Citation
Management of Early Broad Mite [Polyphagotarsonemus latus (Banks)] Infestations with Neoseiulus californicus McGregor in Greenhouse-Grown Bell Pepper (Capsicum annuum L.)

Material Information

Title:
Management of Early Broad Mite [Polyphagotarsonemus latus (Banks)] Infestations with Neoseiulus californicus McGregor in Greenhouse-Grown Bell Pepper (Capsicum annuum L.)
Creator:
JOVICICH, ELIO
Copyright Date:
2008

Subjects

Subjects / Keywords:
Crops ( jstor )
Fruits ( jstor )
Greenhouses ( jstor )
Infestation ( jstor )
Leaves ( jstor )
Mites ( jstor )
Peppers ( jstor )
Pests ( jstor )
Plants ( jstor )
Seedlings ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Elio Jovicich. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
5/31/2017
Resource Identifier:
660161433 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

1 MANAGEMENT OF EARLY BROAD MITE [ Polyphagotarsonemus latus (Banks)] INFESTATIONS WITH Neoseiulus californicus McGregor IN GREENHOUSE-GROWN BELL PEPPER ( Capsicum annuum L.) By ELIO JOVICICH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Elio Jovicich

PAGE 3

3 ACKNOWLEDGMENTS I would like to express my gratitude to Dr Daniel Cantliffe, my supervisory committee chair, for his help, guidance, support, and patience I would also like to give my thanks to Dr. Lance Osborne, cochair, Dr. Peter Stoffella, Dr. Eric Simonne, and Dr. John VanSickle for their advice, encouragement, discussions and for shar ing their knowledge. Fi nancial support by the University of Florida made it possible for me to pursue my graduate studies. I am also thankful to the Tropical/Subtropical Agri culture Research Program for funding most of my research project. I am very thankful to Dr. Cantliffe and Stoffella, and to the UF Office of Graduate Minority Programs for their financial support during the last term of my studies. I extend my gratitude to all staff, students a nd faculty of the Horticultural Sc iences Department who in some way or another enriched my work and life during both my masters and doctoral programs here in Gainesville. My gratitude also extends to the Department of Horticultural Sciences, Institute of Food and Agricultural Sciences, College of Agri cultural and Life Sciences, and the Graduate Student Council at the University of Florida, as well as the Am erican Society for Horticultural Science for providing me with funds to attend scientific conferences. I thank Dr. Carl Welborn (Fla. Division of Plant Industries) for mite specimens identification and advice on m ite recovery; Dr. Marjory Hoy (UF Entomology and Nematology) for providing broad mites to start my colonies; Dr. Ramon Littell (UF IFAS Statistics) for statistical advice; Richard Cullen (UF Plant Dise ase Clinic) for plant di sease diagnostics; and Avri Saragosti (Netafim USA) for helping with training and re pairs of irrigation and greenhouse controllers. I thank Biotactics for their prompt and timely shipments of predatory mites, and acknowledge companies that donate d materials (Rogers-Syngenta, Speedling, and Cerexagri). I am thankful to John Thomas, Cecil Shine, Ge ne Hannah, Scott Taylor, and the many people in

PAGE 4

4 IFAS Facilities who helped with the construc tion and repairs needed in the greenhouses and growth chambers used in my research. Having been a teaching assistant, I want to also thank all the students in the classes I helped; I learned much from them. I want to thank friends, colleagues, and people who helped during my studies and research: Jorge Baldessa ri, Gonzalo Estavillo, Shubin Saha, Hctor Nuez-Palenius, Roco Daz de la Garza, Ni cacio Cruz Huerta, Elizabeth Thomas, Juan Rodriguez, Ji-Young Hong, Jeanmarie Mitchell, Nicole Pratt, Dzengai Rukuni, Jimmy and Melissa Webb, Brian McCollum, James Tully, Richar d Rogers, and David Norden. I also extend my thanks to Dr. Carlos Parera, Juan Aguilera and Dr. Daniel Leskovar for their friendship and support. Finally, I am forever tha nkful to my parents Mara and Jo rge, my brother Jorge, and my wife Samantha for their constant encouragement and love.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................3 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .......11 ABSTRACT....................................................................................................................... ............15 CHAPTER 1 INTRODUCTION..................................................................................................................17 1.1 Broad mite in Greenhouse-Grown Peppers.....................................................................17 1.2 Greenhouse Production of Bell Pepper in Florida...........................................................24 2 MANAGEMENT OF Polyphagotarsonemus latus (Banks) (ACARI: TARSONEMIDAE) WITH Neoseiulus californicus McGregor (ACARI: PHYTOSEIIDAE) ON BELL PEPPER ( Capsicum annuum L.) TRANSPLANTS....................................................................................................................30 2.1 Introduction.............................................................................................................. ........30 2.2 Materials and Methods....................................................................................................33 2.2.1 Treatments.............................................................................................................33 2.2.2 Plant Growing System, Environment, and Irrigation............................................34 2.2.3 Source of Mites, P. latus Infestation, and N. californicus Release.......................35 2.2.4 Plant Sampling and Mite Recovery.......................................................................36 2.2.5 Estimation of Plant Damage and Measurements of Seedling Growth..................37 2.2.6 Experimental Design and Data Analysis...............................................................38 2.3 Results................................................................................................................... ...........39 2.3.1 Polyphagotarsonemus latus Infestations with Absence of N. californicus ...........39 2.3.2 Polyphagotarsonemus latus Infestations with Releases of N. californicus ...........40 2.3.2.1 Seedlings with Unfolded Cotyledons Infested with P. latus .......................40 2.3.2.2 Seedlings with Two Unfolded Leaves Infested with P. latus .....................42 2.3.2.3 Seedlings with Four Unfolded Leaves Infested with P. latus .....................43 2.3.3 Neoseiulus californicus Released with Absence of P. latus ..................................43 2.3.4 Responses of Transplant Growth Variables and Damage Index to P. latus Cumulative Mite-Days.................................................................................................44 2.4 Discussion................................................................................................................ ........45 2.5 Summary................................................................................................................... .......51

PAGE 6

6 3 EFFECTIVE MANAGEMENT OF Polyphagotarsonemus latus (Banks) (ACARI: TARSONEMIDAE) WITH Neoseiulus californicus McGregor (ACARI: PHYTOSEIIDAE) RELEASED IN GREENHOUSE-GROWN PEPPER ( Capsicum annuum L.) CROPS INITIATED WITH INFESTED TRANSPLANTS............68 3.1 Introduction.............................................................................................................. ........68 3.2 Materials and Methods....................................................................................................73 3.2.1 Experiments and Treatments.................................................................................73 3.2.2 P. latus Infestation, N. californicus Release, and Sulfur Sprays...........................74 3.2.3 Transplant Production...........................................................................................74 3.2.4 Greenhouse Plant Arrangeme nt and Crop Management.......................................75 3.2.5 Leaf Sampling and Mite Recovery........................................................................77 3.2.6 Plant Damage Estimation......................................................................................77 3.2.7 Fruit and Plant Measurements...............................................................................78 3.2.8 Environmental Conditions.....................................................................................79 3.2.9 Cost and Benefit Estimation..................................................................................79 3.2.10 Experimental Design and Data Analysis.............................................................80 3.3 Results................................................................................................................... ...........81 3.3.1 Untreated P. latus Infestations in Fall-2004 and Spring-2005..............................81 3.3.2 Polyphagotarsonemus latus Management in Fall-2004........................................82 3.3.2 Polyphagotarsonemus latus Management in Spring-2005....................................84 3.3.5 Cost and Benefit of P. latus Management.............................................................87 3.3.6 Relationships Between Cumulative Mite-Days and Plant Damage Index, Plant Growth Variables and Marketable Fruit Yield...................................................88 3.4 Discussion................................................................................................................ ........89 3.5 Summary................................................................................................................... .......95 4 GREENHOUSE-GROWN COLO RED PEPPERS: A PROF ITABLE ALTERNATIVE FOR VEGETABLE PRODUCT ION IN FLORIDA?..........................................................112 4.1 Introduction.............................................................................................................. ......112 4.1.1 Greenhouses for Production of Vegetables in Florida........................................112 4.1.2 Pepper Imports to the U.S...................................................................................113 4.1.3 Field Production of Be ll Pepper in Florida..........................................................114 4.1.4 Research and Greenhouse Production of Bell Pepper in Florida........................115 4.1.5 Returns of Greenhouse-Grown Peppers..............................................................117 4.2 Methods................................................................................................................... ......117 4.2.1 Greenhouse Structure..........................................................................................117 4.2.2 Crop Cycle...........................................................................................................118 4.2.3 Plant Production System......................................................................................118 4.2.4 Pollinators, Pests, and Diseases...........................................................................119 4.2.5 Fruit Yields..........................................................................................................120 4.2.6 Pepper Fruit Prices..............................................................................................120 4.2.7 Budget Analysis...................................................................................................121 4.2.8 Sensitivity Analysis.............................................................................................123 4.2.9 Break-Even Analysis...........................................................................................124

PAGE 7

7 4.3 Results................................................................................................................... .........124 4.3.1 Prices for Mature Ri pened Bell Pepper Fruits.....................................................124 4.3.2 Investment and Costs...........................................................................................125 4.3.3 Budget Analysis...................................................................................................125 4.3.4 Sensitivity Analysis and Break-Even Analysis...................................................126 4.4 Discussion................................................................................................................ ......127 4.5 Summary................................................................................................................... .....130 APPENDIX A ADDITIONAL FIGURES AND TA BLES FOR CHAPTER 2...........................................147 B ADDITIONAL FIGURES AND TA BLES FOR CHAPTER 3...........................................154 LIST OF REFERENCES.............................................................................................................169 BIOGRAPHICAL SKETCH.......................................................................................................180

PAGE 8

8 LIST OF TABLES Table page 2-1. Pest-predator scenarios evalua ted during pepper seedling growth....................................54 2-2. Number of P. latus and N. californicus recovered from pepper seedlings with mites introduced at three seedling gr owth developmental stages................................................55 2-3. Bell pepper growth variables and seedli ng visual damage index in 42-days old transplants.................................................................................................................... ......56 2-4. Regression equations for the reduction of transplant growth variables relative to non-infested transplants, a nd for visual damage index as affected by increased cumulative m ite-days.........................................................................................................66 3-1. Biological ( N. californicus releases) and pesticide (s ulfur sprays) treatments evaluated for management of P. latus in greenhouse-grown pepper plants in Fall-2004 and Spring-2005................................................................................................98 3-2. Polyphagotarsonemus latus cumulative mite-days, relativ e plant growth variables to non-infested plants, and plant visual da mage index in greenhouse-grown pepper plants in Fall-2004 (106 DAT)........................................................................................104 3-3. Polyphagotarsonemus latus cumulative mite-days, relativ e plant growth variables to non-infested plants, and plant visual da mage index in greenhouse-grown pepper plants in Spring-2005 (88 DAT)......................................................................................105 3-4. Marketable fruit yield, P. latus -damaged fruit, and marketable fruit set in greenhouse-grown pepper plants in Fall-2004 (106 DAT)..............................................106 3-5. Marketable fruit yield, P. latus -damaged fruit, and marketable fruit set in greenhouse-grown pepper plants in Spring-2005 (88 DAT)...........................................107 3-6. Estimated pepper fruit gross return, pest management cost, return relative to no pest and benefit-cost ratios fo r crop scenarios in greenhous e experiments Fall-2004 and Spring-2005.................................................................................................................... ..108 3-7. Regression equations for plant visual dama ge index, fruit yield, fruit set, and plant growth variables, as affected by P. latus cumulative mite-days at 40 DAT on top canopy leaves of greenhousegrown pepper plants in Fa ll-2004 and Spring-2005 ........111 4-1. Area with greenhouse-grown bell peppers in selected countries that export colored fruits to the U.S. and area with gr eenhouse-grown peppers in the U.S...........................131 4-2. Volumes, values, and origins of bell pe ppers imported into the U.S., and volumes and values of the production, exports, and use of bell peppers in the U.S. in the year 2002........................................................................................................................... .......133

PAGE 9

9 4-3. Yields of bell pepper fruits produced in greenhouses, and in field crops in Florida.......134 4-4. Total harvested area, production, yiel d, and dollar values for field-grown bell peppers and greenhouse-grown peppe rs (season 1997) in Florida............................135 4-5. Structure dimensions of one of the tw o identical greenhouse un its, and pepper plant arrangement used in the en terprise budget analysis.........................................................136 4-6. Monthly marketable fruit yields, aver age wholesale market prices, and gross revenues in a typical fall to spring gr eenhouse-grown bell pepper crop in Florida with a total estimat ed yield of 13 kgm (2.7 lb/ft2).......................................................138 4-7. Summary of the investment costs and depreciation of a 0.78-ha (1.927-acre) greenhouse operation planned fo r bell pepper production...............................................139 4-8. Estimated fixed costs to produce bell pepper in a 0.78-ha (1.927-acre) greenhouse in north central Florida.........................................................................................................140 4-9. Estimated variable costs to produce be ll pepper in a 0.78-ha (1.927-acre) greenhouse in north central Florida.....................................................................................................141 4-10. Enterprise budget for greenhouse-grow n bell pepper [0.78 ha (1.927 acres)] in north-central Florida........................................................................................................144 4-11. Sensitivity analysis for bell pepper retu rns to capital and management per unit area and in the 0.78-ha ( 1.927-acre) greenhouse.....................................................................146 4-12. Estimated break-even prices for a range of marketable bell pepper fruit yields of 517 kgm-2 (1.0.5 lb/ft2).......... ...................................................................................146 A-1. Regression equations for growth variables and for visual damage index as affected by increased P. latus cumulative mite-days, in 42-days old pepper transplants..............149 A-2. Regression equations for seedling relative growth reduction and for visual damage index as affected by increased P. latus cumulative mite-days, in 42-days old pepper seedlings...................................................................................................................... .....150 B-1. Cost estimates for one sulfur spray on tran splanted bell pepper pl ants or seedlings for a 1-hectare greenhouse crop.............................................................................................156 B-2. Cost estimates for one release N. californicus on transplanted bell pepper plants or seedlings for a 1-hectare greenhouse crop.......................................................................157 B-3. Polyphagotarsonemus latus cumulative mite-days, plant growth variables, and plant visual damage index in greenhouse-grow n pepper plants in Fall-2004 (106 DAT)........158

PAGE 10

10 B-4. Polyphagotarsonemus latus cumulative mite-days, plant growth variables, and plant visual damage index in greenhouse-grown pepper plants in Spring-2005 (88 DAT).....159 B-5. Regression equations for plant visual dama ge index, fruit yield, fruit set, and plant growth variables, as affected by P. latus cumulative mite-days (in a log.-scale) at the time of first fruit harvest on top canopy leaves of gree nhouse-grown pepper plants in Fall-2004 and Spring-2005..............................................................................................163

PAGE 11

11 LIST OF FIGURES Figure page 1-1. Phoretic relationship betw een broad mite and whitefly.....................................................25 1-2. Life stages of P. latus .........................................................................................................26 1-3. Damage by P. latus on bell pepper shoots and fruit..........................................................27 1-4. Damage on bell plants one month after tr ansplanting infested seedlings with two P. latus ................................................ .............................................................................28 1-5. Neoseiulus californicus on pepper leaves..........................................................................29 2-1. Visual damage ratings on pepper seedling used to assess damage from P. latus feeding. ..............................................................................................................................57 2-2. Number of P. latus and N. californicus recovered after mites were introduced to pepper seedlings at three grow th developmental stages.................................................... 58 2-3. P. latus cumulative mite-days in pepper seedlings after P. latus and N. californicus were introduced at three seedli ng growth developmental stages.......................................59 2-4. Changes in visual damage indices in terminal shoots of pe pper seedlings after P. latus and N. californicus were introduced at three seedling growth developmental stages..................................................................................................................................60 2-5. Dry weight increase in pepper seedlings after P. latus and N. californicus were introduced at three seedling gr owth developmental stages................................................ 61 2-6. Stem length increase in pepper seedlings after P. latus and N. californicus were introduced at three seedling gr owth developmental stages................................................ 62 2-7. Number of leaves increa se in pepper seedlings after P. latus and N. californicus were introduced at three seedling grow th developmental stages. .............................................63 2-8. Leaf area increase in pepper seedlings after P. latus and N. californicus were introduced at three seedling gr owth developmental stages................................................ 64 2-9. Reduction of relative transplant growth variables and visual damage index as affected by P. latus cumulative mite-days in pepper seedlings that had pest infestations initiated with two P. latus at three seedling growth developmental stages.... 65 2-10. Transplant growth variables and dama ge index relationshi ps as affected by P. latus cumulative mite-days in 42-days old pepper seedlings..................................................... 67 3-1. Pepper plant damage indices used to assess damage from P. latus feeding...................... 99

PAGE 12

12 3-2. Number of P. latus and N. californicus recovered per unit of area of sampled leaf in greenhouse-grown pepper plants in Fall-2004.................................................................100 3-3. Number of P. latus and N. californicus recovered per unit area of sampled leaf in greenhouse-grown pepper plants in Spring-2005............................................................101 3-4. Polyphagotarsonemus latus cumulative mite-days per uni t of area of sampled leaf and plant damage index in greenhouse-grown pepper plants in Fall-2004.....................102 3-5. Polyphagotarsonemus latus cumulative mite-days per uni t of area of sampled leaf and plant damage index in greenhouse-gro wn bell pepper plants in Spring-2005..........103 3-6. Regression curves for variables m easured in bell pepper plants and P. latus cumulative mite-days (in a log.-scale) until 40 DAT in greenhouse experiments Fall-2004 and Spring-2005..............................................................................................110 4-1. Total value, volume, and unit value of be ll pepper fruits (all colors) imported into the U.S. from major shipping count ries throughout the years 1996............................132 4-2. Total sales value of bell peppers importe d to the U.S. from selected countries throughout the years 1995......................................................................................133 4-3. Passively ventilated greenhouses at the Protected Agriculture Center in Gainesville, Fla............................................................................................................... .137 4-4. Pepper plants grown in 11.4-L (3 gal) containers and trellised to the Spanish system in a passively ventilated greenhouse at the Protected Agri culture Center in Gainesville, Fla............................................................................................................... .137 4-5. Means and standard deviations for bell pepper fruit wholesale prices from years 1993 to 2002 obtained from Thursdays transacti ons at the Miami terminal market: A) prices for yellow ( ), red ( ), and green ( ) fruits produced in the field (shipped from Florida, California, Georgia, and No rth Carolina), and B) prices for yellow ( ), red ( ), and orange ( ) fruits imported and produced in greenhouses (shipped from The Netherlands, Israel, and Spain).................................................................................143 4-6. Estimated responses of tota l costs, returns to capital a nd management (excluding cost of land), and gross revenues to increased marketable fruit yields of greenhouse-grown bell peppers in north-central Florida.................................................145 A-1. Pest-predator scenarios evaluated on pepper seedlings...................................................147 A-2. Procedure used for recovering m ites from seedlings or leaves........................................148 A-3. Transplant growth variables and damage index as affected by P. latus cumulative mite-days (in a log.-scale) in 42-days old peppers .........................................................149

PAGE 13

13 A-4. Regression curves for seedling relative gr owth reduction and visual damage index as affected by increased P. latus cumulative mite-days (in a log.-scale), in 42-days old pepper seedlings...............................................................................................................150 A-5. Pepper transplants with P. latus infestation and no N. californicus release....................151 A-6. Pepper transplants with N. californicus released on the same day P. latus infested seedlings...................................................................................................................... .....151 A-7. Pepper transplants with N. californicus released before P. latus infested seedlings.......152 A-8. Pepper transplants with N. californicus released after P. latus infested seedlings..........152 A-9. Pepper transplants with a preventive release of N. californicus and no P. latus infestation.................................................................................................................... .....153 A-10. Pepper transplants with no mites.....................................................................................153 B-1. Pepper seedling infested with two P. latus that were transplanted in the greenhouse three days later............................................................................................................... ..154 B-2. Passively ventilated greenhouse and arrangement of plots in Spring-2005....................154 B-3. Daily air temperatures (maximum, m ean, and minimum) inside the passively ventilated greenhouse and over canopie s of pepper plants in Fall-2004 and Spring-2005.................................................................................................................... ..155 B-4. Relative marketable fruit yield in pepper plants infested with P. latus and that had one release of N. californicus at two release densities a nd at three different release times.......................................................................................................................... .......160 B-5. Relationships between selected vari ables measured in pepper plants and P. latus cumulative mite-days in infested plants treated with or without sulfur in Spring-2005.................................................................................................................... ..161 B-6. Regression curves for variables m easured in bell pepper plants and P. latus cumulative mite-days (in a log.-scale) at the time of first fruit harvest in experiments Fall-2004 and Spring-2005..............................................................................................162 B-7. Plant canopies of greenhouse-gr own peppers in Fall-2004 (75 DAT)............................164 B-8. Plant canopies of greenhouse-gr own peppers in Spring-2005 (45 DAT)........................165 B-9. Greenhouse-grown pepper plants with l eaves or fruit removed in Spring-2005 (88 DAT)..........................................................................................................................166

PAGE 14

14 B-10. Fruit on pepper plants th at were not infested with P. latus and that were infested before transplanting (Fall-2004)......................................................................................167 B-11. Fruit on pepper plants treated with and without N. californicus for an early P. latus infestation (Spring-2005).................................................................................................168

PAGE 15

15 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MANAGEMENT OF EARLY BROAD MITE [ Polyphagotarsonemus latus (Banks)] INFESTATIONS WITH Neoseiulus californicus McGregor IN GREENHOUSE-GROWN BELL PEPPER ( Capsicum annuum L.) By Elio Jovicich May 2007 Chair: Daniel J. Cantliffe Cochair: Lance S. Osborne Major Department: Horticultural Science The microscopic phytophagous mite, P. latus can infest pepper seedlings before transplanting in gree nhouses. Detecting P. latus is difficult and mite populations increase rapidly; thus, delayed control significantly decreases fruit yield and fruit quality. Multiple applications of miticides disrupt biological contro l programs. As an alternative to pesticide use, the predaceous mite N. californicus was studied for its effects on pest population, seedling and plant growth, and early fruit yiel d as affected by density and ti ming of predator release when infestations occurred during s eedling development. Under a controlled environment, two P. latus were placed on seedlings at cotyledonary, tw o-leaf, and four-leaf stages, and predators were released before or after the initial pest infestations. Transp lants 42-days old were undamaged when two predators were released 0 24 days pre-infestation and were seriously damaged with predators releas ed 9 days post-infestation. In two crop cycles, when seedli ngs were infested with two P. latus and transplanted three days later in a greenhouse, market able fruit was nil in plants with no pest management. When two predators were released 4 days after transplan ting, yields relative to n on-infested plants were <68%. However, yields were comparable to non-in fested plants when four predators/plant were

PAGE 16

16 used at any of the release times (6 days before, at, or 4 days after tran splanting); or when two predators/plant were released 6 days before transplanting. Exponential decay relationships described transplant and plant growth variable s, and fruit yield as affected by cumulative mite-days. Reductions of 5% in colored-fruit production occurred with 24 cumulative mite-days/leaf-cm2 within 40 days after tr ansplanting, in both s easons. Effective pest management resulted from preventive predator re leases and when initial prey-predator density ratios that were <3:1 seedling mites : seedling mites in seedlings, and <4:1 plant mites : leaf mites in plants 4 days after transplanting. Pest management would be mo re effective and econom ically-feasible with micronized-sulfur sprays than with predator re leases if interventions are delayed until first plant-injury symptoms. Spra ying first on transplants kept P. latus at low populations but more than five weekly-sprays may be needed in a full-season crop. Sulfur spraying was less costly than N. californicus augmentation. However, with integrated pest management of greenhouse-grown peppers, preventive release of N. californicus on transplants would be critical as it would minimize plant injury and prevent yield loss, while be ing compatible with biological control programs.

