategi es for Bemisia argentifolii (Homoptera: Aleyrodidae). J. Econ. Entomol. 91: 820 826. Praveen, S., C. M. Kushwaha, A. K. Mishra, V. Singh, R. K. Jain, and V. Anupam. 2005. Engineering tomato for resistance to tomato leaf curl disease using viral rep gene s equences. Plant Cell Tiss. Org. Cult. 83: 311 318. Price, J. F. 1987. Controlling a new pest. Greenhouse Grower. 570: 72 73. Putnam, A. R., J. Defrank, and J. P. Barnes. 1983 Exploitation of allelopathy for weed control in annual and perennial cropping s ystems. J. Chem. Ecol. 9: 1001 1010.

PAGE 197

197 Ramappa, H. K., V. Muniyappa, and J. Colvin. 1998. The contribution of tomato and alternative host plants to tomato leaf curl virus inoculum pressure in different areas of South India. Ann. Appl. Biol. 133: 187 198. R eay Jones, F. P. F. 2010. Spatial distribution of the cereal leaf beetle (Coleoptera: Chrysomelidae) in wheat. Environ. Entomol. 39: 1943 1952. Reay Jones, F. P. F., M. D. Toews, J. K. Greene, and R. B. Reeves. 2010. Spatial dynamics of stink bugs (Hemipt era: Pentatomidae) and associated boll injury in Southeastern cotton fields. Environ. Entomol. 39: 956 969. Reeves, R. B., J. K. Greene, F. P. F. Reay Jones, M. D. Toews, and P. D. Gerard. 2010. Effects of adjacent habitat on populations of stink bugs (He teroptera: Pentatomidae) in cotton as part of a variable agricultural landscape in South Carolina. Environ. Entomol. 39: 1420 1427. Riis, L. and G. Nachman. 2006. Migration, trapping and local dynamics of whiteflies (Homoptera: Aleyrodidae). Agric. For. E ntomol. 8: 233 241. Riley, D. G. and M. A. Ciomperlik. 1997. Regional population dynamics of whitefly (Homoptera: Aleyrodidae) and associated parasitoids (Hymenoptera: Aphelinidae). Environ. Entomol. 26: 1049 1055. Riley, D. G. and J. C. Palumbo. 1995. A ction thresholds for Bemisia argentifolii (Homoptera: Aleyrodidae) in cantaloupe. J. Econ. Entomol. 88: 1733 1738. Ritchie, M. and D. Pedgley. 1989. Desert locusts cross the Atlantic. Antenna. 13: 10 12. Rojas, M. R., H. Jiang, R. Salati, B. Xoconostle C azares, M. R. Sudarshana, W. J. Lucas, and R. L. Gilbertson. 2001. Functional analysis of proteins involved in movement of the monopartite Begomovirus, Tomato yellow leaf curl virus Virology. 291: 110 125. Rom, M., Y. Antignus, D. Gidoni, M. Pilowsky, an d S. Cohen. 1993. Accumulation of tomato yellow leaf curl virus DNA in tolerant and susceptible tomato lines. Plant Dis. 77: 253 257. Rosell, R. C., I. D. Bedford, D. R. Frohlich, R. J. Gill, J. K. Brown, and P. G. Markham. 1997. Analysis of morphological variation in distinct populations of Bemisia tabaci (Homoptera: Aleyrodidae). Ann. Entomol. Soc. Am. 90: 575 589. Rossi, R. E., D. J. Mulla, A. G. Journel, and E. H. Franz. 1992. Geostatistical tools for modeling and interpreting ecological spatial depen dence. Ecol. Monogr. 62: 277 314.