PAGE 17

17 CHAPTER 1 INTRODUCTION 1.1 Broad Mite in Greenhouse-Grown Peppers The broad mite, Polyphagotarsonemus latus (Banks) (Acari: Tarsonemidae), is a minuscule (approx. 150 m long) phytophagous mite that inhabits numerous plant species (Gerson, 1992; Jeppson et al ., 1975). The polyphagous mite is a pest in agronomic crops such as cotton ( Gossypium spp.), tea [ Camellia sinensis (L.) Kuntze], citrus ( Citrus spp.), beans ( Phaseolus spp.), papaya ( Carica papaya L.), and pepper ( Capsicum spp.). The mite can be found outdoors in most regions with warm climat e, and in greenhouses in temperate regions (Cross, 1979; Gerson, 1992; Ibrahim and Low, 1998; Jeppson et al., 1975; Ramakers, 2005; Silva et al., 1998). In Florida, P. latus is present in field and gr eenhouse crops, including fruits, vegetables, herbs, and ornamentals (de Coss-Rome ro and Pea, 1998; Jovicich et al., 2004a; Fan and Petitt, 1994; Fasulo, 2005; Olson and Si monne, 2006; Osborne et al., 1998; Pea and Bullock, 1995; Pea and Campbell, 2005). Pe pper crops grown in greenhouses frequently become infested with P. latus (Castagnoli and Falchini, 1993; C ho et al., 1996; Dik et al., 1999; Gerson, 1992; Ibrahim and Low, 1998; Laffi, 1982; Mizobe and Tamura, 2004; Nuez et al., 1996; Weintraub et al., 2003). My studi es focused on protecting bell pepper ( Capsicum annuum L.) from P. latus infestation at early crop stages in the greenhouse. Specifically, I investigated the use of augmentative releases of a predaceous mite (Acari: Phytoseiidae) as a biological control method for managing the pest. The warm temperatures that promote rapid cr op growth and lead to great fruit production in vegetable crops in greenhouses are also optimal for increasi ng phytophagous arthropod populations to levels that damage crops (Gullino et al., 1999). During most of the year, passively ventilated greenhouses in the tropics and subtropi cs act as a clear roof over the crop. Crops can

PAGE 18

18 be partially protected from incoming arthropod s when ventilation openings (roof-vents and all-around side walls) are covered with anti-i nsect screens. Despite barriers and crop management practices that are critical for minimizing the introduc tion of pests into greenhouses, pest arthropods such as P. latus eventually appear and become recurrent in pepper crops (Cho et al., 1996; Dik et al., 1999; Gerson, 1992; Ibra him and Low, 1998; Laffi, 1982; Mizobe and Tamura, 2004; Nuez et al., 1996; Weintraub et al., 2003). The appearance of P. latus on pepper crops at early plant developmental stages has been associated with dispersal mechanisms of the pest The small mite can in fest crops by air currents that carry them; insects that transport them [(i.e., whiteflies, Bemisia spp. (Homoptera: Aleyrodidae) and aphids (Hemiptera: Aphididae)]; crawling short distances from infested weeds or plants inside the greenhouse; or by humans who inadvertently carry them into the greenhouse (Fan and Petitt, 1998; Flechtmann et al., 1990; Gerson, 1992; Palevsky et al., 2001; Weintraub, 2003) (Figure 1-1). These sources of infestations typically occur in is olated patches of the transplanted crop. A more problematic source of in festation is the transplanting of seedlings that host a few P. latus These seedlings (which became infested earlier in the nursery or just prior to transplanting) may not exhibit da mage symptoms at transplanting In this situation, almost the entire pepper crop becomes rapidly infested and damage becomes vi sible shortly after transplanting (Cross, 1979; Dik et al., 1999; Hussey, 1985; Laffi, 1982; Ramakers, 2005; Weintraub, 2003). P. latus populations can increase very rapidly in a short period of time (Cho et al., 1996; Ho, 1991; Ibrahim and Low, 1998; Jeppson et al ., 1975; Silva et al., 1998 ; Weintraub et al., 2003). The mite has four life stages (i.e egg, motile nymph, quiescen t nymph, and adult) (Basset, 1985; Gerson, 1992) (Figure 1-2). Re production is arrhenotokous (unfertilized eggs

PAGE 19

19 leading to males), making possibl e that a whole populat ion develops from a single female that infests a plant (Gerson, 1992). Male-female sex ratio is approximately 1:4 (Gerson, 1992). Males have an important role in dispersing the pest as they carry unmated females (quiescent nymphs) to undamaged leaves. The length of th e developmental time (egg to adult) depends mainly on temperature, humidity and plant species (Gerson, 1992). In peppers, it can be as short as 3 to 4 days at 25oC and 80% RH (Almaguel et al., 1984; Ho, 1991; Ibrahim and Low, 1998; Lee et al., 1992; Li and Li, 1986; Li et al., 1985; Silva et al., 1998). Female P. latus can start laying eggs as soon as th ey reach adulthood and they live for up to 3 weeks. During this period they can lay from 20 to 40 eggs (Gerson, 1992; Ho, 1991; Ibrahim and Low, 1998; Silva et al., 1998), with more eggs laid per day when they are young adults (up to 2 eggs/day at ages from 4 to 12 days old) (Ibrahim and Low, 1998). Ibrahim and Low (1998) estimated that a P. latus population is capable of doubling within 1 week, continuing to double every 3 days (intri nsic rate of increase, rm = 0.293 at environmental conditions: 26 29oC, 65% RH and 12 h photoperiod). These aut hors estimated that if a mature female colonizes a pepper plant, the number of mites ca n reach 60 times (all mite stages) the initial population after 2 weeks of infestation. Ibrahim and Low (1998) suggested that a cont rol measure should be implemented within the first week of pest detection to curb the e xponential increase of the pe st population. However, P. latus inhabit cryptic sites on the terminal shoot s and underside of leaves, and detection of mites is difficult when only a few are presen t on plants (Cho et al., 1996; Gerson, 1992). There is no threshold for managing P. latus in pepper. When plants are scouted for pests in greenhouses, P. latus control with pesticides is started wi th the first detection of the mite on plants. The extent of the infestation throughout th e crop determines if control measures need to

PAGE 20

20 be localized or applied to the whole crop. T ypically, the pest is detected because injury symptoms are visible on plants (Cho et al., 1996; Dik et al., 1999; Gerson, 1992; Hussey, 1985; Jeppson et al., 1975; Nuez et al., 1996; Olson et al., 2005; Polack and Mitidieri, 2005; Weintraub et al., 2003). By the time firs t symptoms appear, hundreds of P. latus already inhabit the plant. When growers are not familiar with the pest, symptoms can be mistaken for a virus, plant nutrient toxicity, or damage from an herbicid e spray (in field crops) (Basset, 1985; Gerson, 1992; Jeppson et al., 1975). This makes the problem worse since pest management is delayed. Injury on plants becomes more severe as P. latus populations increase and the pest remains on the pepper plants (de Coss-Romero and Pea 1998). Young plants tissues (buds, leaves, flowers, and small fruits) are the pref erred feeding and re production sites for P. latus (Figure 1-3). The simple and sl ender chelicerae are unsuitable for penetrating the thick-walled or lignified tissue found in developed fruits old leaves, and stems (Basset, 1985; de Coss-Romero and Pea, 1998; Gui et al., 2001). Th erefore, pepper crops that become infested with P. latus at early growth developmental stages ar e at a high risk of being completely damaged if infestations are not managed timely (F igure 1-4). The cost of pepper production in greenhouse systems is high (Jovicich et al., 2005), and losses in early and total fruit production of colored fruit pose a significan t economic loss for growers. Until recently, most of the control recommendations for P. latus were based on the use of pesticides. Pesticides have been an effective control method for P. latus when pests are managed by this method only (Basset, 1985; Dik, 1999; Zhang, 2003). P. latus in greenhouse-grown bell peppers, and other vegetable crops (i.e. cucurbits), appear to be a more frequent and prevalent pest since the use of less toxi c pesticides and implementation of biological control practices (Basset, 1985; Dik, 1999; Hussey, 1985). In the pa st, broad spectrum pesticides used for pests

PAGE 21

21 such as T. urticae used to avoid outbreaks of P. latus (Basset, 1985; Ramakers, 2005). There are a few reports that suggest that P latus may have developed resistance to pesticides in some field crop scenarios (Davies, 1963; Vaissayre, 1986) A list of pestic ides used against P. latus in various countries and crops was report ed by Zhang (2003). For a long time, P. latus have been controlled inexpensively with sulfur sprays alt hough multiple sprays are required (Olson et al., 2005; Ramakers, 2005; Schoonhoven et al., 1978; Sm ith, 1933). In peppers, other commonly used pesticides are abamectin and dicofol, which also have to be applied more than once because these miticides do not kill P. latus eggs and because of the cryptic ha bitat of the pest (Nuez et al., 1996; Hussey, 1985; Olson et al., 2005; Ra makers, 2005; Weintraub et al., 2003). There are numerous examples of effective ma nagement of pests with biological control strategies in greenhouse-grown peppe rs without or with minimum us e of pesticides (Dik et al., 1999; Ramakers, 2005). However, pesticides ar e still needed for a number of phytophagous arthropods, especially when plant virus transm ission by insects is a concern (Ramakers, 2005). Moreover, even if natural enemies exist, pest icides are needed when large pest populations develop and biological control canno t manage the pest effectively and/or becomes very costly. This represents a challenge to implementing an integrated pest management (IPM) program that favors the use of biological control, as pesticides have to be selected to minimize the killing of introduced or naturally existent beneficial arth ropods and pollinators (and in addition, be among the few labeled for use on the plant species). Particularly when P. latus is detected after transplanting, management that in cludes repeated use of pesticides disrupts biological control programs for other pests in pepper (Basset, 1985 ; Dik et al., 1999; Koppe rt Biol. Syst., 2006). However, a few phytoseiids that are commonly released in greenhouse-grown vegetables for management of pests other than P. latus [such as for thrips (Thripid ae: Thysanoptera) or spider

PAGE 22

22 mites (Acari: Tetranychidae)] have been recently evaluated for management of P. latus in pepper plants (Fan and Petitt, 1994; Mizobe and Ta mura, 2004; Pea and Osborne, 1996; Weintraub et al., 2003). One such predatory mite is Neoseiulus californicus McGregor (Acari: Phytoseiidae) (Figure 1-5), which has been a ssessed in detached bindweed ( Convolvulus arvensis L.) leaves, lime [ Citrus aurantifolia Christm. (Swingle)] seedlings, and bean ( Phaseolus vulgaris L.) plants infested with P. latus (Castagnoli and Falchini, 1993; Pea and Osborne, 1996). In laboratory experiments, N. californicus populations were able to increase when preying on P. latus (Castagnoli and Falchini, 1993). However, the increase of N. californicus population feeding on P. latus was slower than that feedi ng on two-spotted spider mites [ Tetranychus urticae Koch (Acari: Tetranychidae)], the pred ators preferred prey (Castagno li and Falchini, 1993). Feeding on P. latus N. californicus developmental time (from egg to a dult) was approximately 6 days at 25oC, with females starting to lay eggs two days later (at a rate of near 2 eggs/day) (Castagnoli and Falchini, 1993). The male to female sex rati o is approximate to 1:1. In a scenario of abundant P. latus N. californicus could multiply its population about 1.2 times per day (Castagnoli and Falchini, 1993). Important characteristics of N. californicus are its tolerance to warm environments (33oC), the ability to survive on non-prey food (i.e. pollen) or, in absence or minimal food for relatively long periods ( 10 to 22 days) (Castagnoli and Simoni, 1991, Castagnoli and Simoni, 1995, cited in Griffiths, 1999; de Courcy Williams et al., 2004; Gotoh et al., 2004; Palevsky et al., 1999; Wal zer et al., 2001). These attribut es of the predator appear to be important for managing P. latus in preventative releases and low release densities of the predator (i.e. low cost). Ho wever, it is not yet known if N. californicus is effective for managing P. latus in young pepper plants grown under greenhouse conditions.

PAGE 23

23 It is important to protect pepper plants from P. latus during early plant developmental stages. Introducing N. californicus before and/or after transplan ting can be an effective method for eliminating P. latus or for maintaining their densities at non-damaging levels on plants. In the studies presented in the following chapters, the predaceous mite N. californicus was evaluated for its effects on pest population, seedling and plant grow th, and early fruit yield as affected by density and timing of predator rele ase when infestations occurred during seedling development. My first study (Chapter 2) inve stigated the effectiveness of N. californicus in producing undamaged P. latus -free pepper transplants. The st udy estimated relationships between transplant growth and visual damage as affected by the cumulative number of P. latus present on seedlings over time. I chose 16 scenarios where pe st infestations and pred ator releases occurred at various times during seedling development. Measurements taken durin g transplant production included changes in P. latus and N. californicus populations on seedlings, visual damage on seedlings, and seedling growth variables. This study was conducted under controlled environmental conditions that were optimal for pest development. Two following studies were conducted with pepp er plants transplanted in a passively ventilated greenhouse (Chapter 3). The main obj ective of these studies was to evaluate the effectiveness of N. californicus released for management of P. latus in scenarios where pepper seedlings became infested just before transplantin g. Various pest-predator scenarios were tested and compared with sulfur spray programs, non-managed infestations, and no infestations. The effectiveness of P. latus management by N. californicus was evaluated using estimates of pest population density, cumulative presence of the pe st on the plant, plant visual damage, plant growth variables and marketable fruit yield. A dditional objectives of these studies were to

PAGE 24

24 examine the economic viability of the pest manage ment scenarios, and to identify relationships among plant growth variables and yield as affected by the cumulative presence of P. latus on plants over time. 1.2 Greenhouse Production of Bell Pepper in Florida Production of bell peppers in passively ve ntilated greenhouses differs from field production (Jovicich et al., 2004a). Fruit in green houses are typically harvested when they ripen to a full red, orange, or yellow color, dependi ng on genotype. In the U.S., the high-quality commodity attracts high prices in terminal market s supplied almost entirely with imported fruit. In the field, delaying a harvest of green peppers until fruit ripen is risky for growers in subtropical and tropical regions. Outdoor high temperatures comb ined with frequent rainfall, insect and disease pressure during the fruit ripening period, and lower yields of colored compared to green fruit in field crops, has meant that pepper fruit in Flor ida fields (7500 ha) are harvested at their mature green stage of devel opment (U.S. Dept. of Agriculture, 2007a). There is, however, a small greenhouse indu stry that produces colored be ll peppers in Florida (Jovicich et al., 2004a; Mitchell and Cantliffe, 2005; Tyson et al., 2004). Research on the production of peppers in greenhouses for warm climate regions ha s been one of the objec tives of the Protected Agriculture Project since 1997 (Cantliffe, 1999, The Protected Agriculture Project. Hort. Sci. Dept., Univ. of Florida, Gainesville, Fla. ; Jovicich and Cantliffe, 2004; Jovicich et al., 1999; 2004a; 2004b; 2005; 2007) There was a lack of information regarding market information and th e potential profitability in the business of growing colored bell peppers for new or existing gr owers in Florida. I a ddress this by reviewing bell pepper imports and historic al market prices, and estimating costs and returns for a small greenhouse operation (Chapter 4). Th e analyses presented in this chapter provide information to

PAGE 25

25 assist growers and agricultural investors in evaluating greenhouse pepper production as a potential business opportunity in Florida. Figure 1-1. Phoretic re lationship between broad mite and whitefly. Six P. latus were counted on the silverleaf whitefly [ Bemisia argentifolii Bellows & Perring (In secta: Homoptera: Aleyrodidae)] (approximately 1.2-mm long), which was on a melon plant ( Cucumis melo L.) densely infested with P. latus

PAGE 26

26 Figure 1-2. Life stages of P. latus The time from egg to adult (passing through motile nymph and quiescent nymph) can be as short as three to four days.

PAGE 27

27 Figure 1-3. Damage by P. latus on bell pepper shoots and fruit. Early symptoms (top), necrosis of terminal shoot (middle), and dama ge on fruit pod and peduncle (bottom).

PAGE 28

28 Figure 1-4. Damage on bell plants 30 days afte r transplanting infested seedlings with two P. latus P. latus -free plants, 85-cm tall (left) and plants that were infested with two P. latus 3 days before they were tran splanted in a greenhouse (right).

PAGE 29

29 Figure 1-5. Neoseiulus californicus on pepper leaves. A) Young adult and B) older adult mite feeding on P. latus and C) adult mites feeding on T. urticae (C photo taken by L.S. Osborne), and adult mite hidde n under leaf domatia (D). AB C D

PAGE 30

30 CHAPTER 2 MANAGEMENT OF Polyphagotarsonemus latus (Banks) (ACARI: TA RSONEMIDAE) WITH Neoseiulus californicus McGregor (ACARI: PHYTOSEIIDAE) ON BELL PEPPER ( Capsicum annuum L.) TRANSPLANTS 2.1 Introduction Broad mite [ Polyphagotarsonemus latus (Banks)] is a polyphagous phytophagous arthropod that frequently infests bell peppers ( Capsicum annuum L.) grown in greenhouses in regions with tropical or subtropi cal climates. The minute mite is transported by air currents, has phoretic associations with whiteflies [ Bemisia tabaci (Gennadius) and B. argentifolii Bellows & Perring; Insecta: Homoptera: Aleyrodidae], an d may be unknowingly trans ported by humans into greenhouses where transplants or fruits are pr oduced (Fan and Petitt, 1994, 1998; Flechtmann et al., 1990; Gerson, 1992; Palevsky et al., 2001; Soroker et al., 2003). P. latus has a body length of circa 150 m and typically inhabits the abaxial side of young leaves and cryptic sites of the terminal shoots (Ho, 1991; Jeppson et al., 1975). Th e mite is difficult to detect and can initially colonize plants at low densities (Fan and Petitt, 1994; Laffi, 1982; Weintraub et al., 2003). After a few seedlings become infested in a transplant nursery, P. latus populations rapidly increase (Cho et al., 1996; Ho, 1991; Ibrahim and Low, 1998; Jeppson et al., 1975) and mites disperse to neighboring seedlings (Cross, 1979; Fan and Pe titt, 1994; 1998; Laffi, 1982). The infested transplants may be damaged and unusable or, still asymptomatic, carriers of P. latus to production greenhouses (Cho et al ., 1996; Laffi, 1982; Weintraub et al., 2003) or the field. P. latus inserts stylet-lik e chelicerae (approx. 43 m long) to young tender plant tissues and extracts cell contents of the top cell layers (Grinberg et al., 2005; Gui et al., 2001). Whole young pepper seedlings exhibit damage symptoms th at are similar to those in shoots of older plants. As the number of mites increase on a te rminal shoot, unfolding leaves exhibit symptoms such as upright blade elongation with cupping upwa rds, zigzagged vein pa tterns, thickening of

PAGE 31

31 the leaf blades, and necrosis (Cho et al., 1996; de Coss-Romero and Pea, 1998; Jeppson et al., 1975). Typical green color tones become da rker, purplish, and ultimately bronzed and dark-brown (Cho et al., 1996; Cross, 1979; Gers on, 1992). The injured shoot apex, axillary buds, and leaf primordia preclude further shoot gr owth, subsequently leav ing the seedling with a stunted appearance. In peppers a nd other crops these differences in symptoms have been used to develop scales with damage indices (de CossRomero and Pea, 1998; de Holanda et al., 1992; Pea, 1990; Pea and Bullock 1994). These sc ales may be used when developing pest management strategies and estim ating loss of production. Moreove r, if pest abundance is known over time, a variable known as cumulative mite -days can be calculated (de Coss-Romero and Pea, 1998) and used to study the ef fects of the additive presence of P. latus on plant growth variables (i.e. leaf area number of leaves, dry weight, and stem length). Undamaged, pest-free transplants are require d for optimum crop establishment and to minimize pests at early plant developmental stages With pepper, a complete loss of fruit yield can occur if P. latus populations are allowed to increas e in transplanted seedlings (de Coss-Romero and Pea, 1998; Jeppson et al., 1975; Mizobe and Tamura, 2004; Weintraub et al., 2003). To control P. latus repeated sprays with sulfur, ab amectin or dicofol are started as curative programs once damage symptoms are per ceived on seedlings or plan ts, or as preventive programs in greenhouses where infestations are r ecurrent (Cho et al., 1996; Dik et al., 1999; Gerson, 1992; Jeppson et al., 1975; Nu ez et al., 1996; Olson et al., 2005 ; Weintraub et al., 2003). However, the use of such pesticides can disrupt biological control programs in greenhouse-grown peppers; therefore, alte rnatives to pes ticide sprays on P. latus are needed when managing crop pests with minimum use of pe sticides or following standards for certified organic production (Brown and Jones, 1983; Dik et al., 1999; Griffiths, 1999; Hussey, 1985;

PAGE 32

32 Koppert Biol. Syst., 2006; Ramakers, 2005; U.S. Dept. of Agriculture, 2007b). Castagnoli and Falchini (1993) reporte d that the phytoseiid Neoseiulus (=Amblyseius) californicus McGregor has the ability to feed and reproduce on P. latus on detached bindweed ( Convolvulus arvensis L.) leaflets, and suggested that the pr edator be evaluated in vegetabl e crops as a biological control agent of the tarsonemid in greenhouses. Pea and Osborne (1996) released N. californicus on bean ( Phaseolus vulgaris L.) plants, after a single leaf in each plant was artificially infested with P. latus at average densities of 367 mites/leaf, and us ed prey-predator density ratios of 12:1 and 14:1 (calculated as prey counts on a leaf and number of predators released pe r plant). Fourteen days later, P. latus densities were reduced to non damagi ng mite densities of 19 mites per sampled leaf. Similarly, on lime [ Citrus aurantifolia Christm. (Swingle)] seedlings P. latus densities of 133 mites/leaf were reduced to one mite per leaf 14 days after N. californicus was released at prey-predator density ratios of 5:1 to 15:1 (Pea and Osborne, 1996). P. latus populations in infested pepper plants were also managed effectively when using multiple releases of other phytoseiids such as Neoseiulus cucumeris (Oudemans) (Acari: Phytoseiidae) (Mizobe and Tamura, 2004; Weintraub et al., 2003) and Neoseiulus barkeri Hughes (Acari: Phytoseiidae) (Fan and Petitt, 1994). The latter predatory m ite is not commercially available in the U.S. (Osborne, L.S., 2007, Biological control of foliar pests. Mid-Florida Research and Education Center, Univ. of Florida, Apopka, Fla. ). N. californicus is commercially reared in the U.S. and is released, mainly on a preventative basis, for management of two spotted spider mites [ Tetranychus urticae Koch (Acari: Tetranychidae)] in greenhouse-gr own vegetables and ornamental s (Gillespie and Raworth, 2005; Griffiths, 1999; Osborne et al., 1998; 2005; Rama kers, 2005). The predator is considered a

PAGE 33

33 viable candidate for biological c ontrol in warm protected environm ents since, relative to other phytoseiids used against spider mites, it is tolerant to temperatures up to 33oC (Castagnoli and Simoni, 1991, Castagnoli and Simoni, 1995, cited in Griffiths, 1999; Gotoh et al., 2004). It is also capable of surviving on pollen (Casta gnoli and Simoni, 1999) and fungal hyphae in the absence of prey, provided that water is present (de Courcy Williams et al., 2004). In transplant nurseries where P. latus problems are frequent, infestations can occur at any seedling developmental stage. Preventive introd uction of predators in transplant nurseries represents a possible method to eliminate P. latus or maintain their densities at non damaging levels. The purpose of this study was to provide information that will aid in determining the effectiveness of N. californicus to produce undamaged, P. latus -free pepper transplants. This was achieved by measuring a nd comparing changes in P. latus and N. californicus populations, visual damage on seedlings, and seedling grow th during a pepper transplant production period when pest infestations and predator releases occurred at various times during seedling development. Prey-predator scenarios were al so used to estimate relationships between transplant growth and visual damage as affected by the cumulative presence of P. latus over time. 2.2 Materials and Methods 2.2.1 Treatments Sixteen scenarios in which pest and predat or were present on pe pper seedlings during a transplant production period were investigate d. Prey-predator scenarios had artificial infestations initiated with two P. latus placed on seedlings at growth developmental stages a) cotyledons unfolded, with predators released 0, +15, or +24 days from initial infestation (DFI), b) two leaves unfolded, with predators released 15, 0, or +9 DFI, or c) four leaves unfolded, with predators released 24, or 0 DFI (Tables 2-1 and A-1). Additionally, separate groups of

PAGE 34

34 seedlings remained mite-free (noninfested control), and other groups were either infested with P. latus only or had releases of N. californicus only at the three seedling developmental stages. With pest and predators introduced on the same day, infestation was carried out in the morning and N. californicus released in the evening. The selected transplant growi ng period was 42 days, from seeding to a mean of seven expanded leaves in mite-free seedlings. The terms preventive or curative are used to indicate if N. californicus were released on seedlings days before or after, respectively, P. latus were placed on seedlings. 2.2.2 Plant Growing System, Environment, and Irrigation Seedlings of bell pepper Leg ionnaire (Rogers-Syngenta, Boise, Idaho) were grown on a 70% peatmoss : 30% vermiculite (v/v) substrate mix (Pro-mix 0463, Premier, Qubec, Canada) in 32-cell (top width top length height: 3.5 3.5 6.5 cm; 35-mL/cell) polystyrene transplant flats sliced from larger commercial nu rsery flats (Speedling Inc., Sun City, Fla.). To ensure mites would not disperse outside the transpla nt flats, the flats were placed in the center of a polyethylene tray (width length height: 23 50 5 cm) that contained water with a couple drops of household bleach to avoid algae developm ent. Seedling roots did not grow into the water as the bottom of the flat was kept elevated 3.5 cm above the water surface level. Seedling flats were placed on the floor, wh ere air movement was negligible to air-transport mites across flats (Jung and Croft, 2001; Palevsky, 2001). A total of 16 trays with transplant flats were replicated three times in a room retrofitted as a walk-in plant growth chamber (width length height: 3 3 3 m; Aluma-Shield, Miami, Fla.) with a controlled environm ent. Light period was 16 h, with mean irradiance of 220 20 mols-1m-2 (PAR) from a combination of eight 2-m-long fluorescent tubes (Sylvania F96T12/CW/SS; 60W) and two high-pressure sodium lamps (Philips Ceramalux ED-37;

PAGE 35

35 1000W). During light (16 h) and dark (8 h) pe riods, the mean air temperatures were 24 1oC and 19 1oC, and air relative humidity 70 10% and 80 10%, respectively (measurements at 5 cm above the seedlings). After the cotyledons unfolded, seedlings were irrigated with a solution with nutrient concentrations (mgL-1) of NO3-N: 70, P: 50, K: 100, Ca: 90, Mg: 40, S: 56, Fe: 2.8, Cu: 0.2, Mn: 0.8, Zn: 0.3, B: 0.7, and Mo: 0.06. Nutrient solu tion was delivered (30 mL/seedling) every second day, using a dispenser made of multiple pi pettes and directing th e solution to the media plug cells without wetting cotyledon s or leaves. After irrigating a flat containing seedlings, the dispenser was submerged in hot water (80oC) to avoid contamination of mites among treatments. 2.2.3 Source of Mites, P. latus Infestation, and N. californicus Release A P. latus colony, initiated with a few mites from a citrus nursery at the Univ. of Florida, Gainesville, Fla., was maintained on bell pepper cv. Legionnaire seedlings grown in a growth chamber (L:16 h, 200 10 mols-1m-2, 25 1oC, and D:8 h, 21 1oC). Non-infested pepper seedlings were added weekly to augment the P. latus stock colony. N. californicus in 1000-mites vials were shipped overnight fr om a commercial rearing facility (Biotactics, Romoland, Calif.) and arrived in Gainesville, Fla. the day before they were released. Seedlings emerged 11 days after seeding ( DAS), and non-infested plants had unfolded cotyledons on 13 DAS, two unfolded leaves on 28 DAS, and four unfolded leaves on 37 DAS. On selected dates, mites were introduced onto each seedling in the transplant flat of the corresponding treatment (Tables 2-1 and A-1). Tw o gravid female mites of one of the mite species were selected under a stereomicroscope and transferred to the top leaves with the aid of a single-hair brush.

PAGE 36

36 2.2.4 Plant Sampling and Mite Recovery Seedlings with no mites (non-infested control) were sampled every third or fourth day from the day cotyledons unfolded until 42 DAS. In all other treatments, seedlings were sampled for the first time on the day N. californicus or P. latus were introduced. Depending at which seedling developmental stage mites were first intr oduced, treatments had a to tal of eleven, six, or three sampling dates. Two seedlings per flat we re randomly selected, severed at the surface of the plug cell media, and immediately transfer red into 50-mL centrifuge tubes (no. 05-539-6, Fisher Scientific, Pittsburgh, Pe nn.). Tubes were then filled with 40 mL of a 70% ethanol solution and capped to preserve seedlings and mites until mites were recovered and counted. To recover the mites, seedlings were first held in place with a pin at th e top of a perforated tube cap, kept inside the 50-mL tube, and then centrifuged at 861 g (2500 rpm) for 10 min (J2-21M/E, Beckman, Fullerton, Calif.). Holding the seedling from the bottom-end of the stem, stem and leaves were rinsed w ith the ethanol solution inside th e same or, if necessary, a second centrifuge tube (for large seedlings ). Mites in the recovered solu tion were allowed to settle to the bottom of the tube for a day. To reduce the solution volume, the supe rnatant 45 mL mite-free solution was then slowly siphoned out from the tube through a screen (pore side: 45 m) using a vacuum apparatus. The solution left in the ce ntrifuge tube contained the mites and was poured and tube walls rinsed into a 6-cm diameter Petri dish (no. 08-757-13A, Fisher Scientific, Pittsburgh, Penn.). With the Petri dish placed over a black paper with 3 2 mm white grids, mites of both species were counted (both sexes a nd all life stages with the exception of eggs) using a stereomicroscope (Leica MZ16, Bannockbur n, Ill.) set to 40 ma gnification (Figure A-2).

PAGE 37

37 Counts of P. latus and N. californicus are presented as number of recovered mites per seedling. Initial and final prey-preda tor density ratios were calculated. P. latus abundance over time is expressed as cumulative mite-days (de Coss-Romero and Pea, 1998), calculated by averaging the count of P. latus per seedling and multiplying by the time between sampling dates: i i i it n n0 12 where ni is the number of P. latus at the i th sampling date after initial infestation, ni+ 1 is the number of P. latus at the subsequent sampling date, and t is interval period (in days) between two consecutive sampling dates. One mite-day represents one mite (any motile stage) on a seedling during one day. For each treatment, the P. latus cumulative mite-days (CMD) were calculated until 42 DAS, which corresponds to commercial transplants ready for greenhouse or field establishment. 2.2.5 Estimation of Plant Damage and Measurement of Seedling Growth A visual damage scale based on symptoms developed in P. latus -infested pepper seedlings (pest not controlled) was develope d to estimate damage on seedlings (Figure 2-1). The scale was similar to others previously used on pepper pl ants and other crop species (de Coss-Romero and Pea, 1998; Echer et al., 2002; Mizobe and Tamura, 2004; Pea and Bullock, 1994). Seedling damage indices were comprised of ascending number s assigned to stages with increased severity of damage symptoms at the terminal shoot regi on (apical bud and 3 upper leaves, if present). Sampled seedlings were assigned one of five levels (Figure 2-1): 0. No injury: youngest leaf flat and at an approximate upright angle range 10oo 1. Light injury: youngest leaf narrower, light gr een color, and at an upright angle >30o 2. High injury, youngest leaf starting to curl downwards or upwards; dull green color on youngest leaves 3. Severe injury: at least two younge st leaves curled with central leaf veins in a crisscross pattern; dull gr een color on youngest leaves

PAGE 38

38 4. Very severe injury: youngest leaves thickene d, not expanded, with centr al leaf veins in a crisscross pattern, and possibly with necrosis of terminal growth bud; basal leaves thickened and may appear shiny. Plant growth variables, assessed after mites we re recovered from seed lings were number of leaves longer than 1.5 cm, leaf area (LI-3100 area meter, LI-COR, Lincoln, Nebr.), stem length, and dry weight (measured after an ov en-drying period of four days at 70oC). 2.2.6 Experimental Design and Data Analysis Control efficacy by N. californicus in each of the imposed pest/predator/seedling-age scenarios was evaluated by comparing P. latus number, CMD, seedling growth variables, and damage index at selected sampling dates during the transplant growing period. Treatments (as individual trays with seedlings) were randomized w ithin three blocks that divided the floor area of the walk-in growth chamber. A complete randomized block design with the time variable (sampling days after the seedlings were seeded ) was analyzed using a mixed model with PROC MIXED procedure in SAS (SAS Inst., 1999). Treatment, time, and the treatment time interaction were fixed effects and blocks were considered to be random. Comparison of covariance structures with a goodness of fit cr iteria was done by PROC MIXED and correlation between time points was specified as an autoregre ssive process with either a lag period of one [ar(1)] or with heterogeneous variance [arh(1)] (Littell et al., 1996). The mixed model procedure provides Type III-based F -values but does not provide means square values for each element within the analysis or error terms. Means of measured variables were separated after selected pairwise contrasts using the TukeyKramer test (Kramer, 1957) at = 0.05. The PDMIX800 SAS macro was used to assign letter groupings to the PROC MIXED LSMeans output (Saxton, 1998). Mite number and CMD were log-transformed ( y = ln [ x +1]) to normalize data distribution prior to analysis. Non-transformed means and sta ndard errors of the mean s (SEM) are presented. Linear and non-linear regression analyses (SigmaPlot; SPSS Science, 2002) were performed to

PAGE 39

39 estimate relationships between CMD and transp lant growth variable s (and their reductions relative to non-infested seedli ngs) and damage index in 42-days old seedlings. The analyses used pooled data from treatments and blocks that had either infestations initiated on unfolded cotyledons, two unfolded leaves, or four unfol ded leaves, or from all treatments combined. 2.3 Results There were significant treatment time interactions ( P < 0.001) for mite numbers, CMD, seedling growth variables, and visual damage i ndices. Selected means (Tables 2-2 and 2-3) and all means of the measured variables are presented (Figures 2-2). In all treatments, on the first day mites were introduced to seedlings, densities of P. latus or N. californicus were not significantly different ( P = 1.000 in both mite species). On th ese days the per cent of recovered mites in relation to the two mites introduced was 55 for P. latus and 65 for N. californicus 2.3.1 Polyphagotarsonemus latus Infestations with Absence of N. californicus Treatments PC P2 and P4 On 22 DAS, nine days after infestation on cotyledons ( PC ), a mean damage index of 2.3 was significantly different from the index zero (0) for the first time in non-infested seedlings ( P < 0.001; Figure 2-4). That same day, each seedling had 94 26 P. latus and 347 43 CMD (Figures 2-2A, 2-3). On 42 DAS the transplants hosted 242 110 P. latus and had reached 5336 1070 CMD (Table 2-2 and Figure 2-3). Transplants had a mean damage index of 4.0, and as compared with non-infested seedlings (index 0), had a dry we ight reduction of 65.2% (101 vs. 290 mg; P = 0.022), stem length by 59.7% (58 vs. 144 mm; P < 0.001), leaf area by 76% (18 vs. 75 cm2; P < 0.001), and number of leaves by 47.8% (3.5 vs. 6.7; P = 0.033) (Table 2-3).

PAGE 40

40 When P. latus was introduced on seedlings with two unfolded leaves ( P2 ), a mean visual damage index of 1.5 was observed 9 days after th e initial infestation ( 37 DAS); the first time damage index was significantly differe nt from non-infested seedlings ( P = 0.008; Figure 2-4). On 37 DAS, seedlings hosted 102 15 P. latus and had accumulated 295 27 mite-days (Figures 2-2A, 2-3). By 42 DAS, transplants had a mean of 1551 149 CMD, which was not significantly different from infest ations initiated on cotyledons ( P = 1.000; Table 2-2). Transplants infested at unfolded two leaves ha d a mean damage index of 1.7 and significantly shorter stems than non-infested one s (109 vs. 144 mm; a 24.3% reduction; P = 0.043; Table 2-3). However, even without predatory mites, th ese transplants had signif icantly longer stems, larger leaf areas, and smaller damage indices than transplants that had been infested initially at the unfolded cotyledon stage ( P = 0.009, < 0.001, and < 0.001, for stem length, leaf area, and damage index, respectively; Table 2-3). When infestations were initiat ed at four unfolded leaves ( P4 ), 42-d old transplants accumulated only 34 4 mite-days, had a mean damage index of 0.3, and had growth variables that were not significantly different th an non-infested seedlings (all with P > 0.985; Tables 2-2 and 2-3). However, these transplants hosted 15.7 2.2 P. latus (Table 2-2 and Figure A-2). 2.3.2 Polyphagotarsonemus latus Infestations with Releases of N. californicus 2.3.2.1 Seedlings with Unfolded Cotyledons Infested with P. latus Treatments PC-NC PC-N2 and PC-N4 When the P. latus infestation was initiated on se edlings with unfolded cotyledons (13 DAS) and N. californicus were released that same day ( PC-NC ), P. latus number first increased significantly from 13 DAS to 34 DAS (0.7 0.2 vs. 107 96 mites per seedling; P = 0.002) but later, at the end of the transplant production period, de creased after an increase in the number of N. californicus (Figure 2-2C). The number of P. latus on 42 DAS was 12.8 3.3

PAGE 41

41 mites per transplant, and prey-predator density ratio was 7.1:1 (Figure 2-2C and Table 2-2). Although CMD were 982 741 on 42 DAS (Table 2-1) damage index on the terminal shoot (0.3) and transplant growth variables at that time were not signifi cantly different from non-infested seedlings (all with P > 0.914) (Table 2-3). When N. californicus was released 15 or 24 days after P. latus infestations were initiated on cotyledons ( PC-N2 and PC-N4 respectively), prey-predator density ratios were 116:1 and 387:1 at the time of release (Table 2-2). P. latus on seedlings had 1158 187 CMD (15 days delayed release) and 4302 998 CMD (24 days de layed release), which had created seedlings with damage indices of 3.0 0.2 and 4.0 0.0, resp ectively (Figures 2-3, 2-4). Throughout the transplant growing period, thes e curative treatments decreased P. latus populations (Figure 2-2DE and Table 2-2). Nonetheless, these tr ansplants did not recover from the early damage caused by large P. latus populations (Figures 24 and Table 2-3). S eedlings with delayed releases of N. californicus (15 and 24 days after the infestat ion) had leaf areas and dry weights significantly smaller than non-infested seedlings shortly after the releas e of predatory mites (release delayed 15 days had P = 0.001 on 34 DAS, and P = 0.001 on 37 DAS, respectively; delayed 24 days had P < 0.001, and 0.001 on 37 DAS, respectiv ely; Figures 2-5, 2-6). Compared to non-infested transplants, transplants with N. californicus released 15 days after the P. latus infestation had significantly reduced stem length (40.3%; 86 vs. 144 mm; P = 0.001) and leaf area (60.0%; 30 vs. 75 cm2; P = 0.010), and had a mean damage index of 1.8 (Table 2-3). The release of N. californicus that was delayed 24 days from the infestation initiated on cotyledons resulted in transplant growth vari ables and damage index (4.0) that were not significantly different from an infest ation initiated on cotyledons without N. californicus ( P = 1.000; Table 2-3).

PAGE 42

42 2.3.2.2 Seedlings with Two Unfolded Leaves Infested with P. latus Treatments P2-NC P2-N2 and P2-N4 With infestations initiated on seed lings with two unfolded leaves and N. californicus released preventively when cotyledons unfolded (15 days earlier; P2-NC ), or N. californicus released on the same day the infestation was initiated ( P2-N2 ), P. latus densities remained at less than 7 mites per seedling during the transplant production period (Figure 2-2FG). On 42 DAS, the two treatments had CMD (43 7.4 and 83 7.3, respectively) that we re not significantly different ( P > 0.050), and had prey-predator density ratios of 4.0:1 and 5.2:1 (Table 2-2). During the seedling growing period, growth variables and damage indices in these two treatments were not significantly different from noninfested transplants (all with P > 0.050) (Figures 2-4). Release of N. californicus on cotyledons, or on the same day when P. latus infested seedlings with two unfolded leaves, led to 42-days old transplants with no visual damage and growth variables that were not signi ficantly different from non-infe sted seedlings (all with P > 1.000; Table 2-3). With P. latus infestations initiated on seedlings w ith two unfolded leaves and a delayed N. californicus release of 9 days ( P2-N4 ), P. latus density had, by 37 DAS, increased significantly to 92 16 mites per se edling from 1.7 0.2 mites on 28 DAS ( P < 0.001; Figure 2-2H). When N. californicus were released, prey-predator de nsity ratio was 54:1 and CMD had significantly increased to 328 66 ( P < 0.001; Table 2-2 and Figure 2-3). On 42 DAS, this delayed release resulted in CMD (1359 214) that were not signifi cantly different from uncontrolled P. latus infestations initiated on seedlings with two unfolded leaves, or unfolded cotyledons (both with P > 1.000; Table 2-2 and Fi gure 2-3). Damage reached an index of 2.5 and, as compared to non-infested transplant s, stem length was reduced by 38.2% (89 vs.

PAGE 43

43 144 mm; P < 0.001) while other growth variables we re not significantly different (all with P > 0.970; Table 2-3). 2.3.2.3 Seedlings with Four Unfolded Leaves Infested with P. latus Treatments P4-NC P4-N2 and P4-N4 When N. californicus were released preventively, 9 or 24 days before an infestation on seedlings with four unfolded leaves ( P4-N2 and P4-NC respectively), or with a release on the same day of the infestation ( P4-N4 ), P. latus densities did not exceed 4 mites per seedling (Figure 2-2IK). Cumulative mite-days on 42 DAS did not differ significantly among the three treatments ( P > 0.050) and were smaller than 25 mite-day s per transplant (Table 2-3 and Figure 2-3). On 42 DAS, P. latus and N. californicus were present on the seedlings of respective treatments at prey-predator de nsity ratios of 2.5:1, 1.4:1, and 1.2: 1 (Table 2-3). Transplant growth variables and damage indices in these three treatments were not significantly different from non-infested seedlings or from seedlings th at had been infested at 4-leaves and left uncontrolled (all with P > 1.000; Table 2-3 and Figures 2-4). 2.3.3 Neoseiulus californicus Released with Absence of P. latus Treatments NC N2 and N4 On 42 DAS, a few N. californicus were recovered from transplants that had N. californicus released at the three seedling growth de velopmental stages in the absence of P. latus However, these final N. californicus densities were not significantly different from zero mite number in non-infested transplants ( N. californicus was never recovered from non-infested seedlings) ( P > 0.050) (Table 2-2). Densities of N. californicus in these three release times were not significantly different from each othe r on their release day or on 42 DAS ( P = 1.000; Table 2-2). Moreover, densities of the predatory mite were similar on the release day compared to 42 DAS (all with P > 0.899; Table 2-2 and Figure 22B). Transplants where only N. californicus were

PAGE 44

44 introduced had growth variables that were not significantly different from non-infested transplants (all with P > 0.854; Table 2-3 and Figures 2-5). 2.3.4 Responses of Transplant Growth Variables and Damage Index to P. latus Cumulative Mite-Days Relationships between 42-days old transplant growth variables and damage index and CMD were calculated for grouped treatments which had P. latus infestations initiated on seedlings with unfolded cotyledons, two unfolded leaves, or four unfolded leaves. Cumulative mite-days reached greater values when P. latus infestations were initiated on seedlings with unfolded cotyledons as compared with two or f our unfolded leaves (Figure 2-9, Table 2-4). When P. latus infestations were initiated on seedlings with unfolded co tyledons, exponential rise-to-maximum functions were si gnificant for all growth variab les measured and damage index (all with P < 0.001; Figure 2-9 and Table 2-4). With these infestations occurring on cotyledons, up to 7301 CMD could be reached after an infesta tion period of 29 days. After this period, and relative to non-infested transplants, these transp lants had reductions of 69% of dry weight, 80% of leaf area, 49% of leaf numb er, and 62% of stem length; and damage index reached 3.8. When infestations were initiated on seedlings with two unfolded leaves, up to 1828 CMD were reached after a 14-day period, which led to reductions of up to 51% in dry weight and 24% in leaf number, both in linear relationships Damage index reached 2.4. With mite-days at levels up to 1828, dry weight, and leaf numb er in transplants where P. latus infestations were initiated on two unfolded leaves were reduced to a less extent th an when infestations were initiated on unfolded cotyledons (Figure 2-9). P. latus infestations initiated in seedli ngs with four unfolded leaves had up to 41 CMD in a 5-day period, and had no signi ficant responses for relative reduction of transplant growth variables (overall means < 4%) and damage index (overall mean < 0.07; all with P > 0.100; Figure 2-9 and Table 2-4).

PAGE 45

45 After pooling data from all treatments, responses for growth variables and damage index as affected by CMD were calculated for 42-days old transplants. Measurements of transplant growth on 42 DAS and their CMD had significant exponential decay relationships (all with P < 0.001; Figure 2-10). With y = a e-bx, y was the growth variable measured, a the mean growth variable in mite-free transplants, and b the decay factor determining how rapidly the growth variable decreased ra pidly with increased CMD ( x ) (Figure 2-10). Based on the values of b leaf area and dry weight d ecreased with the increase of P. latus abundance followed in descending order by stem length and leaf numbe r (Figure 2-10). With visual damage of transplants, the increase of indices was f it to an exponential rise to a maximum, y = a (1e-bx) (Figure 2-10). This relationshi p indicated that even in transp lants rated with minimum visual damage, such as with a level of one, transplant growth variab les could have been reduced by P. latus infestation. In Figure A-3, the same data plotted in Fi gure 2-10 is presented but with CMD in a log-scale (equations presented in Table A-1). In Figure A-4, seedling growth reductions (data grouped from all treatments) as affected by CMD we re plotted as percentage values relative to non-infested seedlings (equations presented in Table A-2). 2.4 Discussion The level of damage in peppe r transplants attributed to P. latus depended on the plant developmental stage at infestation, the length of the infestation period, and the presence of N. californicus With no pest management, initial densities as low as two female P. latus infesting either unfolded cotyledon s or unfolded two leaves increas ed rapidly to peaks of up to 370 mites/seedling during seedling development a nd led to CMD as high as 5336 in 42-days old transplants. These transplant s were seriously damaged and unusable for planting. When P. latus infested seedlings with cotyle dons unfolded, the population of P. latus decreased after 37 DAS.

PAGE 46

46 This may have been the result of both overcro wding and seedlings becoming an unsuitable plant host (due to necrosis of the shoot) fo r the mites to feed on (Gerson, 1992). P. latus left the seedlings (mites were observed walking on the plug tray) and may have died from starvation. In a real crop scenario (without a water barri er around the transplant trays), these P. latus would have dispersed to neighboring plants. By contra st, when infestations occurred shortly before an assumed transplanting date (e.g., five days befo re the 42-day transplant age), seedling damage from 34 CMD was visually undetectable. However, if only a few mites are present in transplants (e.g., 16 mites/seedling) soon after the infestation is initiated (damage is not visible yet), P. latus are difficult to detect unless plan ts are carefully inspected with a magnifying lens. For the mean population to be 16 mites/seedling after five days, suggested that the two female P. latus laid 3 to 4 eggs/day in the first days and that developmen tal time was probably as s hort as three or four days (Almaguel et al., 1984; Ho, 1991; Ibrahim and Low, 1988; Lee et al., 1992; Silva et al., 1998). This pest problem on seedlings becomes evident only after tran splanting in a fruit production greenhouse (Cho et al., 1996; Weintra ub et al., 2003), where yield reductions in plants with early uncontrolled in festations are greater than in plants where infestations were initiated after fruits had set (de Coss-Romero and Pea, 1998). In this study, it was demonstrated that the commercially reared predatory mite N. californicus was effective for producing 42-days old undamaged pepper transplants with seven leaves when two gravid female predators/se edling were released from 0 to 24 days before the day that seedlings were infested with two P. latus Under these scenarios, initial prey-predator density ratios (fir st time both mite species were pr esent on seedlings) were 3.3:1 or less, and P. latus populations were maintained at number s smaller than 11 mites during seedling

PAGE 47

47 growth. The undamaged transpla nts (42-d old), however, hosted P. latus at densities within a range of 1.2 mites. Pea and Osborne (1996) released densities of N. californicus of 9 and 27 per lime seedling, and 26 and 30 per bean pl ant in plants with initial P. latus densities that were higher (artificial infestations of 133 m ites/leaf in lime and 367 mites/leaf in bean) than those found in the pepper seedlings in this study. In the st udies by Pea and Osborne (1996), suppression of P. latus populations to non damaging levels was achieved with initial prey-predator density ratios smaller than 15:1 which, in a two week pe riod, did not completely eliminate the pest. In this study, curative releases of two predators on P. latus densities from 92 to 271 mites/seedling (e.g. releases delayed 9 to 24 days from the initial P. latus infestation) were too late to be effective for producing undamaged transplants. Even with predators released preventively, higher release densities may be needed to produce P. latus -free pepper transplants in a completely infested plug tray. In this st udy, every seedling in the plug tray was infested artificially with P. latus In most natural infestations this may not be the case and, initially, only a small number of seedlings in the plug tray will most likely host P. latus N. californicus will search to locate the prey (A uger et al., 1999; Collier et al ., 2001), leading to a smaller prey-predator density ratio in the infested seed lings, and thus to a more effective and rapid control of the pest. Conversely, high densities of predators released in every pepper seedling in the nursery may be costly unless they are used as curative releases restricted to highly infested areas (along with lower densit ies used preventively thr oughout the entire greenhouse). N. californicus released at a density of two predator s/seedling costs 1.1 / seedling (costs for shipping and mite release not in cluded) based on a current cost of US$ 5.50 per shaker vial

PAGE 48

48 (1000-mites) when large quantities are purchased The vials contain pr edators of both sexes (female-male ratio is approx. 1:1) wh ich have similar predation rates. The controlled environmental conditions in this study were favorable for mite population increase in N. californicus (Castagnoli and Simoni, 1991; 1994; Gotoh et al., 2004) and P. latus (Jones and Brown, 1983; Li et al., 1985; Li and Li, 1986; Silva et al., 1998). However, further studies are needed to dete rmine the effectiveness of N. californicus to manage P. latus in transplants grown in nursery greenhouses where air temperature can temp orarily reach 35C and relative humidity 100% (Castagnoli and Sim oni, 1991; 1994; 1995) and, in addition, when overhead misting is the irrigation system used (re sults from this study would apply for seedlings irrigated with a subsurface system) (Fitch et al., 2006). In the preventive release scenarios evaluated, N. californicus most likely laid eggs and some survived during the absence of P. latus However, mean predator populations were never higher than two mites per seedling befo re the infestation. The ability of N. californicus to become established in the absence of prey and no n-prey food has been reported in studies where the phytoseiid was evaluated for survival and for control of spider mites (de Courcy Williams et al., 2004; Palevsky et al., 1999; Walzer et al., 2001). In laboratory experiments, N. californicus survived without prey for up to 18 days (with water provided) and 22 days (with water and presence of fungal hyphae) at mean temperatures of 25oC (de Courcy Williams et al., 2004). Others have reported survival without prey as 50 % lethal time values of 10.4 days (Palevsky et al., 1999) and periods of up to 15 days (Walzer et al., 2001). In this study, molds could have been present on pepper seedlings and the surface of the container substrate. Pollen, such as of Quercus species, is an alternative food for N. californicus and will maintain the mite alive but it might not support an increase of the predator population (Castagnoli and Simoni, 1999). If it is

PAGE 49

49 feasible to artificially dust collected pollen (from Quercus or other satisfactory plant species) on seedling trays, this non-prey food source might help the predator survive in the absence of prey. Seedlings may withstand temporary dama ge (Trumble et al., 1993) and, with P. latus suppressed, attain similar growth as in non-infest ed seedlings and exhibit no damage symptoms. Seedlings treated with predators released preventively, or on the same day as P. latus infestations were initiated on two or four unfolded leaves, had seedling growth variables similar to non-infested seedlings. P. latus CMD did not exceed 83 in these transplants, and never had damage indices higher than zero. However, tem porary damage was greater with an infestation and release on unfolded cotyledons. Possibly, the mean of 982 CMD in these early-infested seedlings was near the maximum for transplant growth variables to be smaller (e.g., with P < 0.050) than in non-infested transplants. In treatments that led to undamaged transplants, N. californicus was recovered at densities that were similar to the init ial release density. Establishm ent and increase in number of N. californicus on pepper cotyledons or a few unfolded leaves might be less successful than on leaves in larger plant canopies. Leaf domatia, hair tufts in the axils formed by the central and lateral veins at the basal leaf portion, were not present in leav es of the 42-days old pepper transplants in the present study. Phytoseiids are known to lay eggs and hide in these protective sites (Walter, 1996; Walter and ODowd, 1992). Fara ji et al. (2002) observed that domatia were present on leaves above the eight leaf node in a pepper cultivar, a nd that eggs and larvae of the predatory mite Iphiseius degenerans (Berl.) (Acari: Phytoseiidae) were almost absent on leaves without domatia. In the present study, the absence of these she ltered sites for oviposition could have been a factor that impaired predator col onization when the prey was absent or when the predator was released on cotyledons. It also ma y have reduced its potential to increase in number