PAGE 198

198 Rubinstein, G. and H. Czosnek. 1997. Long term association of tomato yellow leaf curl virus with its whitefly vector Bemisia tabaci : effect on the insect transmission capacity, longevity and fecundity. J. Gen. Virol. 78: 2683 2689. Rubinstein, G., S. Morin, and H. Czosnek. 1999. Transmission of Tomato Yellow Leaf Curl Geminivirus to imidacloprid treated tomato plants by the whitefly Bemisia tabaci (Homoptera: Aleyrodidae). Horticultural Entomology. 92: 658 662. Russell, L. M. 1957. Synonyms of Bemisia tabaci (Gennadius) (Homoptera, Aleyrodidae). Bulletin of Brooklyn Entomology 52: 122 123. Russell, L. M. 1975. Collection records of Bemisia tabaci (Gennadius) in the United States (Hemiptera: Homoptera: Aleyrodidae). Coope rative Economic Insect Report. 25: 229 230. Rybicki, E. P. and G. Pietersen. 1999. Plant virus disease problems in the developing world. Adv. Virus Res. 53: 127 175. Saikia, A. K. and V. Muniyappa. 1989. Epidemiology and control of tomato leaf curl virus in India. Tropical Agriculture (Trinidad). 66: 350 354. Salas, J. and O. Mendoza. 1995. Biology of the sweetpotato whitefly (Homoptera: Aleyrodidae) on tomato. Fla. Entomol. 78: 154 160. Salati, R., M. K. Nahkla, M. R. Rojas, P. Guzman, J. Jaquez, D. P. Maxwell, and R. L. Gilbertson. 2002. Tomato yellow leaf curl virus in the Dominican Republic: Characterization of an infectious clone, virus monitoring in whiteflies, and identification of reservoir hosts. Phytopathology. 92: 487 496. Sanchez Campos, S., J. Navas Castillo, F. Monci, J. A. Diaz, and E. Moriones. 2000. Mercurialis ambigua and Solanum luteum : two newly discovered natural hosts of tomato yellow leaf curl geminiviruses. Eur. J. Plant Pathol. 106: 391 394. Sappington, T. W. and W. B. Showers. 1992. Reproductive maturity, mating status, and long duration flight behavior of Agrotis ipsilon (Lepidoptera: Noctuidae) and the conceptual misuse of the oogenesis flight syndrome by entomologists. Environ. Entomol. 21: 677 688. SAS. 2002. SAS/STAT user' s guide. 9.2. Cary, NC. Schell, S. P. and J. A. Lockwood. 1997. Spatial analysis of ecological factors related to rangeland grasshopper (Orthoptera: Acrididae) outbreaks in Wyoming. Environ. Entomol. 26: 1343 1353.

PAGE 199

199 Schotzko, D. J. and L. E. O'Keeffe. 1 989. Geostatistical description of the spatial distribution of Lygus hesperus (Heteroptera, Miridae) in lentils. J. Econ. Entomol. 82: 1277 1288. Schotzko, D. J. and L. E. O'Keeffe. 1990. Effect of sample placement on the geostatistical analysis of the sp atial distribution of Lygus hesperus (Heteroptera: Miridae) in lentils. J. Econ. Entomol. 83: 1888 1900. Schuster, D. J. 2001. Relationship of the silverleaf whitefly population density to severity of irregular ripening of tomato. HortScience. 36: 1089 10 90. Schuster, D. J. 2004. Squash as a trap crop to protect tomato from whitefly vectored tomato yellow leaf curl. Int. J. Pest Manage. 50: 281 284. Schuster, D. J. 2007. Whitefly resistance update. In A. Whidden, P. Gilreath and E. Simonne [eds.], Florid a Tomato Institute Proceedings. Florida Tomato Institute, Naples, FL. Schuster, D. J. and R. D. Caballero. 2010. Monitoring resistance of the sweetpotato whitefly to insecticides. Gainesville, Florida. Schuster, D. J., T. G. Mueller, J. Kring, and J. F. Price. 1990. Relationship of the sweetpotato whitefly to a new tomato fruit disorder in Florida. HortScience. 25: 1618 1620. Schuster, D. J., J. B. Kring, and J. F. Price. 1991. Association of the sweetpotato whitefly with a silverleaf disorder of squas h. HortScience. 26: 155 156. Schuster, D. J., J. E. Polston, and J. F. Price. 1992. Reservoirs of the sweetpotato whitefly for tomatoes in West Central Florida. Proceedings of the Florida State Horticulture Society. 105: 311 314. Schuster, D. J., J. E. F underburk, and P. A. Stansly. 1996a. IPM in tomatoes, pp. 387 411. In D. Rosen, J. L. Capinera and F. D. Bennet [eds.], Integrated Pest Management A Florida Perspective. Intercept Ltd, Andover, Hants, UK. Schuster, D. J., P.A. Stansly, and J. E. Polst on. 1996b. Expressions of plant damage by Bemisia pp. 153 165. In D. Gerling and R. T. Mayer [eds.], Bemisia 1995: Taxonomy, Biology, Damage, Control and Management. Intercept Ltd, Andover, Hants. Schuster, D. J., P.A. Stansly, J. E. Polston, P. R. Gilre ath, and E. McAvoy. 2007a. Management of whiteflies, whitefly vectored plant virus, and insecticide resistance for vegetable production in Southern Florida. EDIS .Entomology and Nematology Dept., Florida Cooperative Extension Service, IFAS, University of Fl orida.Gainesville. http://edis.ifas.ufl.edu/in695.3/10/2011