PAGE 50

50 when P. latus were present. If there are pepper cultivars which develop domatia on the first leaves, an early establishment of predatory m ites may be possible in transplant nurseries, therefore making low-density releases of N. californicus more effective. In addition, older and larger transplants (with 8 to 10 unfolded leaves) than the one selected in this study are also grown for planting in greenhouses, and th ey might provide a better shelter for N. californicus when released shortly before planting. Transplant growth variables on 42 DAS, which had P. latus infestations within the period 13 DAS, were a function of CMD. In exponentia l decay functions, rapi d reductions occurred for photosynthetic area and dry weight as CM D increased. Stem length and leaf number followed in descending order of level of reducti on. For all the growth variables, it took approximately 100 CMD for an estimated reduction in the order of 5% in relation to measurement in transplants that had not been infested with P. latus When separate regression analyses were calculated for each developmental stage infested by P. latus relative transplant growth reductions were generally reduced to a gr eater extent by a unit of CMD in those seedlings that had been infested with P. latus at younger developmental stages (e.g. infestations initiated in seedlings with cotyledons unfolded compared to seedlings with two unfolded leaves). In pepper and other horti cultural crops, the decrease of gr owth variables have also been described as a function of visual damage i ndices (de Coss-Romero and Pea, 1998; Pea and Bullock, 1994). In pepper seedlings, the use of a damage index scale do es not appear to be useful as an indicator for a curative release of N. californicus at a density of two mites/seedling. With a damage index of one, CMD can be 300 and the number of P. latus can exceed 100 mites, thus requiring a release of more than two predators for an e ffective control. This may not be economically feasible in transp lant production. Moreover, transp lant growth variables can be

PAGE 51

51 reduced even if at transplant age seedlings are ra ted with a visual damage index close to one, an index level which is not easy to disc ern from non-infested seedlings. P. latus can increase in number very rapidly and damage pepper s eedlings. It is easy to fail to notice P. latus on seedlings with in cipient infestations, when dama ge is not evident. Scouting for the presence of this pest will be critical in the greenhouse nursery and before seedlings are transplanted in the pr oduction greenhouse. P. latus densities are generally not used by growers to decide on control actions, si nce counting these mites is neither simple nor rapid to conduct. Without the use of pesticides, management of P. latus in transplant nu rseries could use preventive releases of N. californicus on plug trays to avoid pest outbreaks and to lead to undamaged transplants. Some N. californicus will survive on seedlings even if broad mites are absent during transplant production. The predator will add to seedling pe st protection by also preying on other phytophagous mites th at might be present, such as T. urticae If P. latus outbreaks occur in small areas of the nursery th ey can be treated with localized sprays of micronized-sulfur or abamectin (Ramakers 2005; Weintraub, 2004). The fact that the preventive strategies evaluated in th is study did not result in P. latus -free transplants raises the possibility of pest outbreaks occurring later when these seedlin gs are planted in a fruit production greenhouse. More than two N. californicus per seedling might be needed in preventive releases to obtain P. latus -free transplants with scenarios similar to t hose in this study. In addition, one or more supplementary releases of the predator may be required to avoid possi ble later yield loss. Outcomes of these scenarios are investig ated in subsequent greenhouse studies. 2.5 Summary In warm-climate regions, broad mites [ Polyphagotarsonemus latus (Banks)] infest bell pepper ( Capsicum annuum L.) seedlings grown in greenhouses and repeated pesticide sprays are the sole means to ensure production of undama ged transplants. Biological control of P. latus

PAGE 52

52 was evaluated releasing two predatory mites Neoseiulus californicus (McGregor) on bell pepper seedlings grown under semi-controlled environmental conditions (L: 16 h, 24oC, 70% RH and D: 8 h, 19oC, 80% RH). Prey-predator scenarios had ar tificial infestations initiated with two P. latus on seedlings at growth developmental stag es a) cotyledons unfolded (13 days after seeding, DAS), with N. californicus released 0, +15, or +24 days from initial infestation (DFI), b) two leaves unfolded (28 DAS), with N. californicus released 0, or +9 DFI, or c) four leaves unfolded (37 DAS), with N. californicus released , or 0 DFI. Seedlings were also grown mite-free, or had only P. latus or N. californicus introduced at one of the three growth developmental stages. Prey-predator density ratios seedling mites : seedling mites on the day both mite species were first present on seedlings were in the range s of 0.7:1 to 3.3:1 with predators released at pre-infestation days, 0.6:1 to 1.5:1 with predators released on the same days of infestation, and 54:1 to 387:1 with predators released at post-infe station days. Transplant s (42-days old) grown from infested seedlings had seven leaves unfol ded and neither damage symptoms nor growth variables differed from non-infest ed seedlings in the following scenarios (which still hosted P. latus ): a) predators released at pre-infe station days (hosted 3 to 5 P. latus /seedling) or on the same days of infestation (hosted 4 to 13 P. latus /seedling), and b) infestat ion initiated at four leaves with no predators released (hosted 16 P. latus /seedling). At the time predators were released at post-infestation days (e.g. from +9 to +24 DFI), P. latus was already at densities from 92 to 271 mites/seedling and severe damage on seed lings made these releases too late to be effective for producing usable 42-days old transp lants. Combining data from all infestation scenarios, transplant growth variables were reduced with exponen tial decay functions by increased abundance of P. latus over time (estimated as cumulative mite-days/seedling); reductions were 5% with 100 cumulative mite-days/ seedling. Overall, when a single

PAGE 53

53 preventative release of two N. californicus per seedling was used on pl ug trays, transplants had accumulated less than 83 P. latus mite-days and were considered undamaged. Because P. latus infestation time in the nursery is unpredictable, management with N. californicus should be initiated with a release as early as cotyledons unf old. However, in scenarios similar to those in this study, release densities greater than two N. californicus per seedling might be needed in preventive releases to obtain broad-mite-free tran splants, and to avoid fu rther plant damage after transplanting in fru it production greenhouses.

PAGE 54

54 Table 2-1. Pest-predator scenar ios evaluated during pepper seed ling growth. Seedlings were infested with two P. latus at three growth developmental stages and two N. californicus were released before, at, or afte r the day of infestation. Additional seedling groups had only one mite species introduced at one of each developmental growth stage, or no mites introduced. Seedling developmental stage (days after seeding) P. latus infestation N. californicus release Release delay from infestation (days) Treatment codea Cotyledons (13) none -PC Cotyledons (13) Cotyledons (13) 0 PC-NC Cotyledons (13) 2-leaves (28) +15 PC-N2 Cotyledons (13) 4-leaves (37) +24 PC-N4 2-leaves (28) none -P2 2-leaves (28) Cotyledons (13) P2-NC 2-leaves (28) 2-leaves (28) 0 P2-N2 2-leaves (28) 4-leaves (37) +9 P2-N4 4-leaves (37) none -P4 4-leaves (37) Cotyledons (13) P4-NC 4-leaves (37) 2-leaves (28) P4-N2 4-leaves (37) 4-leaves (37) 0 P4-N4 none Cotyledons (13) -NC none 2-leaves (28) -N2 none 4-leaves (37) -N4 none none -Non-infestedb aP : P. latus infestation, N : N. californicus release, C : unfolded cotyledons, 2 : unfolded two leaves, and 4 : unfolded four leaves in Non-infested seedlings. bControl, mite-free seedlings.

PAGE 55

55Table 2-2. Number of P. latus and N. californicus recovered from pepper seedlings with m ites introduced at three seedling growth developmental stages. Mite number and pr ey:predator density ratios are indicated fo r the first day both mite species were present and, densities, density ratios and cumulative broad mite-days are in dicated for 42-days old transplants. 1st day for both P. latus and N. californicus 42 days after seeding Mite number (no./seedling) Mite number (no./seedling) Treatment codea Release delay from infestation (days) P. latus N. californicus Ratiob P : N (no. : no.) P. latus N. californicus Ratio P : N (no. : no.) Cumulative P. latus days (mite-days/seedling) PC -1.0 0.3b 0.0 0.0c --242.4 109.5 a 0.0 0.0c --5336 1071 a PC-NC 0 0.7 0.2b 1.2 0.4bc 0.6 : 112.8 3.3 c 1.8 1.4bc 7.1 : 1982 741 b PC-N2 +15 197.5 37.0a 1.7 0.2abc 116.2 : 116.0 3.8 bc 3.2 1.9abc 5.0 : 12077 393 a PC-N4 +24 271.2 73.3a 0.7 0.4bc 387.4 : 191.3 11.9 ab 6.8 4.4ab 13.4 : 15621 1097 a P2 -1.5 0.3b 0.0 0.0c --357.5 50.5 a 0.0 0.0c --1551 149 ab P2-NC 1.7 0.4b 0.5 0.3bc 3.3 : 15.2 0.9 cd 1.0 0.5bc 5.2 : 183 7 c P2-N2 0 2.0 1.0b 2.0 0.3a 1.0 : 14.0 1.3 cd 1.0 0.3bc 4.0 : 143 7 cd P2-N4 +9 91.7 15.9ab 1.7 0.4abc 53.9 : 1226.3 23.4 a 9.2 3.2a 24.6 : 11359 214 ab P4 -0.8 0.4b 0.0 0.0c --15.7 2.2 c 0.0 0.0c --34 3.8cd P4-NC 1.2 0.2b 0.7 0.2bc 1.8 : 13.8 0.9 cd 1.5 0.5abc 2.5 : 125 6.9cd P4-N2 0.7 0.4b 1.0 0.5bc 0.7 : 13.3 0.6 cd 2.3 0.9abc 1.4 : 111 0.5d P4-N4 0 1.2 0.4b 1.2 0.2bc 1.0 : 13.7 0.3 cd 3.0 1.6abc 1.2 : 113 2.9d NC -0.0 0.0b 1.2 0.2b --0.0 0.0 e 0.3 0.2bc --0 0.0e N2 -0.0 0.0b 1.7 0.4b --0.0 0.0 e 0.7 0.2bc --0 0.0e N4 -0.0 0.0b 1.5 0.0b --0.0 0.0 e 1.0 0.5bc --0 0.0e Non-infested -0.0 0.0b 0.0 0.0c --0.0 0.0 e 0.0 0.0c --0 0.0e aP : P. latus infestation, N : N. californicus release, C : unfolded cotyledons, 2 : unfolded two leaves, and 4 : unfolded four leaves in Non-infested seedlings (13, 28, and 37 days after seeding, respectively). bMean prey:predator ( P. latus : N. californicus ) density ratio calculated for treat ments with both mite species. Selected means from the treatment time significant interactions ( P. latus number: F = 14.63; df = 101, 66; P = 0.0001; N. californicus number: F = 3.52; df = 101, 149; P = 0.0001; P. latus CMD: F = 36.74; df = 101, 86; P < 0.001) are presented here for days 13, 28 and 37, and for 42 DAS. Letter groupings were assigned to the means in each column using the PDMIX800 SAS macro (Saxton, 1998) after post-hoc Tukey-Kra mer ( P < 0.050) comparisons were made on LSMeans across days with the PR OC MIXED procedure in SAS. Within a column, means followed by different letters are significantly different at P < 0.050.

PAGE 56

56 Table 2-3. Bell pepper growth variables and s eedling visual damage index in 42-days old transplants. Seedlings had two P. latus and/or two N. californicus introduced at three seedling growth developmental stages. Treatment codea Release delay from infestation (days) Dry weight (mg/ seedling) Stem length (mm) Leaf number (no./ seedling) Leaf area (cm2/ seedling) Damage index (0 to 4) PC -101 d 58 d 3.5 cd 18.0 c 4.0 a PC-NC 0 222 abcd 114 abc 6.2 abc 57.8 ab 0.3 cd PC-N2 +15 122 bcd 86 cd 6.2 abc 29.6 bc 1.8 b PC-N4 +24 81 d 53 d 2.8 d 17.6 c 4.0 a P2 -161 bcd 109 b 5.8 abc 55.6 ab 1.7 bc P2-NC 297 ab 139 ab 7.0 ab 76.1 a 0 d P2-N2 0 248 ab 111 abc 6.3 ab 59.7 ab 0 d P2-N4 +9 154 cd 89 cd 5.0 bcd 44.7 abc 2.5 b P4 -274 a 135 ab 6.5 ab 71.6 a 0.3 cd P4-NC 283 abc 140 ab 6.8 ab 77.7 a 0 d P4-N2 281 a 132 ab 7.2 ab 76.1 a 0 d P4-N4 0 287 a 145 ab 7.0 a 75.1 a 0 d NC -279 abc 143 ab 6.7 ab 72.1 a 0 d N2 -301 ab 145 ab 6.8 ab 78.7 a 0 d N4 -305 a 147 a 7.0 a 75.1 a 0 d Non-infested -290 abc 144 ab 6.7 ab 75.1 a 0 d aP : P. latus infestation, and N : N. californicus release. C : unfolded cotyledons, 2 : unfolded two leaves, and 4 : unfolded four leaves in Non-infested seedlings (13, 28, and 37 days after seeding, respectively). bNo injury (0), very sev ere injury (4). See Figure 2-1 and de tailed description in the Materials and Methods section of this manuscript. Means from the treatment time significant ( P < 0.001) interactions are presented for 42 DAS. Letter groupings were assigned to the means of 42 DAS usin g the PDMIX800 SAS macro (Saxton, 1998) after post-hoc Tukey-Kramer ( P < 0.050) comparisons were made on LSMeans across all sampling days with the PROC MIXED procedure in SAS. Within a column, means followed by different letters are significantly different at P < 0.050.

PAGE 57

57 Figure 2-1. Visual damage ratings on pe pper seedling used to assess damage from P. latus feeding. Scale indices ranged from zero (no injury) to four (very seve re injury). A detailed description is incl uded in the Materials and Methods section.

PAGE 58

58 Figure 2-2. Number of P. latus and N. californicus recovered after mites were introduced to pepper seedlings at three grow th developmental stages. Two P. latus ( P ) infested seedlings at unfolded cotyledons (13 DAS), two unfolded leaves (28 DAS) or four unfolded leaves (37 DAS). Two N. californicus ( N ) were released at a seedling developmental stage that preceded (9, 15, or 24 days before), was at the same (same day), or followed (9, 15, or 24 days after) the developmental stage of P. latus infestation. P. latus infestations with no N. californicus (A), N. californicus released and absence of P. latus (B), and releases of N. californicus with infestations at unfolded cotyledons (CE), two leaves (FH), and four leaves (IK) Plotted for each sampling date: mean SEM, n = 3. Treatment time interactions for P. latus number: F = 16.51; df = 101, 66; P < 0.001; and for N. californicus number: F = 3.52; df = 101, 149; P < 0.001.

PAGE 59

59 Figure 2-3. Polyphagotarsonemus latus cumulative mite-days in pepper seedlings after P. latus and N. californicus were introduced at three seedli ng growth developmental stages. Seedlings were infested with two P. latus ( P ) at A) unfolded cotyledons ( C ) without ( PC ) or with N. californicus ( N ) released 0, +15, or +24 days from initial infestation (DIF) ( PC NC PC N2 PC N4 respectively), B) two leaves ( 2 ) without ( P2 ) or with N. californicus released 0, or +9 DFI ( P2 NC P2 N2 P2 N4 respectively), or C) four leaves ( 4 ), without ( P4 ) or with N. californicus released , or 0 DFI ( P4 NC P4 N2 P4 N4 respectively). Cumulative mite-days were zero in seedlings grown mite-free or, that had only N. californicus introduced at each seedling growth developmental stage (not plotted in this fi gure). Plotted for each sampling date: mean SEM, n = 3. Treatment time: F = 36.74; df = 101, 86; P < 0.001.

PAGE 60

60 Figure 2-4. Changes in visual damage indices in terminal sh oots of pepper seedlings after P. latus and N. californicus were introduced at three se edling growth developmental stages. A) seedlings were grown mite-free or, had only N. californicus ( N ) introduced at unfolded cotyledons ( C ), two unfolded leaves ( 2 ), or four unfolded leaves ( 4 ) ( NC N2 and N4 ); B) seedlings were infested with two P. latus ( P ) at unfolded cotyledons without ( PC ) or with N. californicus released 0, +15, or +24 days from initial infestation (DIF) ( PC NC PC N2 PC N4 respectively); C) seedlings were infested with two P. latus at unfolded two leaves without ( P2 ) or with N. californicus released 0, or +9 DFI ( P2 NC P2 N2 P2 N4 respectively); D) seedlings were infested with two P. latus at unfolded four leaves, without ( P4 ) or with N. californicus released , or 0 DFI ( P4 NC P4 N2 P4 N4 respectively). Visual damage was rated with a 0 scale (see Figure 2-1). Pl otted for each sampling date: mean SEM, n = 3. Treatment time: F = 6.62; df = 150, 245; P < 0.001.

PAGE 61

61 Figure 2-5. Dry weight increas e in pepper seedlings after P. latus and N. californicus were introduced at three seedling growth developm ental stages. A) seedlings were grown mite-free or, had only N. californicus ( N ) introduced at unfolded cotyledons ( C ), two unfolded leaves ( 2 ), or four unfolded leaves ( 4 ) ( NC N2 and N4 ); B) seedlings were infested with two P. latus ( P ) at unfolded cotyledons without ( PC ) or with N. californicus released 0, +15, or +24 days from initial infestation (DIF) ( PC NC PC N2 PC N4 respectively); C) seedlings were infested with two P. latus at unfolded two leaves without ( P2 ) or with N. californicus released 0, or +9 DFI ( P2 NC P2 N2 P2 N4 respectively); D) seedlings were infested with two P. latus at unfolded four leaves, without ( P4 ) or with N. californicus released , or 0 DFI ( P4 NC P4 N2 P4 N4 respectively). Plotted for each sampling date: mean SEM, n = 3. Treatment time: F = 2.95; df = 101, 71; P < 0.001.

PAGE 62

62 Figure 2-6. Stem length increase in pepper seedlings after P. latus and N. californicus were introduced at three seedling growth developm ental stages. A) seedlings were grown mite-free or, had only N. californicus ( N ) introduced at unfolded cotyledons ( C ), two unfolded leaves ( 2 ), or four unfolded leaves ( 4 ) ( NC N2 and N4 ); B) seedlings were infested with two P. latus ( P ) at unfolded cotyledons without ( PC ) or with N. californicus released 0, +15, or +24 days from initial infestation (DIF) ( PC NC PC N2 PC N4 respectively); C) seedlings were infested with two P. latus at unfolded two leaves without ( P2 ) or with N. californicus released 0, or +9 DFI ( P2 NC P2 N2 P2 N4 respectively); D) seedlings were infested with two P. latus at unfolded four leaves, without ( P4 ) or with N. californicus released , or 0 DFI ( P4 NC P4 N2 P4 N4 respectively). Plotted for each sampling date: mean SEM, n = 3. Treatment time: F = 3.39; df = 101, 62; P < 0.001.

PAGE 63

63 Figure 2-7. Number of leaves in crease in pepper seedlings after P. latus and N. californicus were introduced at three seedling growth de velopmental stages. A) seedlings were grown mite-free or, had only N. californicus ( N ) introduced at unfolded cotyledons ( C ), two unfolded leaves ( 2 ), or four unfolded leaves ( 4 ) ( NC N2 and N4 ); B) seedlings were infested with two P. latus ( P ) at unfolded cotyledons without ( PC ) or with N. californicus released 0, +15, or +24 days from initial infestation (DIF) ( PC NC PC N2 PC N4 respectively); C) seedlings were infested with two P. latus at unfolded two leaves without ( P2 ) or with N. californicus released 0, or +9 DFI ( P2 NC P2 N2 P2 N4 respectively); D) seedlings were infested with two P. latus at unfolded four leaves, without ( P4 ) or with N. californicus released , or 0 DFI ( P4 NC P4 N2 P4 N4 respectively). Counted leaves were longer than 1.5 cm. Plotted for each sampling date: mean SEM, n = 3. Treatment time: F = 3.43; df = 101, 54; P < 0.001.

PAGE 64

64 Figure 2-8. Leaf area increase in pepper seedlings after P. latus and N. californicus were introduced at three seedling growth developm ental stages. A) seedlings were grown mite-free or, had only N. californicus ( N ) introduced at unfolded cotyledons ( C ), two unfolded leaves ( 2 ), or four unfolded leaves ( 4 ) ( NC N2 and N4 ); B) seedlings were infested with two P. latus ( P ) at unfolded cotyledons without ( PC ) or with N. californicus released 0, +15, or +24 days from initial infestation (DIF) ( PC NC PC N2 PC N4 respectively); C) seedlings were infested with two P. latus at unfolded two leaves without ( P2 ) or with N. californicus released 0, or +9 DFI ( P2 NC P2 N2 P2 N4 respectively); D) seedlings were infested with two P. latus at unfolded four leaves, without ( P4 ) or with N. californicus released , or 0 DFI ( P4 NC P4 N2 P4 N4 respectively). Plotted for each sampling date: mean SEM, n = 3. Treatment time: F =3.65; df = 101, 83; P < 0.001.

PAGE 65

65 Figure 2-9. Reduction of relative transplant growth variables and visual damage index as affected by P. latus cumulative mite-days in pepper seedlings that had pest infestations initiated with two P. latus at three seedling growth developmental stages. Only regression lines with P < 0.05 are plotted. Equations and values for R2, F and P-values are presented in Table 2-4. Visual damage was rated with a 0 scale (Figure 2-1).

PAGE 66

66 Table 2-4. Regression equations for the reduction of transplant growth variables relative to non-infested transplants, a nd for visual damage index as affected by increased cumulative mite-days. P. latus infestations initiated in seedlings with unfolded cotyledons, two unfolded leaves, a nd four unfolded leaves. Growth reduction (%) and visual damagea Equation R2 F P-value Range for cumulative mite-days (x) P. latus on cotyledonsb Dry weight = 68.91 (1 e0.0008x) 0.75 39.20.0001 001 Leaf area = 80.07 (1 e0.0008x) 0.83 63.90.0001 001 Leaf number = 95.02 (1 e0.0001x) 0.68 27.00.0002 001 Stem length = 63.44 (1 e0.0005x) 0.82 59.40.0001 001 Damage index = 4 (1 e.0004x) 0.86 78.10.0001 001 P. latus on 2 unfolded leaves Dry weight = 0.028x 0.67 26.60.0002 028 Leaf area = 0.017x 0.19 3.10.1045 028 Leaf number = 0.013x 0.33 6.40.0251 028 Stem length = 0.015x 0.11 1.70.2170 028 Damage index = 4 (1 e.0005x) 0.68 28.00.0001 028 P. latus on 4 unfolded leaves Dry weight = 0.062x 0.03 0.50.4905 0 Leaf area = 0.104x 0.00 0.11.0 0 Leaf number = 0.134x 0.00 2.01.0 0 Stem length = 0.002x 0.00 2.01.0 0 Damage index = 0.005x 0.14 2.10.1657 0 aGrowth reduction relative to 42-days old transplants that had no P. latus introduced. The mean values in transplants were: dry weight = 294 mg, leaf area = 75.3 cm2, leaf number = 6.8, stem length = 115 mm, and damage index = 0. bCombined data from treatments PC PC-NC PC-N2 PC-N4 for infestations on unfolded cotyledons; P2 P2-NC P2-N2 P2-N4 for infestations on unfolded 2 leaves and P4 P4-NC P4-N2 P4-N4 for infestations at unfolded 4 leaves. Thr ee replications per treatment. See Table 2-1 for explanation of treatment codes.

PAGE 67

67 Figure 2-10. Transplant growth variables a nd damage index relationships as affected by P. latus cumulative mite-days in 42-days old pepper seedlings. Analyses included data from all treatments and three replications (n = 48). Growth variab les were fit with non linear regressions of the exponential type y = a ebx: leaf area (; a = 73.58; b = 0.0003; R2 = 0.74; df = 1, 46; F = 129.9; P < 0.001), stem length (; a = 136.46; b = 0.0002; R2 = 0.74; df = 1, 46; F = 130.2; P < 0.001), dry weight (; a = 281.18; b = 0.0003; R2 = 0.75; df = 1, 46; F = 136.8; P < 0.001), and leaf number (; a = 6.80; b = 0.0001; R2 = 0.68; df = 1, 46; F = 97.6; P < 0.001). Damage index ( ) was fit to a exponential functi on that rises to a maximum, y = a (1 ebx), with a = 4; b = 0.0004; R2 = 0.86; df = 1, 46; F = 284.0; P < 0.001). For a description of treatments see Table 2-1. Visual damage wa s rated with a 0 scale (see Figure 2-1).