PAGE 200

200 Schuster, D. J., J. E. Taylor, C. D. Stanley, J. E. Polston, and S. Grunwald. 2007b. Evaluating factors affecting movement of the silverle af whitefly and tomato yellow leaf curl virus. Tomato Research Report for 2006 2007. 56 63. Schuster, D. J., S. K. Thompson, L. D. Ortega, and J. E. Polston. 2009. Laboratory evaluation of products to reduce settling of sweetpotato whitefly adults. J. Eco n. Entomol. 102: 1482 1489. Scott, J. W., M. R. Stevens, J. H. M. Barten, C. R. Thome, J. E. Polston, D. J. Schuster, and C. A. Serra. 1996. Introgression of resistance to whitefly transmitted geminiviruses from Lycopersicon chilense to tomato. In D. Ger ling and R. T. Mayer [eds.], Bemisia : 1995 Taxonomy, Biology, Damage, Control and Management Intercept, Andover. Seal, S. E., F. vandenBosch, and M. J. Jeger. 2006. Factors influencing Begomovirus evolution and their increasing global significance: Impli cations for sustainable control. Crit. Rev. Plant Sci. 25: 23 46. Secker, A. E., I. D. Bedford, P. G. Markham, and M. E. C. William. 1998. Squash, a reliable field indicator for the presence of B biotype of tobacco whitefly, Bemisia tabaci pp. 837 842, B righton crop protection conference pests and diseases. British Crop Protection Council, Farnham, UK. Setzer, R. W. 1995. Spatio temporal patterns of mortality of Pemphigus populitransversus and P. populicaulis on cottonwoods. Oecologia. 67: 310 321. Shah, D. A., G. C. Bergstrom, and P. P. Ueng. 2001. Foci of Stagonospora nodorum blotch in winter wheat before canopy development. Phytopathology. 91: 642 647. Sharaf, N. 1986. Chemical control of Bemisia tabaci Agriculture Ecosystems & Environment. 17: 1 11 127. Sharov, A. A., A. M. Liebhold, and E. A. Roberts. 1996. Spatial variation among counts of gypsy moths (Lepidoptera: Lymantriidae) in pheromone baited traps at expanding population fronts. Environ. Entomol. 25: 1312 1320. Shepherd, R. F., G. A. Va n Sickle, and D. H. L. Clarke. 1988. Spatial relationships of douglas fir tussock moth defoliation within habitat and climatic zones. Proceedings Lymantriidae: A comparison of features of New and Old World tussock moths. 381 400. Showers, W. B. 1997. Migr atory ecology of the black cutworm. Annu. Rev. Entomol. 42: 393 425.