PAGE 68

68 CHAPTER 3 EFFECTIVE MANAGEMENT OF Polyphagotarsonemus latus (Banks) (ACARI: TARSONEMIDAE) WITH Neoseiulus californicus McGregor (ACARI: PHYTOSEIIDAE) RELEASED IN GREENHOUSE-GROWN PEPPER ( Capsicum annuum L.) CROPS INITIATED WITH INFE STED TRANSPLANTS 3.1 Introduction Broad mite, Polyphagotarsonemus latus (Banks), is a recurrent pest of bell peppers ( Capsicum annuum L.) grown in greenhouses in regions with tropical, dry, or with mild winter climate (Castagnoli and Falchini 1993; Cho et al., 1996; Dik et al., 1999; Gerson, 1992; Ibrahim and Low, 1998; Laffi, 1982; Mizobe and Tamura 2004; Nuez et al., 1996; Weintraub et al., 2003). In these regions, the probability of greenhouse crops becoming infested with P. latus is high because outdoors the polyphagous mite inha bits numerous cultivated and non-cultivated plant species (plants in more than 60 fam ilies) many of which can grow year-round (Gerson, 1992; Ibrahim and Low, 1998; Jeppson et al., 1975; Silva et al., 1998). P. latus populations thrive in young pepper plants growing unde r warm and humid environments (e.g. 25oC and 80% RH) which are common inside greenhouses (C ho et al., 1996; Coss-Ro mero and Pea, 1998; Ibrahim and Low, 1998; Jeppson et al., 1975; Jones and Brown, 1983; Li et al., 1985). Therefore, even in regions with cold winters, pest outbreaks may occur in greenhouses (Basset, 1985; Cross, 1979; Gerson, 1992; Hussey, 1985; Jeppson et al., 1975; Ramakers, 2005). How P. latus finds its way into fruit production gr eenhouses is directly related to its dispersal mechanisms. Phoresy appears to be an important means of dispersion for this minuscule mite (with a body length circa 150 m) and it has been observed to occur on whiteflies ( Bemisia spp.; Insecta: Homoptera: Aleyrodida e) and aphids (Inse cta: Hemiptera: Aphididae) (Fan and Petitt, 1998; Flechtmann et al., 1990; Palevsky et al ., 2001; Soroker et al., 2003). However, the pest can also be carried to crops by air currents, workers that have touched infested plants, or mites that have crawled onto cr ops from infested plants present either in the

PAGE 69

69 vicinity of or inside the greenhouse (Fan and Petitt, 1994, 1998; Flechtmann et al., 1990; Gerson, 1992). In addition, using transplants that host P. latus is a common means of introducing the pest to greenhouses (Cho et al., 1996; Cross, 1979; Dik et al., 1999; La ffi, 1982; Weintraub et al., 2003). P. latus infestations that occur late during th e transplant production in the nursery greenhouse can result in transplants which appear undamaged while hosting a few mites in the apical shoot (Chapter 2). Undetected P. latus will then rapidly devel op into large populations on the transplanted crop (Cho et al., 1996; Ho, 1991; Ibrahim and Low, 1998; Jeppson et al., 1975; Laffi, 1982; Silva et al., 1998; Weintr aub et al., 2003). Only a few P. latus -infested plants may be needed to infest an entire greenhouse, as crop activities which invo lve touching plants (e.g. placing irrigation emitters, extending strings fo r trellising canopies, de-flowering, and shoot pruning) rapidly disperse mites. P. latus infestations on pepper are almost exclusively managed with miticides, and control is recommended to be started either as s oon as mites are detected on the underside of young leaves, or preventively (Cho et al., 1996; Di k et al., 1999; Gerson, 1992; Hussey, 1985; Nuez et al., 1996; Ramakers, 2005). When plants are not regul arly and carefully inspected for pests, it is not until plant injury symptoms are noticed (narrowing and curling downwards of youngest leaves) that an intervention with a miticide spra y (most commonly, sulfur, abamectin or dicofol) is carried out (Cho et al., 1996; Dik et al ., 1999; Gerson, 1992; Hussey, 1985; Jeppson et al., 1975; Nuez et al., 1996; Olson et al., 2005; Polack and Mitidieri, 2005; Weintraub et al., 2003). Since the latter miticides do not kill eggs and P. latus are hidden in the plant shoots, multiple sprays are repeated at periods that are close to the developmental time of P. latus (4 to 7 days from egg to adult) (Gerson, 1992; Weintraub et al., 2003). If the pepper production system relies mostly on the use of natural en emies for control of pests, continuous spraying of the current

PAGE 70

70 miticides labeled for use on P. latus will disrupt biological control programs and the activity of pollinators, particularly spraying with dicofol (Basset, 1985; Brown and Jones, 1983; Dik et al., 1999; Fan and Petitt, 1994; Griffiths, 1999; Koppert Biol. Syst., 2006; Ramakers, 2005). However, for P. latus management, there are few pest management methods that are an alternative to miticides (Dik et al., 1999; Gers on, 1992; Fan and Petitt, 1994; Pea et al., 1996; Ramakers, 2005; Weintraub et al., 2003). Several phytoseiids have been reported to feed on P. latus (Badii and McMurtry, 1984; McMurtry et al., 1984; Pea, 1992; Steiner et al., 2003a; 2003b) and a few species have been evaluated for P. latus control in greenhouse-grown peppers. Neoseiulus cucumeris (Oudemans) (Acari: Phytoseiidae) [reared commercially and us ed for control of nymphs of western flower thrips, Frankliniella occidentalis (Pergande) (Thripidae: Thysa noptera)] (Mizobe and Tamura, 2004; Weintraub et al., 2003) and Neoseiulus barkeri Hughes (Acari: P hytoseiidae) (not currently available in th e U.S.) (Fan and Petitt, 1994; Pea and Osborne, 1996) were studied for management of P. latus in greenhouse-grown peppers using multiple releases or slow-release systems (plant-hanging sachets th at contain predator, prey, and food for the prey). Another predacious mite with potential use is Neoseiulus (= Amblyseius ) californicus McGregor (Acari: Phytoseiidae) (Castagnoli and Falchini, 1993). This mite is commercially reared in the U.S. and E.U. for management of two-spotted spider mite [ Tetranychus urticae Koch (Acari: Tetranychidae)] in vegetable a nd ornamental crops in greenhous es (Gillespie and Raworth, 2005; Griffiths, 1999; Osborne et al., 1998; 2005). N. californicus can feed on different prey and non-prey food sources (a predator with a life-style Type II, as categorized by McMurtry and Croft, 1997). This predatory mite has been ev aluated as a biological control candidate for P. latus in experiments under laborato ry conditions (Castagnoli a nd Falchini, 1993); in pepper

PAGE 71

71 seedlings under controlled environment (Chapter 2); and, in short-term greenhouse experiments, in bean ( Phaseolus vulgaris L.) plants, and in lime [ Citrus aurantifolia Christm. (Swingle)] and pepper ( Capsicum annuum L.) seedlings (Pea and Osborne, 1996; L.S. Osborne, personal communication). Results from these previous studies indicated that N. californicus could survive and increase in number while feeding on P. latus and that pest populatio ns could be rapidly reduced with post-infestation releases of N. californicus at prey-predator re lease density ratios on leaves less than 15:1 plant mites : leaf mites in lime seedlings and bean plants (Pea and Osborne, 1996), or 3:1 in pepper seedlings seedling mites : seedling mites (Chapter 2). N. californicus has been reported to tole rate temperatures up to 33oC and air relative humidity as low as 60% (Castagnoli and Sim oni, 1994; Castagnoli and Simoni, 1995, cited in Griffiths, 1999; Gotoh et al., 2004), and to be able to survive for long peri ods in the absence of prey or when non-prey food is present (Castagno li and Simoni, 1999; de Courcy Williams et al., 2004; Griffiths, 1999; Palevsky, 1999; Wa lzer, 2001). The release of two N. californicus on pepper seedlings either 0, 9 or 24 days before an initial infestation with two P. latus /seedling, resulted in 42-d old seedlings that appeared visually undamaged and had growth variables that were not different from non-infested seedli ngs (Chapter 2). These undamaged seedlings, however, hosted as many as five P. latus Therefore, it is important to determine if N. californicus could be effective for managing P. latus when greenhouse bell pepper crops are started with transplants that appear undamaged, but host P. latus and to examine the effect of the pest on plant growth and fruit yi eld as pest population increases. While feeding, P. latus extracts cell contents from the upper cell layers of young plant tissues and, it is thought, secret s substances that are toxic for the plant (Basset, 1985;

PAGE 72

72 Coss-Romero and Pea, 1998; Gerson, 1992; Gri nberg et al., 2005; Gui et al., 2001). As a consequence of feeding, plant injury during ve getative and flowering stages can severely decrease fruit yield due to the reduction of leaf area, diminuti on in terminal shoot growth, and abscission and injury of flowers and young fru its (Cho et al., 1996; Co ss-Romero and Pea, 1998). In pepper plants where P. latus infestations were left unt reated, Coss-Romero and Pea (1998) estimated that pepper fruit yield decrea sed linearly with increa sed levels of damage indices, visually assessed on plan ts. Similar relationships have been calculated in beans and citrus plants damaged by P. latus (de Holanda et al., 1992; Pea 1990; Pea and Bullock 1994). Information on P. latus density on young pepper plants and its effect on fruit production are needed to develop effective pest manageme nt strategies at cr op initiation. If P. latus populations can be determined on leaves, the abundance of P. latus over time (a variable known as cumulative mite-days; see Chapter 2) could be used to estimate reductions on early fruit yield and plant growth (Pea, 1990). Know ledge of the impact of early-crop P. latus infestations on production of greenhouse-grown peppers is needed for decision-making in pest management using N. californicus. The objective of this study was to evaluate the effectiveness of N. californicus released for management of P. latus in scenarios where bell pepper seed lings became infested just prior to transplanting in a greenhouse. Various pest-predator scenarios we re tested and compared with sulfur spray programs, non-managed infestations and no infestations. The effectiveness of P. latus management by N. californicus was evaluated using estimates of pest population density, cumulative presence of the pest on plant, plan t visual damage, plant growth variables and marketable fruit yield. Additional objectives of the study were to examine the economic

PAGE 73

73 viability of the pest management scenarios a nd to identify relations hips among plant growth variables and yield as affected by cumulative presence of the P. latus on plants over time. 3.2 Materials and Methods 3.2.1 Experiments and Treatments Two experiments (Fall-2004 and Spring-2005) we re conducted in a passively ventilated greenhouse (Top Greenhouses Ltd., Rosh Haayin, Isra el) located at the Protected Agriculture Center, Plant Sciences and Research Unit of the Univ. of Florida, in Citra, Fla. Treatments in each experiment are listed in Table 3-1. In both e xperiments there were cohor ts of seedlings that were either kept pest-free or that were artificially infested with two P. latus three days before transplanting. In seedlings that were infeste d, one group was transplanted with the pest left untreated and other groups were tr eated to suppress the increase of P. latus populations. In the Fall-2004 experiment, plan ts were treated with a) two N. californicus released on each P. latus -infested seedling, either on ce (4 days after transplanting, DAT), twice (4 and 21 DAT when flowers on second-level nodes were at anthesis), or three times (4, 21, and 30 DAT when fruit set on second-level nodes were approxim ately 2 cm in diameter), or b) four sulfur sprays applied weekly, beginning when first plant injury symptoms were evident [16 DAT; plants exhibiting a visual damage index of 2 (Figure 3-1)]. In the Spring-2005 experiment, preventative ma nagement strategies and increased predator release densities were added for evaluation against P. latus Infested seedlings were treated with a) two and four N. californicus released once, either before seedlings were infested with P. latus ( DAT), or after the initial infestation (0 DAT or 4 DAT) or, b) five sulfur sprays applied weekly, with the first spray either before ( DAT ) or after (0 DAT) seedlings were infested with P. latus

PAGE 74

74 3.2.2 P. latus Infestation, N. californicus Release, and Sulfur Sprays A P. latus colony, initiated with mites from a citr us nursery at the Univ. of Florida in Gainesville, Fla., was maintain ed on bell pepper cv. Legionn aire (Rogers-Syngenta, Boise, Idaho) seedlings in a plant growth chamber [L:16 h, irradiance: 220 20 mols-1m-2 (PAR), air temperature: 25 1oC, and D:8 h, 21 1oC]. N. californicus in vials containing 1000 mites were shipped overnight from a commercial rearing facility (B iotactics, Romoland, Calif.) and arrived in Gainesville, Fla. the day before they were released. In both experiments, two gravid female P. latus were transferred onto the top leaves of pepper (cv. Legionnaire) seedlings with seven to eight leaves unfolded. Seedlings were transplanted in the greenhouse three days later (Figure B-1). On selected dates (Table 3-1), two and four gravid female N. californicus were released onto the top leaves of each seedli ng or plant. Mites were chosen using a stereomicroscope and handled with a single-hair brush. For the purpose of this experiment, N. californicus were temporarily stored in gelatin cap sules (size 0; Capsuline, Pompano Beach, Fla.) for 2 h before capsules were uncoupled ove r each seedling (in the evening) in the growth chamber or plant in the greenhouse. Micronized-sulfur (Microthiol Disperss 80WS, Cerexagri, King of Prussia, Penn.) was sprayed in the evenings at a rate of 3.4 kgha-1 (a low rate for avoiding plant spray injury in greenhouses where temperatures can exceed 32oC) with an 8-L hand-held pump sprayer equipped with a single hollo w-cone nozzle (flow 400 mLmin-1 at 172 kPa). Plants were sprayed thoroughly to cover the whole canop y and underside of top leaves w ith volumes that ranged from 0.01.20 L/plant depending on plant deve lopmental growth stage. 3.2.3 Transplant Production Pepper seedlings were grown on a 70% peatmo ss : 30% vermiculite (v/v) substrate mix (Pro-mix 0463, Premier, Qubec, Canada) in 128-cell (top width top length height: 3.5 3.5

PAGE 75

75 6.5 cm; 35-mL/cell) polystyrene transplant flats (Speedling Inc., Sun City, Fla.) until they had four unfolded leaves. Seedlings were then replan ted into larger polyethyl ene container cells (top width top length height: 6 6 9 cm; 200-mL/cell) and pl aced inside trays with width length height: 23 50 5 cm (16 containers/tray). Seedlings were grown in an insulated room (width length height: 3 3 3 m; Aluma-Shield, Miami, Fla.) retrofitted to a plant growth chamber with controlled environment [L:16 h, irradiance 220 20 mols-1m-2 (PAR), air temperature 24 1oC; air relative humidity 70 10%, and D:8 h, 19 1oC; 80 10%]. In the Spring-2005 experiment, cohorts of seedlings were placed in four separate growth chambers (maintained at the same prev ious environmental conditions) one week before transplanting in the greenhouse. Th e chambers had seedlings with 1) P. latus only, 2) P. latus and N. californicus (the two predator density treatments separated by wa ter and a vertical plank as barriers), 3) a preven tive spray of sulfur and P. latus and 4) neither P. latus nor N. californicus. After cotyledons unfolded, seedlings were irrigated using a solution with nutrient concentrations (mgL-1) of NO3 --N: 70, P: 50, K: 100, Ca: 90, Mg: 40, S: 56, Fe: 2.8, Cu: 0.2, Mn: 0.8, Zn: 0.3, B: 0.7, and Mo: 0.06. The nutrient solution was delivered (30 to 40 mL/seedling, every one to two days) first on the top of the media and later, once mites were introduced, into the trays containi ng the plug cells (subirrigation). 3.2.4 Greenhouse Plant Arrangement and Crop Management Seedlings were transplanted on 17 Sept. (F all-2004) and 9 Mar. (Spring-2005) into 11.3-dm3 black polyethylene nursery pots (diamete r: 25 cm, height: 30 cm, model A-30, Lerio Corp., Kissimmee, Fla.) containing composted pine bark as soilless media. Treatment plots consisted of plants aligned in 2.5-m long rows (20 plants in double ro ws for a density of 4

PAGE 76

76 plants/m2 in Fall-2004 and 10 plants in single rows, for a density of 3 plants/m2 in Spring-2005) (Figure B-2). Each plant row was in the center of a 3.2 m 5 m (width length) area which was surrounded by a vertical white polyethylene curt ain 1.50-m tall. This wind barrier around the plot was an effective method for minimizing mite contamination during th e period that lasted from transplanting until first fruit harvest. To develop larger plant canopies, flowers at the first branching node were severed (since flowers could host mites, they were left in the pots). In each plot, poles and strings were used to support the non-pruned plants vertically using the Spanish trellis system (Jovicich et al., 2004b). Pollinator arthropods were not released and, other than P. latus no other pest (e.g. whitefly) was noticed on plants (two yellow and one blue sticky cards were hanged above plants in each plot). Treatment plots were visited in a sequence, starting first with those non-in fested, and followed by those with sulfur, non-treated, and with predators. Contact with the plants was alwa ys followed by spraying hands and tools with an ethanol solution. Plants were irrigated with a complete nutrient solution for 2-min events with pressure-compensating drip emitters (average flow/emitter: 37 mLmin-1 at 139 kPa; Netafim, Altamonte Springs, Fla.). Two fertilizer pr oportional injectors (Model DI 16-11; Dosatron International Inc., Clearwater, Fla.) placed in a series were used to pump concentrated stock solutions into the irrigation water (v:v dilution ra te 1:100). Nutrient levels for peppers were adapted from those formulated for hydroponi c greenhouse-grown tomatoes (Hochmuth, 1991; Jovicich et al., 2004b; Jovicich et al., 2007). The irrigation so lution pH ranged from 5.5 to 6.5 and EC from 1.4 to 2.3 mScm-1. Plants received daily volumes that ranged from 0.4 to 2.8 L, with irrigation events programme d on a time schedule and cumula tive solar radiation basis. Changes in delivered volumes were made accordi ng to plant growth developmental stage and on

PAGE 77

77 the volume of solution that drained from the plant containers. This draina ge was kept within a range of 20% to 40% of the irrigation volume delivered the previous day. 3.2.5 Leaf Sampling and Mite Recovery On the first sampling date (0 DAT in Fa ll-2004 and 4 DAT in Spring-2005), mites were counted on top leaves of plants (5 leaves/plo t; non-destructive sampli ng) using a hand-held magnifying lens (10 magnification). Th ereafter, approximately every week, five leaves (third or fourth unfolded leaf below the tip of the shoot, approximately 5-cm long, 9 cm2 when not seriously injured by P. latus ; Figure 3-1) were randomly coll ected from each plot (until 78 DAT in Fall-2004 and 88 DAT in Spring-2005). Leaves we re collected in the mo rnings and stored in a 50-mL vial with a 70% ethanol solution. Mites were later recovered from these leaves and counted under the microscope following the same protocol described in Chapter 2 for recovering mites in pepper seedlings. Leaf area of the sampled leaves was measured (LI-3100 area meter, LI-COR, Lincoln, Nebr.) and used to express mite density (no./cm2). P. latus abundance over time was expressed as cumulative mite-days pe r unit of leaf area (C MD) using calculation procedures discussed in Chapter 2. 3.2.6 Plant Damage Estimation A visual plant injury scale, based on plants infested with P. latus was used to assess the level of damage on plants (Figure 3-1). The pl ant damage indices developed were similar to ones used to evaluate P. latus infestations in other agronomic crops (de Coss-Romero and Pea, 1998; Echer et al., 2002; Pea and Bullock, 1994; Mizobe and Tamura, 2004). The scale for plant damage was comprised of ascending numbers as signed to stages with increased severity of injury symptoms on shoots at the top of the plant canopies (Figure 3-1). A ll plants in the plot were inspected weekly, assigned one of seven da mage indices (an average was calculated for the plot):

PAGE 78

78 0. No injury: youngest leaves light green color, re latively flat, and at an approximate upright angle range 10o (possibly a few P. latus visible on top leaves of the plant) 1. Minor injury: youngest leaves narrower, light green color, and at an upright angle >30o (a few P. latus visible on top leaves of the plant) 2. Moderate injury: one or two young leaves upright and leaves dire ctly below are starting to curl downwards; dull green color on young leaves ( P. latus visible mostly on top leaves but some migrating to lower leaves of the plant) 3. High injury: top leaves curled downwards with dull darker green color; curled leaves with central leaf veins in a cr isscross pattern; shoot tip and flower buds still green ( P. latus visible on all leaves of the plant, a nd on small fruits if present) 4. Severe injury: all top leaves curled downward s with a dark green co lor; youngest leaves thickened, not expanded, with central leaf veins in a crisscross pattern; shoot tip may show necrosis; flowers and flower buds abscising; sm all fruits near the injured shoot may exhibit russet-type scarring ( P. latus visible anywhere on the plant canopy) 5. Very severe injury: all top leav es curled downwards with a dark green color; necrosis of the shoot tip (black color) and flow er buds; no flowers left on shoots; all small fruits near the injured shoot exhibit russet-type scarring ( P. latus visible anywhere on the plant), 6. Extreme severe injury: plant or top of the plant canopy appears necrotic with only some lower (older) leaves remaining with color dark green, brown and/or purpl e tones; small fruits remaining near the injured shoot exhibiting ru sset-type scarring (at this stage only small numbers P. latus on the plant). Plants that reached da mage indices from 4 to 6 grow few or no new shoots, even after P. latus population naturally decl ined or was suppressed. 3.2.7 Fruit and Plant Measurements Fruit were harvested, weighed, and c ounted (Fall-2004: 106 DAT, 31 Dec., center 16 plants of the plot and Spring2005: 88 DAT, 4 June, center 8 plants of the plot). Fruit were graded as marketable (> 60% su rface red, diameter > 56 mm, and blemish-free) or unmarketable fruit that were damaged by P. latus (scars anywhere on the pod surface or peduncle). Fruit with no scars made by P. latus but with physiological disorders that may appear on greenhouse-grown peppers were few and graded as unmarke table (Jovicich et al., 2004a; 2007). Plant growth measurements were taken on 4 plants/plot located at the center of the plant row. These included fresh and dry weight of le aves and stems (dry-weight measurements taken after an oven-drying period of 5 d at 70oC), number of leaves (those with length > 1.5 cm), leaf

PAGE 79

79 area, plant height (measured from the container media surface to the tip of the longest branch), stem diameter (measured at the first leaf node after the cotyledonary node), and number of flower nodes. Fruit set was calculated as the pe rcentage of fruit harvested (those marketable and damaged) relative to the total number of flow er nodes located above the first branching node. 3.2.8 Environmental Conditions Air temperature 5-cm over the plant canopies was recorded from portable data-loggers (WatchDog 100-Temp 2K, Spectrum Technologies, Pl ainfield, Illinois) pl aced in two of the three treatment replications (Figure B-3). Greenhouse curt ains operated automatically (Galileo-Elgal 2000, Galcon, Kfar Blum, Israel) on temperature settings, closing at air temperatures < 18oC. Under the environmental cond itions that existe d throughout both experiments, curtains remained closed during the early mornings, even ings, and nights; when outside wind speed exceeded 16 kmh-1; and during rainfall. 3.2.9 Cost and Benefit Estimation Costs of using N. californicus and sulfur sprays were estimate d for the pest scenarios in the present study assuming a 1-ha crop with 30,000 plants The cost of the management strategies included inputs for predatory mite or pesticide, labor required to re lease the predator or to spray the pesticide, and maintenance and depreciation of a backpack sprayer (Table B-1 and B-2). The cost for scouting plants was not included as it would be the same in pest management programs that use either natural enemies or pesticides (or both pest management systems) (Blockmans, 1999). The gross return ( GR ) for fruit harvested was estimat ed using market price data ( GR = Yield Price ) reported by Jovicich et al. (2005) (Chapter 4). Produce price was not differentiated with regard to pest manageme nt methods (e.g. organic, pesticide-free, or conventional production). Sulfur can be used as a fungicide or miticide when fruit production will be certified as organically-gro wn (U.S. Dept. of Agriculture, 2007b).

PAGE 80

80 Returns relative to production in non-infested plants was calculated with the following formula: nopest pestGR PMC GR where GRpest was the gross return with pest infestation, PMC was the pest management cost, and GRnopest was the gross return in scenar ios with no pest infestation. The coefficient takes a value of one (1) in a scen ario where plants are pest-free. Therefore, coefficients from pest management treatments that are close to one indicate that with a P. latus infestation on transplants, the strategy can lead to returns similar to a scenario with no pest. Additionally, a benefit-cost ratio was calculated as PMC GRpest. This ratio indi cated the expected benefit from paying for selected a pest management strategy. 3.2.10 Experimental Design and Data Analysis Control efficacy by N. californicus in the selected scenario s was evaluated by comparing P. latus number and CMD, and damage index at select ed sampling dates. In both experiments, treatment plots were randomized within three bloc ks that divided the floor area of two bays in the greenhouse. Variables such as mite counts, CMD, and damage index, were measured over time and were analyzed using a mixed model with the SAS PROC MIXED procedure in SAS (SAS Inst., 1999). Treatment, time (DAT), and the treatment time interaction were tested as fixed effects while blocks were treated as random effects. Comparison of covariance structures with a goodness of fit cr iteria was done by PROC MIXED, a nd correlation between time points was specified as an autoregressive process with either a lag period of one [ar(1)] or with heterogeneous variance [arh(1)] (Litt ell et al., 1996). Mite counts and P. latus CMD were log-transformed ( y = ln [ x +1]) to normalize data distribution pr ior to analysis. Actual means and standard errors of the means (SEM) are presen ted in tables and figures. The PDMIX800 SAS macro was used to assign letter groupings to the PROC MIXED LSMeans output (Saxton, 1998).