PAGE 201

201 Simmons, A. M., H. F. Harrison, and Kai Shu Ling. 2008. Forty nine new host plant species for Bemisia tabaci (Hemiptera: Aleyrodidae). Entomol. Sci. 11: 385 390. Simmons, A. M., C. S. K ousik, and A. Levi. 2010. Combining reflective mulch and host plant resistance for sweetpotato whitefly (Hemiptera: Aleyrodidae) management in watermelon. Crop Protect. 29: 898 902. Sinisterra, X. H., C. L. McKenzie, W. B. Hunter, C. A. Powell, and RG Sha tters, Jr. 2005. Differential transcriptional activity of plant pathogenic begomoviruses in their whitefly vector ( Bemisia tabaci Gennadius: Hemiptera Aleyrodidae). J. Gen. Virol. 86: 1525 1532. Smith, H. A. and R. McSorley. 2000. Potential of field corn as a barrier crop and eggplant as a trap crop for management of Bemisia argentifolii (Homoptera: Aleyrodidae) on common bean in North Florida. Fla. Entomol. 83: 145 158. Smith, H. A., R. L. Koening, H. L. McAuslane, and R. McSorley. 2000. Effect of silv er reflective mulch and a summer squash trap crop on densities of immature Bemisia argentifolii (Homoptera: Aleyrodidae) on organic bean. J. Econ. Entomol. 93: 726 731. Sonka, S. T., M. E. Bauer, E. T. Cherry, J. W. Colburn, R. E. Heimlich, D. A. Joseph, J. B. Leboeuf, E. Lichtenberg, D. A. Mortensen, S. W. Searcy, S. L. Ustin, and S. J. Ventura. 1997. Precision Agriculture in the 21st Century. Geospatial and Information Technologies in Crop Management. National Academy Press, Washington D.C. Stansly, P. A. 1995. Seasonal abundance of silverleaf whitefly in southwest Florida vegetable fields, pp. 234 242, Proceeding of the Florida Horticultural Society. Florida State Horticultural Society. Stansly, P. A. and D. J. Schuster. 1990. Whitefly Update, pp. 20 4 2. In W. M. Stall [ed.], Proceedings of the Florida Tomato Institute Vegetable Crops Special Series SS VEC 001. IFAS, University of Florida, Gainesville. Stansly, P.A., T. X. Lui, D. J. Schuster, and D. E. Dean. 1996. Role of Biorational Insecticides in M anagement of Bemisia In D. Gerling and R. T. Mayer [eds.], Bemisia 1995: Taxonomy, Biology, Damage, Control and Management. Intercept, Andover, UK. Stansly, P.A., T. X. Lui, and C. S. Vavrina. 1998. Response of Bemisia argentifolii (Homoptera: Aleyrod idae) to imidacloprid under greenhouse, field and laboratory conditions. J. Econ. Entomol. 91. Steinberg, D. and P. Colla. 1995. CART: Tree Structured Non Parametric Data Analysis. San Diego, CA.

PAGE 202

202 Suwwan, M. A., M. Akkawi, A. M. Al Musa, and A. Mansou r. 1988. Tomato performance and incidence of tomato yellow leaf curl (TYLC) virus as affected by type of mulch. Scientific Horticulture. 37: 39 45. Taylor, L. R. 1961. Aggregation, variance and the mean. Nature. 189: 732 735. Taylor, L. R., I. P. Woiwod, and J. N. Perry. 1978. The density dependence of spatial behaviour and the rarity of randomness. J. Anim. Ecol. 47: 383 406. Taylor, R. A. J., S. Shalhevet, I. Spharim, M. J. Berlinger, and S. Lebiush Mordechi. 2001. Econo mic evaluation of insect proof screens for preventing tomato yellow leaf curl virus of tomatoes in Israel. Crop Protect. 20: 561 569. Thomas, C. F. G., L. Parkinson, G. J. K. Griffiths, A. Fernandez Garcia, and E. J. P. Marshall. 2001. Aggregation and tem poral stability of carabid beetle distributions in field and hedgerow habitats. J. Appl. Ecol. 38: 100 116. Thomson, M. C. and S. J. Connor. 2000. Environmental information systems for the control of arthropod vectors of disease. Med. Vet. Entomol. 14: 22 7 244. Thresh, J. M. 1976. Gradients of plant diseases. Ann. Appl. Biol. 82: 381 406. Tillman, P. G., T. D. Northfield, R. F. Mizell, and T. C. Riddle. 2009. Spatiotemporal patterns and dispersal of stink bugs (Heteroptera: Pentatomidae) in peanut cotton farmscapes. Environ. Entomol. 38: 1038 1052. Tonhasca, Jr. A., J. C. Palumbo, and D. N. Byrne. 1994. Aggregation patterns of Bemisia tabaci in response to insecticide applications. Entomol. Exp. Appl. 72: 265 272. Toscano, L. C., A. L. Boica Junior, and W. I. Maruyama. 2002. Nonpreference of whitefly for oviposition in tomato genotypes. Scientia Agricola. 59: 677 681. Tsai, J. H. and K. Wang. 1996. Development and reproduction of Bemisia argentifolii (Homoptera: Aleyrodidae) on five host plants. Environ Entomol. 25: 810 816. Tu, Z., D. N. Byrne, and H. H. Hagedorn. 1997. Vitellin of the sweet potato whitefly, Bemisia tabaci : biochemical characterization of titer changes in the adult. Arch. Insect Biochem. Physiol. 34: 223 237. Tubajika, K. M., E. L. C iverolo, M. A. Ciomperlik, D. A. Luvisi, and J. M. Hashim. 2004. Analysis of the spatial patterns of Pierce's disease incidence in the lower San Joaquin Valley in California. Phytopathology. 94: 1136 1144.

PAGE 203

203 Turechek, W. W. 2010. Environmental and geographi cal variables associated with TYLCV epidemics in southwest Florida. In M. Ozores Hampton and C. Snodgrass [eds.], 2010 Florida Tomato Institute Proceedings. University of Florida, Naples, Florida. Turechek, W. W. and L. V. Madden. 1999. Spatial pattern an alysis of strawberry leaf blight in perennial production systems. Phytopathology. 89: 421 433. Ucko, O., S. Cohen, and R. Ben Joseph. 1998. Prevention of virus epidemics by a crop free period in the Arava Region of Israel. Phytoparasitica. 26: 313 321. V alverde, R. A., P. Lotrakul, and A. D. Landry. 2001. First report of tomato yellow leaf curl virus in Louisiana. Plant Dis. 85: 230. Van Lenteren, J. C. and L. P. J. J. Noldus. 1990. Whitefly Plant Relationships: Behavioural and Ecological Aspects. In D. Gerling [ed.], Whiteflies: their Bionomics, Pest Status and Management. Intercept, Andover, UK. Van Schelt, J., J. Klapwijk, M. Letard, and C. Aucouturier. 1996. The use of Macrolopus caligionosus as a whitefly predator in protected crops. In D. Gerling a nd R. T. Mayer [eds.], Bemisia 1995: Taxonomy, Biology, Damage, Control and Management. Intercept, Andover, UK. Van Sickle, G. A. 1989. GIS A tool in forest pest management pp. 213 219. In G. Wygul, J. McCoy and J. Smith [eds.], A Wider Perspective, Pro ceedings. GIS 89, Forestry Canada. vanEngelsdorp, D., N. Speybroeck, J. D. Evans, B. K. Nguyen, C. Mullin, M. Frazier, D. Cox Foster, Y. Chen, D. R. Tarpy, E. Haubruge, J. S. Pettis, and C. Saegerman. 2010. Weighing risk factors associated with bee colony collapse disorder by classification and regression tree analysis. J. Econ. Entomol. 103: 1517 1523. Varma, A. and V. G. Malathi. 2003. Emerging geminivirus problems: A serious threat to crop production. Ann. Appl. Biol. 142: 145 164. Vidavsky, F. and H Czosnek. 1998. Tomato breeding lines resistant and tolerant to tomato yellow leaf curl virus issued from Lycopersicon hirsutum Phytopathology. 88: 910 914. Vidavsky, F., S. Leviatov, J. Milo, H. D. Rabinowitch, N. Kedar, and H. Czosnek. 1998. Response o f tolerant breeding lines of tomato, Lycopersicon esculentum originating from three different sources ( L. peruvianum L. pimpinellifolium and L. chilense ) to early controlled inoculation by tomato yellow leaf curl virus (TYLCV). Plant Breeding. 117: 165 1 69.

PAGE 204

204 Vidavsky, F., H. Czosnek, S. Gazit, D. Levy, and M. Lapidot. 2008. Pyramiding of genes conferring resistance to Tomato yellow leaf curl virus from different wild tomato species. Plant Breeding. 127: 625 631. Vieira, S. R., J. L. Hatfield, D. R. Niel sen, and J. W. Biggar. 1983. Geostatistical theory and application to variability of some agronomical properties. Hilgardia. 51: 1 75. Villar, A., E. Gomez, F. Morales, and P. Anderson. 1998. Effect of legal measures to control Bemisia tabaci and geminiv iruses in the Valley of Azua. Report from the National Integrated Pest Management Program. 16. Visser, J. H. 1988. Host plant finding by insect: orientation, sensory input and search patterns. J. Insect Physiol. 34: 259 268. Walker, G. P. and E. T. Natwi ck. 2006. Resistance to silverleaf whitefly, Bemisia argentifolii (Hem., Aleyrodidae), in Gossypium thurberi a wild cotton species. J. Appl. Entomol. 130: 429 436. Warner, D. J., L. J. Allen Williams, S. Warrington, A. W. Ferguson, and I. H. Williams. 2 003. Mapping, characterization, and comparison of the spatio temporal distributions of cabbage stem flea beetle ( Psylliodes chrysocephala ), carabids, and Collembola in a crop of winter oilseed rape ( Brassica napus ). Entomol. Exp. Appl. 109: 225 234. Watso n, T. F. 1993. Chemical control of the sweetpotato whitefly in cotton, pp. 221 239. Arizona Agricultural Experiment Station. Watson, T. F., J. C. Silvertooth, A. Tellez, and L. Lastra. 1992. Seasonal dynamics of sweetpotato whitefly in Arizona. Southwest. Entomol. 17: 149 167. Webb, S. E. and S. B. Linda. 1992. Evaluation of spunbonded polyethyline row covers as a method of excluding insects and viruses affecting fall grown squash in Florida. J. Econ. Entomol. 85: 2344 2352. Wege, C. 2007. Movement and l ocalization of Tomato yellow leaf curl viruses in the infected plant, pp. 185 206. In H. Czosnek [ed.], Tomato yellow leaf curl disease. Springer, Dordrecht. Weintraub, P. G. and M. J. Berlinger. 2004. Physical control in greenhouses and field crops. In A R. Horowitz and I. Ishaaya [eds.], Insect Pest Management Field and Protected Crops. Springer, Berlin.