PAGE 81

81 Means of measured variables were separated after selected pairwise contrasts using the Tukey-Kramer test at = 0.05 (Kramer, 1957). Plant measurements and fruit yield were analyzed by PROC MIXED for randomized complete block design. Measurements expressed as percentages were analyzed after an arcsine tran sformation was performed. Regression analyses were used to estimate relationships between CMD and a) damage index at harvest time, b) plant growth variables, marketable fruit yield, and fr uit set (as percentages relative to plants not infested with P. latus ). These analyses were conducted in each experiment, with pooled data from all treatments and blocks (SPSS Sci., 2002). 3.3 Results 3.3.1 Untreated P. latus Infestations in Fall-2004 and Spring-2005 In Fall-2004 and Spring-2005, treatment ti me interactions were significant ( P < 0.001) for P. latus density (Figures 3-2 and 3-3), and for cu mulative mite-days and visual damage index (Figures 3-4 and 3-5), respectively. In both experiments, plants w ith the untreated infestation of P. latus had pest densities that in creased from 0.14.01 mites/cm2 (0 DAT) to 109 mites/cm2 at 25 DAT in Fall-2004 ( P < 0.05; Figure 3-2A), and from 2.4.23 mites/cm2 (4 DAT) to 76 mites/cm2 at 37 DAT in Spring-2005 ( P < 0.05; Figure 3-3A). The P. latus populations declined shortly th ereafter, and were 0 mites/cm2 at 78 DAT in Fall-2004 (Figure 3-2A) and 0.8.77 mites/cm2 at 88 DAT in Spring-2005 (Figure 3-3A). The average P. latus densities (SD) throughout the sa mpling periods were 20 mites/cm2 in Fall-2004 and 20 mites/cm2 in Spring-2005. Plant damage indices were significantly differe nt from non-infested plants (0) for the first time at 15 DAT (2.9) in Fall-2004 ( P < 0.05; Figure 3-4A) and at 25 DAT (2.3) in Spring-2005 ( P < 0.05; Figure 3-5A). However, the firs t plants with evidence of injury from P. latus were noticed as early as 8 DAT in Fall-2004 a nd 21 DAT in Spring-2005. In both experiments,

PAGE 82

82 damage indices changed concurrently with change s in CMD on plants (Figures 3-4A and 3-5A). In Fall-2004, plants were rated with a da mage index 6 from 30 DAT to 78 DAT ( P > 0.05), and CMD was 1468 at 78 DAT (Figure 3-4A). In Spring-2005, damage indices were 5 from 48 DAT to 88 DAT ( P > 0.05), and CMD was 1529 at 88 DAT (Figure 3-5A). In Fall-2004 and Spring-2005, not managing P. latus on infested transplants led to reductions on all plant growth variables which, with the exception of stem diameter, were 50% or less than measurements in non-infested plants (100%) (Table 3-2, 3-3, B-3, and B-4). In both experiments, infested plants that were not treated for P. latus had no marketable fruit (Table 3-4 and 3-5). In Fall-2004, fruit se t in plants w ith untreated P. latus infestations was 7% (all damaged fruit), and significantly lower than the 22% fruit set in non-infested plants (all marketable fruit) ( P < 0.05; Table 3-5). In Spring-2005, fruit set in plants with untreated P. latus infestations was 14%, similar to non-infested plants ( P > 0.05; Table 3-5). However, in plants not treated for P. latus all set fruit were dama ged (Table 3-5). 3.3.2 Polyphagotarsonemus latus Management in Fall-2004 In Fall-2004, pest densities at transplanting (0 DAT) were not significantly different among treatments that had seedlings infested with P. latus three days before transplanting ( P > 0.05; mean among treatments: 0.13.01 mites/cm2; Figure 3-2A to E). Peak densities of P. latus were not significantly differe nt among plants that had two N. californicus released once at 4 DAT (44.4 mites/cm2 at 20 DAT), twice at 4 and 21 DAT (33.1 mites/cm2 at 25 DAT), or three times at 4, 21 and 30 DAT (39.1 mites/cm2 at 15 DAT) ( P > 0.05; Figure 3-2C, D, E). All three peaks of P. latus density in plants treated with N. californicus were significantly smaller than the peak density in plants with an untreated P. latus infestation (109 mites/cm2 at 25 DAT; all with P < 0.05; Figure 3-2A, C, D, E). At 40 DAT and later, all plants tr eated with N. californicus had P. latus densities of less than 3 mites/cm2 (Figure 3-2C, D, E). In plants

PAGE 83

83 treated with one, two or three releases of N. californicus changes in damage index followed changes in CMD throughout the crop season (Fi gure 3-4C, D, E). After 40 DAT, damage indices on plants in any of the three treatments with N. californicus had large variability and were in the range of 3, with means not significantly different from each other ( P > 0.05; Figure 3-4C, D, E). Plants treated with N. californicus once, twice or three times, had mean plant damage indices that ranged from 2.6 to 4.0 at harvest time (106 DAT). However, these indices were neither different among each other, nor from the index of 6 in plants with untreated P. latus infestations ( P > 0.05; Table 3-2). The treatment time interaction was not significant for density of N. californicus in Fall-2004 ( P = 0.053; P < 0.001 for treatment or time; Figure 32). In the three predator release treatments, N. californicus densities were at less than 0.3 mites/cm2 on top leaves throughout the sampling period (78 d) (Figure 3-2C, D, E). Pr edators were more abunda nt on top leaves after peak densities of P. latus were measured (Figure 3-2C, D, E). When N. californicus was released twice at 4 and 21 DAT, the predator was present at a higher density at 51 DAT (0.3 mites/cm2) than at 8 DAT (0.01 mites/cm2) ( P < 0.05; Figure 3-2D). When N. californicus was released three times (4, 21, and 30 DAT), the pr edator was at higher de nsities at 20 DAT (0.3 mites/cm2) than at 8 DAT (0.01 mites/cm2) ( P < 0.05; Figure 3-2E). Two N. californicus released 4 DAT led to plants with relative plant growth variables that were significantly smaller than in non-infested plants (100%) (Tables 3-2 and B-3). These included fresh weight of leaves (70%), leaf ar ea (60%), stem dry weight (61%), node number (61%), stem diameter (91%), and plant height (58%) ( P < 0.05; Tables 3-2 and B-3). Relative values of most of these plant growth variables did not differ si gnificantly from percentages in plants with an untreated P. latus infestation ( P > 0.05; Table 3-2), nor fr om plants that had two

PAGE 84

84 N. californicus released twice (4 and 21 DAT) or three times (4, 21, and 30 DAT) ( P > 0.05; Table 3-2). Plants that had two N. californicus released once (4 DAT), twic e (4 and 21 DAT) or three times (4, 21, and 30 DAT) had similar marketable fruit yields ( P > 0.05; Table 3-4). These yields were greater than in plants not treated for P. latus (0 kgm-2; P < 0.05), but smaller than in plants that were not infested with P. latus (4.6 kgm-2 or 100%; P < 0.05) (Table 3-4). When two N. californicus were released on plants 4 DAT, relative marketable yield (45%) was significantly smaller than in plants not infested with P. latus ( P < 0.05; Table 3-4). When two N. californicus were released twice (4 and 21 DAT) or three time s (4, 21, 30 DAT), relative yields were reduced to 68% and 64%, respectively ( P < 0.05; Table 3-4). Four sulfur sprays initiated 16 DAT (mean damage index 2.4) kept P. latus at pest population densities at less than 5 mites/cm2 beyond 25 DAT, after a peak density of 34.0 mites/cm2 at 15 DAT (Figure 3-2B). At 78 DAT, plan ts treated with sulfur had 390 CMD which was less than 1529 CMD in plants with untreated P. latus infestations ( P < 0.05; Table 3-2), but not different from CMD in any of the plants treated with N. californicus ( P > 0.05; Table 3-2). In plants treated with the four sulfur sprays, yield relative to noninfested plants was reduced to 87%, which was greater than yiel ds in plants treated with N. californicus released once (4 DAT), twice (4 and 21 DAT), or three times (4, 21, and 30 DAT) ( P < 0.05; Table 3-4). 3.3.2 Polyphagotarsonemus latus Management in Spring-2005 In Spring-2005, pest densities at 4 DAT (first sampling date for all treatments) were not significantly different among treatments that had seedlings infested with P. latus ( P > 0.05; mean among treatments 1.9.97 mites/cm2; Figure 3-3A to I). During th e 88-d crop, plants that had four N. californicus released 0, or 4 DAT, or had two N. californicus released DAT, did not have significantly different P. latus densities on top-canopy leaves ( P > 0.05; Figure 3-3D,

PAGE 85

85 G, H, I). Mean pest density (SD) was 1.0.3 mites/cm2 in the 88-d crop, and densities did not exceed 9 mites/cm2 (Figure 3-3D, G, H, I). At 88 DAT the latter treatments did not have significantly different CMD; CMD ranged from 18 to 117 ( P > 0.05; Table 3-3 and Figure 3-5D, G, H, I). Throughout the season, damage indices in these treatments were not different from the index 0 in non-infested plants ( P > 0.05; Figure 3-5D, G, H, I). When two N. californicus were released on plants 0 DAT or 4 DAT, P. latus density increased from 2.3.05 mites/cm2 or 3.1.46 mites/cm2 at 4 DAT, respectively, to 22.1 mites/cm2 or 32.3 mites/cm2 at 33 DAT, respectively ( P < 0.05; Figure 3-3E, F). At 88 DAT, CMD in plants that had two N. californicus released at 0 DAT (321) or at 4 DAT (524) were not significantly different from each other ( P > 0.05), nor were they different from 1468 CMD in plants not treated for P. latus ( P > 0.05; Table 3-3). However, these plants had lateral shoots (undamaged) that grew after P. latus densities had decreased 50 DAT (Figure 3-3E, F). Moreover, at 88 DAT, the damage index on top canopies was 0.2, which was not significantly different from the index 0 in non-infested plants ( P > 0.05; Table 3-3 and Figure 3-3E, F). The treatment time interaction for density of N. californicus was significant in Spring-2005 ( P < 0.002; Figure 3-3). N. californicus densities were similar and less than 0.05 mites/cm2 throughout the sampling period (88 d) in plants that had four N. californicus released 0, or 4 DAT, or that had two N. californicus released DAT ( P > 0.001; Figure 3-4D, G, H, I). When N. californicus was released on plants 4 DAT (Figure 3-3F), greater densities of predators were de tected at 37 DAT (0.11 mites/cm2) and at 40 DAT (0.22 mites/cm2) than in other scenarios with N. californicus (< 0.08 mites/cm2) ( P < 0.05; Figure 3-3D, E, G, H, I).

PAGE 86

86 Plants that had four N. californicus released DAT had all measured plant growth variables that were comparable to those in plants not infested with P. latus (Tables 3-3 and B-4). However, with four N. californicus released 4 DAT, or with tw o predators released 6 DAT, plants had some of the plant gr owth variables that were smalle r (i.e., average reductions of 19% in fresh weight of leaves and stems, number of leaves, and leaf area) than non-infested plants (Table 3-3). Plants with two predators rele ased 4 DAT had growth variables that were significantly smaller than non-infested plants (100 %). These included fresh and dry weight of leaves (both 61%), number of leaves (64%), leaf area (47%), and fresh an d dry weight of stems (49 and 53%, respectively) ( P < 0.05; Table 3-3). The single releases of N. californicus led to plants with fruit yields that were comparable to non-infested plants (5.0 kgm-2) when four predators per plant were released 0, or 4 DAT, or when two predators per plan t were released DAT ( P > 0.05; Table 3-4). When two N. californicus were released 4 DAT, marketable yiel d relative to non-infe sted plants (100%) was reduced significantly to 25% ( P < 0.05; Table 3-4). The number of damaged fruits on plants that had two N. californicus released on plants 4 DAT was 7. 4 per square meter, which was greater than the mean 0.3 damaged fruit per square meter in plants that had four predators released 0, or 4 DAT ( P > 0.05; Table 3-4). In Spring-2005, five sprays of sulfur, app lied weekly, starting or 0 DAT maintained P. latus densities on top-canopy leaves at less than 2 mites/cm2 from 4 DAT to 37 DAT (Figure 3-3B, C). Nonetheless, there were indications that the pest could resurge af ter the sulfur spray programs were terminated at 28 or 30 DAT. P. latus increased only in some plants in sulfur-treated plots, therefore mean densities after 33 DAT were not significantly different from densities before day 33 after transplanting ( P > 0.05; Figures 3-3B, C, and 3-5B, C). By 37

PAGE 87

87 DAT, set fruits that were later harvested at 88 DAT were not damaged by the resurging P. latus populations (Table 3-5). At 88 DAT plants treated with five sulf ur sprays started or 0 DAT had damage indices, a majority of plant grow th variables, and relative yield that were comparable to plants that were not infested with P. latus ( P > 0.05; Tables 3-3, 3-5, and B-4). 3.3.5 Cost and Benefit of P. latus Management Cost of using N. californicus changed according to the predator release density (2 or 4/plant) and to the timing of rel eases (Table 3-6). Labor costs we re 60% less when releases were carried out in seedlings prev ious to transplanting, as compared to releases in plants set in the fruit production greenhouse. When using N. californicus (pest scouting not included), shipping costs (handling and air freight from Calif. to Fla.) accounted for 50 to 56% of the total cost of managing P. latus This contrasted with 37 to 41% for th e cost of the predator, and 3 to 13% for the cost of labor required for re leasing the predator (Table B-2). Return relative to non-infested plants was 0.92 when four pr edators/plant were released preventively ( DAT); using this pest management strategy resulted in re turns that were neither different from a scenario with no P. latus nor from scenarios where sulfur or four predators/plant were used at transplanting ( P > 0.05; Table 3-6). In pest management scenarios that led to yields comparable to those in non-infested plants (four N. californicus /plant released 6, 0, or 4 DAT, two N. californicus /plant released or five sulfur sprays started on plants at or 0 DAT) greater benefit-cost ratios were calcu lated for sulfur sprays (387 $/$) than for releases of N. californicus (115 $/$) (Table 3-6). The cost of a preventive release of four N. californicus on seedlings was comparable to the gross value from 7.8 g of fresh colored bell pepper [$0.162/m2 ($4.7/kg 4.4 kgm-2) 103 gkg-1], while the cost of five sprays of sulfur was comparable to the gross value of 1.9 g of fruit [$0.038/m2 ($4.44/kg 4.4 kgm-2) 103 gkg-1], assuming fruit prices for Spring-2005 (Table 3-6).

PAGE 88

88 3.3.6 Relationships Between Cumulative Mi te-Days and Plant Damage Index, Plant Growth Variables and Ma rketable Fruit Yield With data pooled in Fall-2004 and in Spring2005, relationships were determined between CMD at 40 DAT and a) damage index, at harvest time, b) plant growth variables, marketable fruit yield, and fruit set (as percentages relative to plants not infested with P. latus ) (Figure 3-6 and Table 3-7). Since CMD did not increase af ter 40 DAT in all treatments (Figures 3-4 and 3-5), CMD within this early crop period was select ed for determination of relationships [(the two sulfur treatments in Spring-2005 did not reduce mark etable fruit of first harvest when compared to non-infested plants (Table 3-3)]. In general, estimated trends for va riables in Fall-2004 were similar to Spring-2005 (Figure 3-6). Damage indices on plants incr eased with exponential rise to maximum functions ( r2 = 0.54 in Fall-2004 and r2 = 0.60 in Spring-2005), with damage indices slightly higher in Fall-2004 than in Spring-2005 (Fig 3-6A and Table 3-7). Marketable yield followed exponential decay functions as CMD increased ( r2 = 0.81 in Fall-2004 and r2 = 0.78 in Spring-2005) (Figure 3-6B and Table 3-7). For ex ample, in the Spring-2005, plants not infested with P. latus had a marketable fruit yield of 5.0 kgm-2. However, yields in P. latus -infested plants were estimated to decrease 5, 10, 50, a nd 95% (resulting in gro ss return losses of 1.18, 2.22, 11.10 and 21.11$/m2, respectively) with 24, 48, 315, and 1370 CMD at 40 DAT, respectively ( r2 = 0.78). Marketable fru it set decreased rapidly af ter 500 CMD with quadratic functions ( R2 = 0.52 in Fall-2004 and R2 = 0.66 in Spring-2005) (Figur e 3-6C and Table 3-7). Plant growth variables were redu ced at different rates by CMD (Fi gure 3-6D to L). For example, in Spring-2005 at 40 DAT, 200 CMD led to esti mated reductions of 12% in leaf area ( r2 = 0.65), 5% in plant height ( r2 = 0.68), and 2% in stem diameter ( r2 = 0.59).

PAGE 89

89 3.4 Discussion Plants were severely damaged in less than a month after transplanting when P. latus was left unmanaged (Figs. 3-4A and 3-5A). Pest populations later decreased in both experimental crops as the necrotic plants were no longer a su itable host. The production of any marketable fruit was expected to be minimal after the first harvest of damaged fruit. P. latus density on top-canopy leaves followed changes similar to thos e reported in pepper plants where the pest was left unchecked (Coss-Romero and Pea, 1998; Fan and Petitt, 1994; Mizobe and Tamura, 2004; Weintraub et al., 2003). In both seasons, temp eratures over the crop canopy were optimum for P. latus populations to develop (Jones and Brown, 1983; Li et al., 1985; Li and Li, 1986; Silva et al., 1998). The more rapid increase of P. latus CMD early in the season of Fall-2004, as compared to Spring-2005 (Figs. 3-4A and 3-5A ), may have been associated with warmer weather conditions in Fall-2004, when daily mean air temperature was 32oC as compared to 28oC in Spring-2005 during the first month of growth (Figure B-3). In both of these seasons, CMD values in non-treated plants (1529 and 1468 CMD based on cm2 of leaf; 0% relative yield) were greater than those reported for P. latus -infested bell peppers grown in tunnels with no pest management in a desert region [1438 CMD based on whole leaves, and 29% re lative yield; leaf area sampled not reported (Weintraub et. al, 2003)]. The populations of P. latus and their damaging effects on pl ant growth variables and fruit yield were limited by N. californicus and by sulfur sprays. The effectiveness of N. californicus in controlling P. latus was associated with both the density of predators released and the time that had elapsed between the day when P. latus infested plants and the day when N. californicus were released (Figure B-4). By either delayi ng or releasing fewer pred ators/plant, CMD caused plants to have reduced growth variables at harv est time (Tables 3-2 and 3-3). In some scenarios (i.e. two N. californicus released per plant at DAT, or fo ur per plant at 4 DAT), plants had

PAGE 90

90 leaf and stem fresh weights, number of leaves, a nd leaf area that were reduced in comparison to non-infested plants (a 17% re duction on average); however, repr oductive organs in the plant were not damaged and marketable yield wa s not reduced (Table 3-5). Because two N. californicus released at 4 DAT were ineffective to avoid yield loss, it can be expected that additional releases of predators after 4 DAT would be needed ear lier than 21 DAT (the time of the second predator release in Fall-2004, when plants had flowers on second nodes on anthesis) (Figure 3-4C, D, E). However, increasing the number of predator s released per plant in curative releases would likely produce lo w return-cost ratios. In the Spring-2005 study, an average of two or less motile stages of female P. latus and eggs were expected to be on the plants at transplanting (0 DAT), three days after seedlings were artificially infested with two female P. latus A week after the infestation in Spring-2005 (4 DAT; the first mite count), prey-predator dens ity ratios were 0.85:1 and 0.43:1 (based on cm2 of sampled leaves in relation to the number of predators released on the plant) in plants with two and four N. californicus released, respectively [(mean 1.7 P. latus /cm2) (2 or 4 N. californicus /plant); Figure 3-3]. If pest density on a whole leaf (9 cm2/leaf) were determined (15.3 P. latus /leaf), mean prey-predator ratios values would have corresponded to 7.7:1 and 3.8:1 for releases of two or four N. californicus /plant carried at 4 DAT, re spectively. These ratios on leaves had to be even smaller when either two or four N. californicus were released at transplanting (0 DAT). In anot her study (Chapter 2), initial prey -predator ratios of 3:1 or less (mite numbers based on whole pepper seedlings) were reported to be effective to avoid damaged transplants when seedlings were infested with two P. latus and had preventative releases of two N. californicus The small values of these prey-predator ratios, difficulties associated with measuring such small densities of P. latus on plants, and the significant crop damage that occurs

PAGE 91

91 with delayed predator releases, indi cate that a preventative release of N. californicus may be most effective for managing P. latus at feasible costs. Micronized sulfur sprays were effective in rapidly reducing P. latus number and in maintaining the pest at low dens ities (Figures 3-2B and 3-3B, C). The effectiveness of managing the pest with sulfur was also dependent on when the spray program was started with respect to the initial infestation. Fruit yi eld after the first harvest may be decreased if the weekly spray program is interrupted. In both seasons, mi cronized-sulfur used at the rate of 3.4 kgha-1 did not cause phytotoxicity on pepper. Delaying a first sulf ur application until a visual damage index of 2 was too late to avoid early yi eld reduction (Table 3-4). The estimated cost of releasing four N. californicus before transplanting ($0.16/m2, for a density of 3 plants/m2 after transplanting) was 400% of the co st of five sprays of sulfur started on seedlings ($0.04/m2). Both of these pest management stra tegies led to marketable yields that were similar to yields in plants not infested with P. latus (Tables 3-5 and 3-6), but returns relative to non-infested plants were more similar to returns from a pest-free scenario (non-infested plants) with th e preventive rele ase of four N. californicus at DAT than with the five sulfur sprays (Table 3-6). The difference in cost between the two pest management methods used in these experiments would decrease in a long-season crop if additional sulfur sprays are required. Additionally, if sprays are continued until harvest, sulfur residues on fruits will need to be washed to be marketable, an additional cost which was not considered in this study. In the U.S, a greater demand for N. californicus from industries that use this predator and that are larger industries than greenhouse pe pper (i.e. greenhouse-grown or namentals and field-grown strawberry) may increase the number of companies rearing this predatory mite; thus, leading to reduced prices of the predator. Stevens et al. (2 000) reported costs that were three times greater

PAGE 92

92 with biological control as compared to pesticid es in an ornamental crop grown in greenhouse. However, in regions where biological control is co mmonly used, the differences in cost are either minimal or costs are greater using pesticid es (Blockmans, 1999; van Lenteren, 2006; Van Driesche and Heinz, 2004). Moreover, in an in tegrated pest management program (IPM), the decision to use a particular pest management st rategy is not based on its cost only (Blockmans, 1999). Releasing predators on pepper seedlings is critical in an IP M program that aims to reduce pesticide use and increase the use of natural enem ies. The release of an effective natural enemy of P. latus is critical to avoid repeated pesticide sprays against this pest (Dik et al., 1999; Fan and Petitt, 1994; Gerson, 1992). Occasiona l sprays of micronized sulfur directed to the top of plant canopies (Hassan and Van de Ve ire, 2004) and, possibly, in co mbination with the use of N. californicus might be an effective IPM program for P. latus Fruit yield and plant growth variables were a function of cumulative mite-days (Fig. 3-6). In exponential decay functions, rapid reducti ons occurred for marketable fruit yield, photosynthetic area, and fresh and dry weight of stem as CMD in creased (Table 3-7). In both experiments, stem diameter had the smallest relative reductions by CMD among plant growth variables (Fig. 3-6I and Tabl e 3-7). Yield loss was due to fruit that were damaged by P. latus (Tables 3-4 and 3-5), and after ne ar 500 CMD, yield loss was the re sult of fruit that did not set (i.e., damaged flower buds, flowers, and small fruits that aborted) (Fig. 3-6C). For P. latus infestations initiated at transpla nting, less than 24 CMD (based on a cm2 of top-leaf sampled) should exist on plants 40 DAT to avoid a reduction of early yiel d of near 5% (a loss of $1.11/m2 based on a early fruit harvest of 5 kgm-2 in non-infested plants) (Figure 3-6B). Yield reduction was greater when higher CM D values were reached early in the crop season. For example, in Spring-2005, 339 a nd 384 CMD in sulfur-treated plants had

PAGE 93

93 accumulated mostly past the last sprays (after 40 DAT), and did not reduc e early fruit yield as compared to non-infested plants (Figure 3-5B, C and Table 3-5). By contrast, in the same experiment, a level of 321 CMD at harvest re duced the relative yield to 66% when two N. californicus per seedling were released at transplanting; the CMD in this scenario was accumulated early during the crop season (Figure 3-5E). P. latus infestations will develop more rapidly in young pepper plants, where they also ca use more damage on plant tissue than at older developmental stages (Coss-Romero and Pe a, 1998; Chapter 2). The exponential decay relationships calculated with pooled data from various trea tments indicate the importance of avoiding the presence of P. latus on plants prior to 40 DAT to avoid early plant damage and yield reduction. Estimates of mite populations on top-ca nopy leaves can be more erratic for N. californicus than for P. latus (Figures 3-2 and 3-3). Although P. latus is difficult to detect at low densities, it aggregates on the top young leaves of pepper plants particularly during the initial stages of an infestation (Gerson, 1992). By contrast, N. californicus and other phytoseiids used for P. latus management (i.e. N. cucumeris ), have an ambulatory behavior and frequently inhabit and reproduce on the abaxial blade of larg er leaves, near the ba se of the midrib vein of leaves (i.e. in domatia) (Auger et al., 1999; Castin eiras et al., 1997; Faraji et al.; 2002; Weintraub et al., 2004). Therefore, a sampling protocol for estimating the density of N. californicus for P. latus management would need to include leaves from lower parts of the plant canopy. In this study, even when few or no P. latus were recovered from top leaves of plants that previously had N. californicus released, the predator wa s found on the underside leaves at the middle height of the plant canopy. Weintraub et al ., (2004) reported that leaf temperature had an effect on the distribution of N. cucumeris within pepper plants. Environmen tal variables and leaf structure

PAGE 94

94 need to be considered when deciding on where in the plant N. californicus should be scouted and, when during the day, leaves should be insp ected. This information is essential for monitoring N. californicus and for predicting success or failure in managing P. latus In greenhouses where pepper transplants are likely to host a few P. latus releasing four N. californicus /seedling either a week before transplant ing, or at transplanting, may represent a viable biological contro l strategy to minimize plant injury an d avoid fruit yield loss in early and, potentially, in full-season production. N. californicus survives on seedlings and plants in the greenhouse even if broad mites ar e absent during transplant pr oduction or after transplanting (Chapter 2; Castagnoli and Simoni, 1999; de Cour cy Williams et al., 2004; Palevsky et al., 1999; Walzer et al., 2001). Moreover, N. californicus will contribute to pest protection of seedlings and transplanted plants by also preying on other phytophagous mites that may be present, such as T. urticae (Gillespie and Raworth, 2005; Griffith s, 1999; Osborne et al., 1998; 2005). The scenarios in which P. latus can infest a pepper crop are numerous. Among the many factors that define the pest s cenarios are the initial pest popul ation density and its rate of migration, environmental conditions, developmenta l growth stage of the pepper plant, and the pepper genotype (Chapter 2; Coss-Romero and Pea, 1998; Echer et al., 2002; Fan and Petitt, 1994). A single predator release recommendation is not adequate for all pest scenarios. P. latus infestations initiated on pepper seedlings are recurrent in gree nhouse crops in the tropics and subtropics. With the pest on seedlings, the risk of losing the pepper crop is at its highest. In this study, it was observed that if the entire transpla nted pepper crop becomes infested with initial pest densities as low as two P. latus /seedling, intervention with N. californicus has to be implemented within a short period (e.g. less than a w eek) of the initial day of infestation. In real crop scenarios, the moment when initial pest infestations occur is not known; therefore,