PAGE 205

205 Weisz, R., S. Fleischer, and Z. Smilowitz. 1995. Map generation in high value horticultural integrated pest management appropriate interpolation met hods for site specific pest management of colorado potato beetle (Coleoptera: Chrysomelidae). J. Econ. Entomol. 88: 1650 1657. Westbrook, J. K., W. W. Wolf, P. D. Lingren, J. R. Raulston, J. D. Lopez, J. H. Matis, R. S. Eyster, J. F. Esquivel, and P. G. S chleider. 1997. Early season migratory flights of corn earworm (Lepidoptera: Noctuidae). Environ. Entomol. 26: 3 4. Williams, L., D. J. Schotzko, and J. P. McCaffrey. 1992. Geostatistical description of the spatial distribution of Limonis californicus (Co leoptera, Elateridae) wireworms in the Northwestern United States, with comments on sampling. Environ. Entomol. 21: 983 995. Wilson, L. T. and P. M. Room. 1983. Clumping patterns of fruit and arthropods in cotton, with implications for binomial sampling. Environ. Entomol. 12: 50 54. Xu, X. M. and L. V. Madden. 2004. Use of SADIE statistics to study spatial dynamics of plant disease epidemics. Plant Pathol. 53: 38 49. Yang, Y., T. Sherwood, C. P. Patte, E. Hiebert, and J. E. Polston. 2004. Use of Tomato y ellow leaf curl virus (TYLCV) Rep gene sequences to engineer TYLCV resistance in tomato. Phytopathology. 94: 490 496. Zalom, F. G., E. T. Natwick, and N. C. Toscano. 1985. Temperature regulation of Bemisia tabaci (Homoptera: Aleyrodidae) populations in im perial valley cotton. J. Econ. Entomol. 78: 61 64. Zamir, D., I. Ekstein Michelson, Y. Zakay, N. Navot, M. Zeidan, M. Sarfatti, Y. Eshed, E. Harel, T. Pleban, H. van Oss, N. Kedar, H. D. Rabinowitch, and H. Czosnek. 1994. Mapping and introgression of a to mato yellow leaf curl virus tolerance gene, TY 1. Theor. Appl. Genet. 88: 141 146. Zang, Lian Sheng, Wei Qiang Chen, and Shu Sheng Liu. 2006. Comparison of performance on different host plants between the B biotype and a non B biotype of Bemisia tabaci fr om Zhejiang, China. Entomol. Exp. Appl. 121: 221 227. Zhang, W., H. L. McAuslane, and D. J. Schuster. 2004. Repellency of ginger oil to Bemisia argentifolii (Homoptera: Aleyrodidae) on tomato. J. Econ. Entomol. 97: 1310 1318.

PAGE 206

206 BIOGRAPHICAL SKETCH Jam was born in Tifton, Ga. and graduated from Tift Co. High School in 2000. He graduated in 2004 from the Univers ity of Georgia with a B. S. in b iological s ciences from the College of Agriculture and Environmental Sciences. He completed a M. S. degree at the University of Georgia in 2006 working on the impact of beet armyworm in tomato He received his Ph.D. from the University of Florida in the summer of 2011. He has worked in entomology for over 10 years in various disciplines includin g row crop, medical, veterinary, fruit and vegetable entomology. His current research interests include the spatial distribution of Bemisia tabaci and Tomato yellow leaf curl virus in Florida tomato. In time away from the office, he enjoy s spending time with his family and enjoying the outdoors. Glory, glory to ole Georgia!