PAGE 95

95 preventative releases of N. californicus will be more effective and economically viable than post-infestation releases. N. californicus will search and locate the prey (Auger et al., 1999; Collier et al., 2001) in the transplant nursery. Therefore when only a small number of seedlings host P. latus a small prey-predator density ratio may o ccur in infested seedlings, leading to an effective and rapid control of the pest. 3.5 Summary If not controlled, a few undetected broad mites [ Polyphagotarsonemus latus (Banks)] in bell pepper ( Capsicum annuum L.) seedlings rapidly increase in numbers and damage plants and reduce yield of high-value colored-fruit produced in greenhouses. In two experiments, Fall-2004 and Spring-2005 (106 and 88-day crops), pest management with the predacious mite Neoseiulus californicus (McGregor) was investigated in pe pper plants that had low-density P. latus infestations (2 mites/seedling) initiated on s eedlings, three days before transplanting in a greenhouse. Various pest-predator scenarios were compared to sulfur spray programs, non-managed infestations, and no infestations until a first-single harvest of fruit. In Fall-2004, plants were treated with a) two N. californicus released on each P. latus -infested seedling, either 4 days after transplant ing (DAT), 4 and 21 DAT, or 4, 21, and 30 DAT, or b) four sulfur (3.4 kgha-1) sprays applied weekly, started at first symptoms of plant in jury (16 DAT). In Spring-2005, seedlings were treat ed with a) two and four N. californicus released either DAT, 0 DAT, or 4 DAT or, b) five sulfur sprays a pplied weekly, started DAT or 0 DAT. Plants with untreated P. latus infestations were seriously damage d and had no marketable yield after 1468 cumulative mite-days/leaf-cm2 (CMD, an estimate of pest presence over time) in Fall-2004 or 1529 CMD in Spring-2005. In Fall-2004, fruit yi eld and plant growth variables relative to non-infested plants (yield 4.6 kgm-2 or 100%; 0 CMD), were reduced in plants treated with two N. californicus released either once 4 DAT (yield 45% ; 735 CMD), twice at 4 and 21 DAT (yield

PAGE 96

96 68%; 566 CMD), or three times at 4, 21 and 30 DAT (yield 64%; 581 CMD). In Spring-2005, yields were reduced when two N. californicus were released either 0 DAT (yield 66%; 321 CMD) or 4 DAT (yield 25%; 534 CMD). Howe ver, in Spring-2005, a single release of N. californicus led to undamaged plants and fruit yields that were comparable to non-infested plants (5.0 kgm-2 or 100%) when four predators/plant were released either 0, or 4 DAT (yields 88%; 18 CMD), or when two N. californicus were released DAT (yield 87%; 105 CMD). In plants with these e ffective preventative releases of N. californicus prey-predator density ratios on terminal leaves were <1:1 plant mites : cm leaf mites2 or equivalent <4:1 plant mites : leaf mites, four days after transplanting, respectively. In Fall-2004, relative yield was reduced to 87% (390 CMD) when the first sulfur spray was delaye d until 16 DAT, but in Spring-2005 yields were comparable to non-infested plants when sulfur sp rays were started either or 0 DAT (yields 87%; 384 CMD). There was indi cation of pest resurgence afte r the last sulfur spray in Spring-2005. With data pooled, plant growth variables and marketable fruit yield were decreased by CMD on the first 40 DAT in expone ntial decay relationshi ps in both seasons; a reduction in marketable yield 5% was estimated with near 24 CMD on top leaves ( R2 = 0.78). Return-cost ratios were great er with sulfur than with N. californicus ; however, since biological control programs are disrupted by the multiple pesticide sprays used against P. latus N. californicus will be critical in IPM of greenhouse-grown peppers. If seedlings to be transplanted in the green house are likely to host two P. latus /seedling or less, release of four N. californicus per seedling either a week before transpla nting or at transpla nting may represent a viable biological control strate gy to minimize plant injury and avoid fruit yield loss in early and, potentially, in full-season prod uction. Intervention with four N. californicus has to be implemented within a short period (e.g. less than a w eek) of the initial day of infestation. In real

PAGE 97

97 crop scenarios, the moment when initial pest infestations occur is not known; therefore, a preventative release of N. californicus will be more effective a nd economically viable than post-infestation releases.

PAGE 98

98 Table 3-1. Biological ( N. californicus releases) and pesticide (s ulfur sprays) treatments evaluated for management of P. latus in greenhouse-grown pepper plants in Fall-2004 and Spring-2005. In both experiments, plants were artificially infested with two P. latus three days before transplanting. Time of predator release or pesticide spray P. latus treatment (predators released per plant) After transplanting (d) After infestation with P. latus (d) Treatment codea Fall-2004b N. californicus (2) +4+7 2N+4 N. californicus (2) +4, +21+7, +24 2N+4,21 N. californicus (2) +4, +21, +30+7, +24, +33 2N+4,21,30 Sulfur +16, +22, +29, +36+19, +25, +32, +39 4S+16 Non-treatedc ---Non-treated Non-infestedd ---Non-infested Spring-2005e N. californicus (2) 2N N. californicus (2) 0+3 2N0 N. californicus (2) +4+7 2N+4 N. californicus (4) 4N N. californicus (4) 0+3 4N0 N. californicus (4) +4+7 4N+4 Sulfur 0, +7, +14, +21, +3, +10, +17, +24 5S Sulfur 0, +7, +14, +21, +28+3, +10, +17, +24, +31 5S0 Non-treated ---Non-treated Non-infested ---Non-infested aFirst number indicates predator release density, or nu mber of pesticide sprays (applied every 7 days); N for N. californicus or S for micronized-sulfur spray (rate: 3.4 kgha-1); + or followed by numbers indicate number of days from transplanting date, when predators were released or sulfur was sprayed for the first time. bTransplanting on 17 Sept. 2004 and first-single fruit harvest on 1 Jan. 2005 (106 DAT). cPlants infested with P. latus and no pest management strategy. dPlants with neither P. latus nor pest management strategy. eTransplanting on 9 Mar. 2005 and first-single fruit harvest on 5 June 2005 (88 DAT).

PAGE 99

99 Figure 3-1. Pepper plant damage indi ces used to assess damage from P. latus feeding. Scale indices ranged from zero (no in jury) to six (extreme severe injury). A description of symptoms is included in the Materials and Methods section. The arrow indicates the leaf sampled for mite recovery and count. 4 0 1 2 3 5 6

PAGE 100

100 Figure 3-2. Number of P. latus and N. californicus recovered per unit of area of sampled leaf in greenhouse-grown pepper plants in Fall-2004. Seedlings were infested with two P. latus 3 d before transplanting and were treated with N. californicus releases and sulfur sprays. Treatments were: infested non-tr eated (A), 4 sulfur sprays initiated 16 d after transplanting (DAT) when symp toms were evident (B), and two N. californicus released after the P. latus infestation at, 4 DAT (C), a second release 21 DAT (D), or a third release 30 DAT (E). Da ta for non-infested plants (no P. latus ) not plotted. First mite count made at 0 DAT. Arrows represent time of release ( N ) or sulfur spray ( S ). Plotted for each sampling date: mean SE, n = 3 (from 3 replications, one treatment replicate is the average of 5 leaves sampled from 16 plants). Treatment time interactions for P. latus density: F = 8.1; df = 75, 23.8; P < 0.0001; and for N. californicus density: F = 1.4; df = 75, 107; P < 0.0526.

PAGE 101

101 Figure 3-3. Number of P. latus and N. californicus recovered per unit area of sampled leaf in greenhouse-grown pepper plants in Spring-2005 Seedlings were infested with 2 P. latus 3 d before transplanting and were treated with N. californicus releases and sulfur sprays. Treatments: infested non-tr eated (A), 5 sulfur sprays initiated d from transplanting (DAT) (B) or 0 DAT (C), and one-time releases of N. californicus at a per plant predator density of either 2 (DF) or 4 (G I), at DAT (D and G), 0 DAT (E and H), or 4 DAT (F and I). Data for non-infested plants (no P. latus ) not plotted. First count at 4 DAT. Mean SE, n = 3 (from 3 replications, 5 leaves/rep. of 8 plants). Treatment time interactions for P. latus density: F = 3.01; df = 108, 66.5; P < 0.0001; and for N. californicus density: F = 1.7; df = 108, 144; P < 0.0020.

PAGE 102

102 Figure 3-4. Polyphagotarsonemus latus cumulative mite-days per uni t of area of sampled leaf and plant damage index in greenhouse-grown pepper plants in Fall-2004. Seedlings were infested with two P. latus 3 d before transplanting and were treated with N. californicus releases and sulfur sprays. Treatm ents were: infested non-treated (A), four sulfur sprays initiated 16 d after tr ansplanting (DAT) when damage symptoms were evident (B), and two N. californicus released 4 DAT (C), a second release 21 DAT (D), or a third release 30 DAT (E). Arrows represent time of release ( N ) or sulfur spray ( S ). Data for non-infested plants (no P. latus ) not plotted. Plotted for each sampling date: mean SE, n = 3 (from 3 replications, one treatment replicate is the average of 5 leaves sampled from 16 pl ants). Treatment time interactions for P. latus cumulative mite-days: F = 12.2; df = 75, 139; P < 0.0001; and for plant damage index: F = 3.1; df = 75, 176; P < 0.0001.

PAGE 103

103 Figure 3-5. Polyphagotarsonemus latus cumulative mite-days per uni t of area of sampled leaf and plant damage index in greenhouse-gro wn bell pepper plants in Spring-2005. Seedlings were infested with 2 P. latus 3 d before transplanti ng and were treated with N. californicus releases and sulfur sprays. Treatm ents were: infested non-treated (A), 5 sulfur sprays initiated d from transplanting (DAT) (B) or 0 DAT (C), and one-time releases of N. californicus at a per plant predator density of either 2 (DF) or 4 (GI), at DAT (D and G), 0 DAT (E and H), or 4 DAT (F and I). Data for non-infested plants (no P. latus ) not plotted. First count at 4 DAT. Mean SE, n = 3 (from 3 replications, 5 leaves/rep. of 8 pl ants). Treatment time interactions for P. latus cumulative mite-days: F = 7.2; df = 108, 236; P < 0.0001; and for plant damage index: F = 3.8; df = 144, 296; P < 0.0001.

PAGE 104

104Table 3-2. Polyphagotarsonemus latus cumulative mite-days, relative plant growth vari ables to non-infested pl ants, and plant visual damage index in greenhouse-grown pepper plants in Fall -2004 (106 DAT). Plants we re treated to manage a P. latus infestation initiated three days before transplanting. Leaves Stems Treatment codea Cumulative mite-days (mite-days/ cm2) Fresh weight (%) Dry weight (%) Number (%) Area (%) Fresh weight (%) Dry weight (%) Node number (%) Diameter (%) Plant height (%) Damage index (0) N. californicus 2N+4 735 ab 70 bc 84 ab 79ab 60bc 54 ab 61b 61bc 91bc 58cd 4.0ab 2N+4,21 566 ab 63 bc 80 b 70ab 66bc 49 ab 55bc 68abc 9 0 bc 55cd 2.6ab 2N+4,21,30 581 ab 70 bc 86 ab 87ab 77abc 71 ab 67b 73abc 8 7 bc 63c 2.9ab Sulfur 4S+16 390 b 84 ab 88 ab 88ab 93ab 82 ab 76b 92ab 9 4 ab 87b 0.3bc Non-treated 1529 a 40 c 47 c 33c 24c 28 c 21c 38c 7 2 c 40d 6.0a Non-infested 0 c 100 a 100 a 100a 100a 100a 100a 100a 10 0 a 100a 0.0c Significance F5, 10 P 12.2 0.0001 11.2 0.0008 3.8 0.0334 10.5 0.0010 8.5 0.0023 13.5 0.0004 19.1 0.0001 7.7 0.0034 9.3 0.0016 71.2 0.0001 5.1 0.0143 Means within the same column followed by different lette rs are significantly different based on Tukey-Kramer test, P < 0.05. a Treatments were: two N. californicus released 4 DAT (2N+4), a second release 21 DAT (2N+4,21), or a third release 30 DAT (2N+4,21,30); four sulfur sprays initiated 16 DAT (4S+16); plants infested with P. latus and no pest management (Non-treated); plants with no P. latus (Non-infested).

PAGE 105

105Table 3-3. Polyphagotarsonemus latus cumulative mite-days, relative plant growth vari ables to non-infested pl ants, and plant visual damage index in greenhouse-grown pepper plants in Spri ng-2005 (88 DAT). Plants we re treated to manage a P. latus infestation initiated three days before transplanting. Leaves Stems Treatment codea Cumulative mite-days (mite-days/c m2) Fresh weight (%) Dry weight (%) Number (%) Area (%) Fresh weight (%) Dry weight (%) Node number (%) Diameter (%) Plant height (%) Damage index (0) N. californicus 2N 105 bcd 83 bcd84abc86bcd81bc 84bc 87 abc80abcd 95ab 94a 0.0b 2N0 321 abc 79 cd 77bc 77cd 71cd 75cd 76 bc 71bcd 92ab 92a 1.3b 2N+4 524 ab 61 de 61cd 64d 47de 49de 53 cd 51cd 91ab 82ab 0.2b 4N 18 d 98 ab 98ab 97ab 98ab 97ab 98 ab 92abc 104a 98a 0.0b 4N0 28 cd 94 abc93ab 87bcd92abc 78bcd75 bc 87abc 95ab 88ab 0.0b 4N+4 117 bcd 82 bcd88abc85bcd82bc 72cd 73 bc 88abc 95ab 94a 0.0b Sulfur 5S 384 abcd 84 bcd88abc84bcd85abc 84bc 81 abc87abc 99a 95a 1.7b 5S0 339 abc 93 abc98ab 95abc93bc 88abc96 ab 88abc 101a 93a 1.0b Non-treated 1468 a 40 e 43d 31e 25e 34e 31 d 31d 78b 56b 5.0a Non-infested 0 e 100 a 100a 100a 100a 100a 100 a 100a 100a 100a 0.0b Significance F9, 18 P 7.2 0.0001 16.9 0.0001 11.4 0.0001 17.6 0.0001 20.8 0.0001 16.5 0.0001 11.1 0.0001 6.9 0.0003 5.0 0.0019 6.1 0.0006 9.7 0.0001 Means within the same column followed by different lette rs are significantly different based on Tukey-Kramer test, P < 0.05. a Treatments were: one-time releases of N. californicus at a per plant predator density of either two (2N; 2N0; 2N+4) or four (4N; 4N0; 4N+4), with release time either at 0, or 4 DAT, r espectively; five sulfur sprays initiated either DAT (5S) or 0 DAT (5S0); plants infested with P. latus and no pest management (Non-treated); plants with no P. latus (Non-infested).

PAGE 106

106 Table 3-4. Marketable fruit yield, P. latus -damaged fruit, and marketable fruit set in greenhouse-grown pepper plants in Fall-2004 (106 DAT). Plants were treated to manage a P. latus infestation initiated three days before transplanting. Marketable fruit yield Weight Number Treatment codea (kgm-2) (%) (no./m2) (%) Damaged fruit number (no./m2) Fruit set (%) N. californicus 2 N +4 2.1 c 45 d 8.6 c 40d 8.6 a 28.2a 2 N +4,21 3.2 bc 68 c 13.2 abc 63c 4.7 ab 26.3a 2 N +4,21,30 3.0 bc 64 c 12.9 bc 60cd 2.3 ab 21.1a Sulfur 4 S +16 4.1 ab 87 b 17.9 ab 84b 1.1 b 18.4a Non-treated 0.0 d 0 e 0.0 d 0e 2.6 ab 6.8b Non-infested 4.6 a 100 a 21.5 a 100a 0.0 b 21.5a Significance F5, 10 P 35.1 0.0001 218.7 0.0001 19.8 0.0001 138.2 0.0001 4.4 0.0221 188.8 0.0010 Means within the same column followed by diffe rent letters are significantly different based on Tukey-Kramer test, P < 0.05. a Treatments were: two N. californicus released 4 DAT (2N+4), a second release 21 DAT (2N+4,21), or a third release 30 DAT (2N+4,21,30); four sulfur sprays initiated 16 DAT (4S+16); plants infested with P. latus and no pest management (Non-treated); plants with no P. latus (Non-infested).

PAGE 107

107 Table 3-5. Marketable fruit yield, P. latus -damaged fruit, and marketable fruit set in greenhouse-grown pepper plants in Spring-2005 (88 DAT). Plants were treated to manage a P. latus infestation initiated three days before transplanting. Marketable fruit yield Weight Number Treatment codea (kgm-2) (%) (no./m2) (%) Damaged fruit number (no./m2) Fruit set (%) N. californicus 2 N 4.3 ab 87 abc 22.6a 94ab 0.3 b 16.6a 2 N 0 3.4 b 66 c 18.3a 73b 2.7 ab 16.5a 2 N +4 1.4 c 25 d 7.1b 28c 7.4 a 15.6a 4 N 4.7 ab 95 ab 23.9a 102ab 0.0 b 15.2a 4 N 0 4.4 ab 88 abc 21.8a 92ab 0.5 b 14.8a 4 N +4 4.5 ab 91 abc 22.7a 96ab 0.3 b 15.9a Sulfur 5 S 4.4 ab 88 abc 21.9a 92ab 0.6 b 15.1a 5 S 0 4.3 ab 87 abc 22.1a 95ab 0.5 b 15.1a Non-treated 0.0 d 0 e 0.0b 0d 7.7 a 14.0b Non-infested 5.0 a 100 a 23.8a 100a 0.0 b 13.7a Significance F9, 18 P 39.1 0.0001 40.5 0.0001 28.1 0.0001 34.5 0.0001 22.6 0.0001 1.0 0.0442 Means within the same column followed by diffe rent letters are significantly different based on Tukey-Kramer test, P < 0.05. a Treatments were: one-time releases of N. californicus at a per plant predator density of either two (2N; 2N0; 2N+4) or four (4N; 4N0; 4N+4), with release time either at 0, or 4 DAT, respectively; five sulfur sprays initiated either DAT (5S) or 0 DAT (5S0); plants infested with P. latus and no pest management (Non-treated); plants with no P. latus (Non-infested).

PAGE 108

108 Table 3-6. Estimated pepper fruit gross return, pe st management cost, return relative to no pest and benefit-cost ratios for cr op scenarios in gree nhouse experiments Fall-2004 and Spring-2005. P. latus infestations on transplants were evaluated for pest control using N. californicus releases and sulfur sprays. P. latus management costc Treatment codea Gross returnb ($/m2) ($/m2) ($/plant) Return relative to Non-infestedd Benefit-cost ratioe ($/$) Fall-2004 2 N +4 10.75 c 0.090 0.030 0.46c 119b 2 N +4,21 16.21 bc 0.181 0.060 0.67cb 90b 2 N +4,21,30 15.29 bc 0.271 0.090 0.63c 56b 4 S +16 20.80 ab 0.048 0.016 0.87b 436a Non-treated 0.00 d 0.000 0.000 0.00d -Non-infested 23.65 a 0.000 0.000 1.00a -Significance F dfn, dfd P 35.15, 10 0.0001 ----86.85, 10 0.0001 28.73, 5 0.0005 Spring-2005 2 N 18.92 ab 0.083 0.028 0.83b 228 c 2 N 0 15.24 b 0.090 0.030 0.64b 169cd 2 N +4 6.02 c 0.090 0.030 0.24c 67e 4 N 20.88 ab 0.162 0.054 0.92ab 129de 4 N 0 19.36 ab 0.169 0.056 0.82b 115de 4 N +4 20.05 ab 0.169 0.056 0.87b 119de 5 S 19.39 ab 0.038 0.013 0.87b 511a 5 S 0 18.87 ab 0.049 0.016 0.86b 387b Non-treated 0.00 d 0.000 0.000 0.00d -Non-infested 22.07 a 0.000 0.000 1.00a -Significance Fdfn, dfd P 39.19, 18 0.0001 ----40.69, 18 0.0001 55.37, 13 0.0001 Means within the same column followed by diffe rent letters are significantly different based on Tukey-Kramer test, P < 0.05. aFall-2004: Two N. californicus released 4 DAT (2N+4), a second release 21 DAT (2N+4,21), or a third release 30 DAT (2N+4,21,30); four sulfur sprays initiated 16 DAT (4S+16); plants infested with P. latus and no pest management (Non-treated); plants with no P. latus (Non-infested). Spring-2005: One-time releases of N. californicus at a per plant predator density of either two (2N; 2N0; 2N+4) or four (4N; 4N0; 4N+4), with release time either at 0, or 4 DAT, respectively; five sulfur sprays initiated either DAT (5S) or 0 DAT (5S0). bMarketable fruit harvested multiplied by the mean w holesale price for greenhouse-grown red peppers at the Miami Terminal Market (GR = Yield Price) at the end of December (Price: $5.12/kg) and beginning of June (Price: $4.44/kg) (Jovicich et al., 2005). Yield in non-infested plants were 4.6 kgm-2 in Fall-2004 (106 DAT) and 5.0 kgm-2 in Spring-2005 (88 DAT). cCost inputs assuming the crop was grown in a 1-ha greenhouse with 30,000 plants. Calculation details are included in Tables B-3 and B-4.

PAGE 109

109 dCalculated as nopest pestGR PMC GR, where GRpest was the gross return with pest infestation, PMC is the pest management cost, and GRnopest is the gross return in scenar ios with no pest infestation. eCalculated as PMC GRpest. Unpredictable ratios result when denomina tor is zero in either scenario where plants are not treated for P. latus or plants are not infested by P. latus; these two treatments were excluded from the statistical analysis.

PAGE 110

110 Figure 3-6. Regression curves for variable s measured in bell pepper plants and P. latus cumulative mite-days (in a log.-scale) until 40 DAT in greenhouse experiments Fall-2004 and Spring-2005. Visual plant damage index (A). Percentages relative to plants not infested with P. latus (100%): marketable fruit yi eld (B), fruit set (C), and plant growth variables (DL). Data form treatments were pooled (n = 18 in 2004 and n = 30 in 2005). Equations and values for r2, F, and Pvalues are presented in Table 3-7.

PAGE 111

111 Table 3-7. Regression equa tions for plant visual damage inde x, fruit yield, fruit set, and plant growth variables, as affected by P. latus cumulative mite-days at 40 DAT on top canopy leaves of greenhouse-grown pepper plants in Fall-2004 and Spring-2005. Data from treatments were pooled. Estimated variable Equation r2 F df P P. latus -free plant (100% value) Fall-2004 (n = 18) Plant injury indexa = 6 (1 e.0012 x) 0.54 18.8 1, 17 0.0007 0 Plant measurements (%)b Marketable yield = 100 e.0010 x 0.81 68.6 1, 17 0.0001 4.6 kgm-2 Marketable fruit set = 100 0.018 x 2.36 10-0.005 x2 0.52 8.4 2, 17 0.0001 22 % Leaf area = 100 e.0007 x 0.72 46.9 1, 17 0.0001 15,914 cm2 Leaf fresh weight = 100 e.0006 x 0.76 49.4 1, 17 0.0001 645.8 g Leaf dry weight = 100 e.0004 x 0.61 25.4 1, 17 0.0001 79.1 g Leaf node number = 100 e.0006 x 0.60 24.4 1, 17 0.0002 100 Leaf number = 100 e.0005 x 0.60 23.6 1, 17 0.0002 317 Stem diameter = 100 e.0002 x 0.72 40.2 1, 17 0.0001 17.8 mm Plant height = 100 e.0007 x 0.78 55.9 1, 17 0.0001 176 cm Stem fresh weight = 100 e.0008 x 0.68 33.9 1, 17 0.0001 567.8 g Stem dry weight = 100 e.0009 x 0.82 72.1 1, 17 0.0001 82.3 g Spring-2005 (n = 30) Plant injury index = 6 (1 e.0009 x) 0.60 41.9 1, 29 0.0001 0 Plant measurements (%) Marketable yield = 100 e.0022 x 0.78 97.9 1, 29 0.0001 5.0 kgm-2 Marketable fruit set = 100 0.023 x 4.71 10-0.005 x2 0.66 25.8 2, 29 0.0001 14 % Leaf area = 100 e.0013 x 0.65 52.8 1, 29 0.0001 19,493 cm2 Leaf fresh weight = 100 e.0008 x 0.55 33.6 1, 29 0.0001 748.7 g Leaf dry weight = 100 e.0008 x 0.56 51.4 1, 29 0.0001 104.5 g Leaf node number = 100 e.0012 x 0.62 45.1 1, 29 0.0001 174 no. Leaf number = 100 e.0009 x 0.70 66.4 1, 29 0.0001 491 no. Stem diameter = 100 e.0002 x 0.59 49.6 1, 29 0.0001 18.3 mm Plant height = 100 e.0005 x 0.68 60.0 1, 29 0.0001 113 cm Stem fresh weight = 100 e.0011 x 0.41 19.7 1, 29 0.0002 512.3 g Stem dry weight = 100 e.0011 x 0.52 30.0 1, 29 0.0003 80.5 g aScale indices ranged from zero (no injury) to six (extreme severe injury). bEstimate percentage relative to plants not infested with P. latus.

PAGE 112

112 CHAPTER 4 GREENHOUSE-GROWN COLORE D PEPPERS: A PROFITAB LE ALTERNATIVE FOR VEGETABLE PRODUCTION IN FLORIDA?1 4.1 Introduction Outside the U.S., colored peppe r fruits are extensively produ ced in greenhouses (Costa and Heuvelink, 2000; Morgan and Lennard, 2000; Nu ez et al., 1996; Resh, 1996). Spain, The Netherlands, Canada, Israel, and Mexico have large greenhouse ar eas dedicated to the production of colored pepper fruits which are shipped to many countries throughout the world, including the U.S. (Table 4-1). Greenhouse production of peppers in the U.S. is minimal; we estimate that the area in 2002 was approximately 50 ha (123.5 acre s). In 1998, 165 greenhouse operations (14 ha) in the U.S. were dedicated to growing peppers with a total wholesale value of $4.8 million (U.S. Dept. of Agriculture, 1998). In the present re port we describe a greenhouse system for peppers, review historical prices, and i nvestigate the potential fo r production of colored bell peppers as a viable vegetable production alte rnative for Florida growers. 4.1.1 Greenhouses for Production of Vegetables in Florida In Florida, although vegetables are mostly produced in the open field, a few growers have been using greenhouses to produce high value ve getable crops [such as colored peppers, Beit Alpha and Dutch types of cucumbers ( Cucumis sativus ), and beefsteak and cluster types of tomatoes ( Lycopersicum esculentum )] that are difficult to produc e outdoors (Tyson et al., 2001). The area of greenhouses dedica ted to vegetable production in Florida was nearly 31.8 ha (78.58 acres) in the year 2000 (Tyson et al., 2001). 1 This chapter is reproduction of publication: Jovi cich, E., VanSickle, J.J., Cantliffe, D.J., and Stoffella, P.J., 2005. Greenhouse-grown colored peppe rs: a profitable altern ative to vegetable production in Florida? HortTechnology 15, 355.

PAGE 113

113 4.1.2 Pepper Imports to the U.S. Consumption and imports of fresh-market bell peppers to the U.S. have been increasing over the past 10 years (U.S. Dept. of Agricult ure, 2001). From the years 1992 to 2002, the total domestic use of bell peppers increased from 657,710 t (724,989 tons) to 903,747 t (996,194 tons) (U.S. Dept. of Agriculture, 2003a) and the annual per capita fru it consumption increased from 2.6 kg (5.73 lb) to 3.2 kg (7.05 lb), respectively (U.S. Dept. of Agriculture, 2003b). The U.S. Dept. of Agriculture (2001) reported that 24 % of Americans consume at least one food containing bell peppers daily, a nd that 10% of these consumers use fresh bell peppers. In the U.S., colored bell pepper fruits are c onsidered specialty commodities (U.S. Dept. of Agriculture, 2001). For a growing segment of pe pper consumers, the demand for colored fruits has increased despite having two or more times hi gher prices than mature green fruits at retail supermarkets (Frank et al., 2001; U.S. Dept. of Agriculture, 2001). C ountries that produce colored peppers have taken advantage of this market opportunity and have been exporting colored peppers to the U.S. (Cantliffe and VanS ickle, 2001; Fintrac, 2003; Lopez and Shwedel, 2001; U.S. Dept. of Agriculture, 2001) (Figure 1). Countries that sh ipped only high quality colored fruits attracted high annual average pr ice values per unit weight of pepper [>$1.50/kg ($0.68/lb)] (Figure 1). In 2002, 26.9% of the volume of the bell pe ppers consumed in the U.S. was imported for a value of $290.6 m illion (40.6% of the total value of bell peppers sold in the U.S.), and 98.9% of this value co rresponded to shipments from Mexi co (green and colored fruits shipped throughout the year, mostly in the peri od NovemberMay), Canada (green and colored fruits shipped in the period MarchJanuary), Th e Netherlands (colored fruits shipped in the period FebruaryDecember), Israel (colored fruits shipped OctoberApril), and Spain (colored fruits shipped in the period Oct oberApril) (Table 4-2) (U.S. Dept. of Agriculture, 2002, 2003a, 2003c). Sales from Canada, Israel, and Spain ha ve increased in the recent years, with more

PAGE 114

114 colored fruits being shipped. Greenhouse area dedicated to peppers has been increasing in Mexico and their sales of high quality colored pe ppers in the U.S. have increased (Cantliffe and VanSickle, 2003). The fluctuations of sales in the U.S. from countries shipping bell peppers (U.S. Dept. of Agriculture, 2003a) are shown in Figure 2. 4.1.3 Field Production of Bell Pepper in Florida Florida is the main U.S. winter supplier of bell pepper fruits to the northern and mid-western states because it benefits from a mild winter climate in the ce ntral and southern part of the peninsula (Florida Agri culture Statistics Service, 2003 ). About 7163.1 ha (17,700 acres) of bell peppers were harvested in the peri od Aug. 2002 to July 2003 in Florida (Florida Agriculture Statistics Service, 2003). Pepper crops are grown on polyethylene-mulched beds, fumigated with methyl bromide, and irrigate d through subsurface or drip irrigation systems (Maynard and Olson, 2003). Fruits are harvested weekly during a period of approximately one month and almost exclusively picked at the matu re green stage. Although field growers receive a premium price for mature ripened peppers, th e production of red, orange or yellow peppers represents a higher risk in harv esting satisfactory quality and ad equate yield as compared with premature harvests of green peppers. In pepper pl ants, development of full color in fruits is completed 2 to 3 weeks after they reach the mature green stage (R ylski, 1986). The length of the period of full color development (green to fu ll color) depends on environmental conditions. During the ripening period, pepper fruits in the fi eld are often exposed to adverse environmental factors such as rainfall, extreme air temperat ures and solar radiati on, and insect pests and diseases, which significantly reduce fruit qualit y and marketable yield. High solar radiation, temperature and humidity characterize Florida s weather during the la te spring and summer (Bureau of Economic and Business Research, 2 000; Winsberg, 1990). In central and north Florida, and occasionally in some southern areas of the peninsula, night temperatures can fall

PAGE 115

115 below 0 C (32.0 F) during the late fall and wint er. Therefore, extending the crop period to harvest full-color fruits is generally not practic al in open-field pepper pr oduction, although in the warmer coastal areas of southern Florida, some growers harvest full-colo r pepper fruits from a portion of their crop. This practi ce can increase income but also in creases the risk of decreased fruit yield and quality (VanSickle, 2003). 4.1.4 Research and Greenhouse Production of Bell Pepper in Florida In the past 5 years, the Protected Agricu lture Project, Horticultural Sciences Dept., University of Florida, Gainesville (lat. 29N, long. 82W) evaluated the production of bell pepper as an alternative crop for local growers and reported promising fruit yields and high fruit quality during extended growing seasons in northcentral Florida using technologies adapted for regions with subtropical and mild winter climates (Cantliffe D.J., 1999, Horticultural Sciences Protected Agriculture Center, Uni v. of Florida, Gainesville, Fla. ; Jovicich, 2001; Shaw and Ca ntliffe, 2002). Fruit yields were higher than field-grown bell peppers, comp arable or higher than average yields of greenhouse-grown peppers in Spain and Israel, but lower than greenhouse-grown peppers in Canada and The Netherlands (Table 4-3). In 1998, there were eight greenhouse operations in Florida growing bell peppers for a total of 9.51 ha (23.5 acres) (U.S. Dept. of Agriculture, 1998). Fruit yields per unit area in Flor ida greenhouses were twice as high as field-grown pepper yields in 1998; moreover, because of the type and qualit y of the commodity, colored fruits from the greenhouses had sales values 11 times greater than that of field-grown peppers (Table 4-4). In the year 2001, bell pepper had the largest ar ea in production among vegetables produced in greenhouses in Florida [pepper, tomato, cucumber, and lettuce ( Lactuca sativa )] (Tyson et al., 2001). In 2003, the area with greenhouse-grown be ll peppers in Florida was estimated to be 14

PAGE 116

116 ha (34.6 acres), consisting of half a dozen grower s with greenhouse areas that ranged from <1 to 8 ha (2.5 to 19.8 acres), and located in central and southern Florida. Most of the technology and management prac tices applied to gree nhouse pepper crops in Florida have been introduced from Canada and The Netherlands. This includes management of nutrients delivered through drip irrigation and plant pruning system s to two stems, with plants stems trained vertically onto hanging twines (Portree, 1996; Resh, 1996). However, pepper greenhouse production in Florida can benefit fr om greenhouse technologies generated in countries with comparable mild winter climates. Compared to the highly intensive systems used in northern latitudes, some of th e low-cost technologies used in mild winter climates (Costa and Heuvelink, 2000; Martnez, 1999; Nu ez et al., 1996) perform well in Florida (Cantliffe, D.J, 1999, Horticultural Sciences Protected Agriculture Center, Univ. of Florida, Gainesville, Fla. ; Jovicich, 2001; Jo vicich et al., 2004). High-roof, passively ventilated greenhouses represent a functional and low-cost structure for subtropical and tropical regions (Wittwer and Castilla, 1995). Shading and large screened openings in these high structures [about 5 m (16. 4 ft) distance from th e floor to the gutter] facilitate ventilation and lowering of air temper atures during periods of warm weather (Monteiro, 1994; Wittwer and Castilla, 1995). In central and s outhern Florida, little fuel is needed for warming the crop during winter. Greenhouse pepper crops in Florida are not grown in soil in order to avoid nematode and soil-born diseases. In the greenhouse, as an alternative to soil and to the use of chemical disinfection, pepper plants are grown in containers with soilless media, which can be reused for several crops as long as no disease contamina tion of the substrate occurs. The plants are irrigate d through drip irrigation, which a llows controlling the amounts of nutrients and water.

PAGE 117

117 4.1.5 Returns of Greenhouse-Grown Peppers Greenhouse area is predicted to e xpand in the near future, in pa rt as a consequence of the greater demand for specialty vege table crops, loss of methyl brom ide, and an increase in urban sprawl and price of arable land (Cantliffe et al ., 1999). In Florida, pr oduction costs and returns have been estimated for tomatoes grown in small greenhouses [with a floor area of around 278.7 m2 (3000 ft2)] (Hochmuth and Belibasis, 1991; Sm ith, et al., 2003). However, no information about potential profitability has been published for peppers grown in Floridas greenhouses. The objective of this report is to estimate the potenti al profitability of a greenhouse enterprise producing mature-ripened bell peppers in Florida. The analyses have the purpose of providing information that will assist growers and ag ricultural investors in evaluating greenhouse pepper production as a potential business opportunity in Florida. 4.2 Methods The origin of the data used in this economi c analysis was from both experiments at the Protected Agriculture Center (Can tliffe, D.J., 1999, Horticultural Sciences Protected Agriculture Center, Univ. of Florida, Gainesville, Fla. ) and from commercial growers. A typical bell pepper production cycle wa s assumed for a greenhouse area of 0.78 ha, approximately the smallest size existing in the state. 4.2.1 Greenhouse Structure Two identical greenhouse units with a total structure floor area of 7840 m2 (1.94 acre) and with a crop growing area of 6,656 m2 (1.64 acre) were assumed in the enterprise budget analysis (Table 4-5). The multiple-bay high-roof greenhous e structures (commercially available and used by local growers) were covered with polyethylene and had retrac table side walls and saw-tooth roofs with a roof-vent on ever y bay (Figure 3). All openings were screened with a 50-mesh insect-proof screen. Since the greenhouses we re located in north-central Florida, minimal

PAGE 118

118 heating (restricted to nights a nd early mornings) was necessary fr om the end of November to the beginning of March in order to have a good plant growth with high quality fruits. Occasional nighttime heating was needed during April and May. Diesel-fueled heaters with convection tubes laid on the greenhouse floor were used in conjunction with aluminized thermal screens to successfully keep minimum temperatures above 13 C (55.4 F) with a low use of fuel. Sidewall openings, fans, and aluminized screens were us ed to reduce temperatures during the beginning and end of the growing season. During August and from March to May, air could be blown through the same heaters convection tubes to impr ove the ventilation from that of the natural passive ventilation provided by the greenhouse structures. 4.2.2 Crop Cycle The pepper crop used as the basis of this budge t analysis lasted 298 d from seeding to the removal of the crop. Pepper seedlin gs were grown in plug trays for 35 d and then transplanted at the beginning of August. Vertical trellising of the plants, scouting plants (for pests, diseases, beneficial arthropods, and bumble bee activity), and monitoring the volumes of water and levels of nutrients supplied with irriga tion were the main labor activiti es until the end of October. Harvesting of ripened fruits started at the end of October and continued during 28 weeks (usually one harvest per week, with a total of 30 harvests), until the end of May. 4.2.3 Plant Production System Plants were trellised to the S panish system. In this trelli s system, plants are supported vertically by twines ex tended along the rows and on both sides of the canopies [up to six levels of twines in 1.8-m-tall (5.91 ft) plants] and by ve rtical poles (Figure 4). With the Spanish trellis system, minimal pruning is done to the plant canopy. Th erefore, the amount of labor needed is about 25% of the labor needed with the V trellis system, another system where plants are regularly pruned to form two stems and where each stem is vertically supported with

PAGE 119

119 individual strings that are w ound around the stems (Jovicich, 2001). In Florida, fruit yield and quality have been similar in bot h trellis systems (Jovicich and Cantliffe, 2001; Jovicich et al., 2004). Plants were grown in 11.4-L (3 gal) nurs ery pots containing perlite and drip-irrigated with a solution of water and nutrients (Hochm uth, 1997; Jovicich, 2001). The frequency and duration of the irrigation events were controlle d by an automated system. The volumes of irrigation solution and levels of nutrients were changed with plan t growth, air temperature, and solar radiation levels. Each plant received a total 240 L (63.4 gal) of water and 480 g (16.9 oz) of fertilizers that include d macro and micronutrients. 4.2.4 Pollinators, Pests, and Diseases Bumblebees ( Bombus impatiens ) were required to aid natural pollination, and thus to improve fruit set and fruit quality. The crop wa s scouted weekly for pest and diseases, and beneficial insects, and to check on the activity of the bumblebees. Pest control combined the use of biological and chemical control measures. Biol ogical control of insect pests included the use of Bacillus thuringiensis [for control of fungus gnats ( Bradysia spp.)], and releases of parasitoid Aphidius colemani [for control of cotton aphid ( Aphis gossypii ) and green peach aphid ( Myzus persicae )] and of the predatory mite Neoseiulus californicus [for control of two-spotted spider mite ( Tetranychus urticae ) and broad mite ( Polyphagotarsonemus latus )]. Miticides for broad mite control were used only if severe infestations occurred and applications were restricted to the affected areas of the crop. Other pests th at may have been present [e.g., pepper weevil ( Anthonomus eugenii ), silverleaf white y ( Bemisia argentifolii ), and western flower thrips ( Frankliniella occidentalis )] would have required other specif ic predators and/or pesticides. Starting in January, preventive fungicide applicat ions were used to control powdery mildew ( Leveillula taurica ).

PAGE 120

120 4.2.5 Fruit Yields Fruit yields of bell pepper hybr ids, with red, orange, or yell ow color were estimated from those obtained in experimental crops in north-central Florida [7 kgm-2 (1.4.9 lb/ft2)] (Jovicich, 2001; Shaw and Cantliffe, 2001; U.S. De pt. of Agriculture, 1998) and in commercial crops in central and southern Florida, where growers currently obtain a similar yield range (Table 4-3). After harvesting, marketable fruits were graded by size, labeled and packed into 5-kg (11.0 lb) cartons, and then stored in a refrig erated room at 7 C (44.6 F) and 95% relative humidity (Jovicich et al., 2003). Cull fruits, su ch as those with blossom-end rot, cracking, or those unshaped, were detached from the plants during the harvest operations. 4.2.6 Pepper Fruit Prices Historical market prices for mature ripened pe ppers were estimated fr om market prices of imported peppers supplied by The Netherlands, Spai n, and Israel, which ha ve been shipping only greenhouse-grown bell pepper fruits to the U.S. Wholesale price values at the Miami terminal market were obtained from U.S. Department of Agriculture data, available in the Wholesale Market Vegetable Report from the Market Info rmation Database System (MIDS) of the Food and Resource Economics Dept., Univ. of Florida. Monthly average bell pepper wholesale prices (unadjusted for inflation) were calculated from maximum and minimum price values in transactions carried out on Thursdays from Dec. 1993 to Jan. 2002. Means and standard deviations were calculated for selected price seri es and are reported as mean price value SD. Prices of fruits were differentiated by fruit orig in, size, and color, and by weight of the packed carton. Prices are presented in units of dollars pe r kilogram, or the equivalent of dollars per 5-kg carton. Monthly average prices were estimated for mature green bell peppers, combining imports (mostly from Mexico) and U.S. production. Aver age prices for mature-ripened fruits produced

PAGE 121

121 in the field were calculated from shipment data for major bell pepper production states (Florida, California, North Carolina, and Ge orgia). Countries such as Th e Netherlands, Spain, and Israel export to the U.S. only mature-ripened bell pepp er fruits from greenhous e-grown crops; thus, average prices for fruits arriving from these th ree countries combined were used to estimate prices for fruits produced in greenhouses. Fiel d peppers are primarily packaged in 12.7-kg (28 lb) cartons (1-1/9 bu), while greenhousegrown fruits are packaged in 5-kg cartons. 4.2.7 Budget Analysis An enterprise budget that consisted of gross re venue, costs, and profit associated with the production of bell peppers in a greenhouse wa s used to estimate annual profitability. Methodology described by Kay and Edwards (19 94) for a budget analysis was adapted for greenhouse production. The budget tabl es included items, quantities, units, and prices used throughout a typical production cycl e in north-centr al Florida. Gross revenue was estimated by multiplying the wholesale market price by the marketable fruit yield. Average monthly prices of gr eenhouse-grown pepper fruits were multiplied by monthly fruit yields during the NovemberMay harvest period and summed to obtain the total gross revenue expressed in dollars per unit area of total greenhouse floor area (Table 4-6). The total fruit yield was estimated to be 13 kgm-2 based on the technology and practices used, and on the length of the crop season. Fruit yield was expr essed in weight units per unit of usable area (85% of the total greenhouse floor area), whereas costs and revenues were based on a unit area of the total greenhouse area (0.78 ha). The formula used to calculate gross revenue was: gross revenue ($/unit or total area) = yield (weight per unit or total gr owing area) 0.85 market price ($/unit weight). Total costs were divided into fixed and va riable costs. Total fixed costs included depreciation of durables and expenses that woul d still exist even when no crop was grown in the

PAGE 122

122 greenhouse. A summary of the items considered in the total investment, with their original costs, expected life, and depreciation is presented in Table 4-7. The costs of the greenhouse structure, irrigation equipment, and all other material i nputs required for the produ ction of the crop were calculated using estimates from manufacturing companies. Annual depr eciation of long-term items was based on a zero salvage value and an es timated useful life in y ears (Table 4-7). Land price was not included in the investment. Other fixed costs included were those incurred from repairs and maintenance, taxes and licenses, insurances, telephone, and office expenses (Table 4-8). Variable or operating costs were those that would be incurred only if the crop was grown. Variable costs were divided into preharvest, harvest, packing and marketing costs, and sales transaction costs. Quantity and prices of i nputs were obtained by contacting input suppliers. The list of activities a nd inputs throughout the crop production cycle are summarized in Table 4-9. Records for types and amounts of labor need ed in a pepper production cycle were estimated from experimental and commercial crops. Employee wages ranged from $6.50 (basic) to $15.00 (specialized) per hour depending on the expertise n eeded for a specific task. An additional cost of 20% of the wage value was estimated for cont ributions to social security and insurance, unemployment tax, and other employee expenses. Th e sales transaction cost was estimated to be 15% of the gross market price. Total cost per unit area was the sum of variable costs and fixed costs. The price received by the grower was assumed to be the average wholesale market price less the sales transaction cost. Gross revenue was the product of th e price received by the grower and the total production. Profit was calculated by subtracting total expenses from gross revenue. The opportunity costs (interest on operating cap ital and on investment capital), land charge

PAGE 123

123 (rent), and cost of managing th e enterprise (managers salary) were not included as specific expenses in this budget, and, thus, were part of the residual net return or loss. Because the opportunity cost for management, operating capital, and investment capital were not incorporated in the enterprise budget, the estimated profit should be interpreted as estimated returns to capital and management. The formula used to calculate profit (in $/unit area or $/total area) was: returns to management and capital ($/unit or total area) = gross revenue ($/unit or total area) total costs ($/unit or total area). The pr ofitability of the 0.78-ha enterprise (excluding the land) was estimated by the relation between the valu e of return and the investment. The internal rate of the return to invested capital and ma nagement from the total investment (IRR) was calculated with the following formula: IRR (%) = returns to capital and management ($/total area) 100 total investment ($/total area). The estimated IRR was compared to the opportunity cost of capital using th e average interest rate for bank loans to business for the year period 1993. From the budget analysis, gross revenues, pr oduction costs, and retu rns were plotted and linear equations calculated for a range of marketable fruit yields (5 kgm-2), assuming the average fruit price for the N ovember May period ($5.29/kg). 4.2.8 Sensitivity Analysis Sensitivity analysis was used to analyze how changes in budget assumptions affect income and cost projections. Returns to management we re calculated for a range of marketable fruit yields (5 kgm-2) and market prices [$15.00.50/5-kg carton ($1.36.95/lb)]. In the sensitivity analysis, market prices were assumed to be average values for the NovemberMay harvest period.

PAGE 124

124 4.2.9 Break-even Analysis The data in the enterprise budget were used to perform a break-even analysis. This analysis was used to calculate various combina tions of possible yield an d prices required to cover anticipated total production co sts. Break-even prices were calculated using the formula: break-even price ($ per unit weight ) = anticipated total costs ($ pe r unit area) anticipated yield (weight per unit area). 4.3 Results 4.3.1 Prices for Mature Ripened Bell Pepper Fruits Historical wholesale market prices at the Miami terminal market indicate that mature ripened greenhouse-grown bell pepper fruits attracted average pric es three to five times higher than mature ripened or green fruits grown in the field, respectively (Fig ure 5). Red and yellow fruits produced in the field had average annual fr uit prices (hereafter, mean price SD) of $1.51 0.41/kg ($0.68 0.19/lb) and $1.69 0.53/kg ($0.77 0.24 /lb), respectively, with peak prices close to $2.00/kg ($0.91/lb) in Jul y, August, and November (Figure 5). The average price for mature green fruits was only $0.91 0.24/kg ($0.41 0.11/lb) and with minimal price fluctuations throughout the year Imported red, yellow, and orange fruits produced in greenhouses averaged $4.68 0.71/kg, $4.52 0.66/kg, and $5.21 0.76/kg ($2.12 0.32/lb, $2.05 0.30/lb, and $2.36 0.34/lb), respectively (F igure 5). For greenhouse-grown fruits, average prices for all fruit colors combined attracted high values during the period November May [$5.29 0.78/kg ($2.40 0.35/lb)] while peak prices were in January [$5.63 0.50/kg ($2.55 0.23/lb)] and April [$6.00 0.82/kg ([ $2.72 0.37/lb)]. For the whole year, the average price for colored fruits produced in greenhouses was $4.80 0.71/kg ($2.18 0.32/ lb). The annual average price per unit weight for ma ture ripened bell pepper fruits produced in

PAGE 125

125 greenhouses was an additional $3.20/ kg ($1.45/lb) compared to mature ripened fruits produced in the field, and an additional $3.89/kg ($1.76 /lb) compared to mature green fruits. 4.3.2 Investment and Costs The total investment for a greenhouse enterpri se with a protected area of 0.78 ha totaled $820,490 or $104.67/m2 ($9.72/ft2) (Table 4-7). The cost of th e greenhouse structure and cover materials combined was $22.27/m2 ($2.07/ft2) and, with construction labor, supervision, and concrete needed for erection, the cost increased to $43.45/m2 ($4.04/ft2) (Table 4-7). The investment on heating and ventilat ion systems was estimated at $9.13/m2 ($0.85/ft2), and on systems and controls for irrigation, ferti lization, drainage, and ventilation was $16.61/m2 ($1.54/ft2). The annual depreciation cost for the total investment was $10.18/m2 ($0.95/ft2). Additional expenses within fixed costs represented $2.56/m2 ($0.24/ft2) (Table 4-8). Total fixed costs added to $12.74/m2 ($1.18/ft2). Preharvest costs were estimated at $11.50/m2 ($1.07/ft2), harvest costs at $2.12/m2 ($0.20/ft2), and packing and marketing at $5.90/m2 ($0.55/ft2) (Table 4-9). These variable costs combined added to $19.52/m2 ($1.81/ft2). The total labor needed to produce, harvest and pack the bell pepper fruits was 5,744 hours [0.73 h/m2 (0.068 h/ft2)] and represented $45,409 for the 0.78 ha [$5.79/m2 ($0.54/ft2)] (Table 4-9). From this total labor, 35.7% of the time was used to grow the crop, 39.2% for harvesting, and the remain ing 25.1% for packing the fruits (Table 4-9). The total cost of diesel fuel used for heating was estimated at $35,284 [$4.50/m2 ($0.42/ft2)]. 4.3.3 Budget Analysis The summary for the budget analysis (Table 4-10 ) included fixed (Table 4-8) and variable costs (Table 4-9). Adding the sales transaction expenses [15% of the gross revenue, equivalent to $8.85/m2 ($0.82/ft2)] to the variable costs of production [$19.52/m2 ($1.81/ft2)], the total variable cost was $28.37/m2 ($2.64/ ft2) (Table 4-10). Total costs (fixed and variable) amounted

PAGE 126

126 to $41.09/m2 ($3.82/ft2) ($322,148 for the 0.78 ha). From this total cost, depreciation cost represented 24.7%, transaction cost 21.5%, cost of total labor 14.1% and cost of fuel 11.0% of the total cost (Table 4-10). With a marketable fruit yield of 13 kgm-2 harvested during the period NovemberMay, the estimated gross revenue was $58.98/m2 ($5.48/ft2) ($462,392 for the 0.78 ha) (Table 4-6). The return to capital and to management was $140,244 for the total greenhouse area, or $17.89/m2 ($1.66/ft2) (Table 4-10). The IRR to invested capital and management (excluding the land) was estimated at 17.1%. Gross revenues, production costs, and returns we re calculated for a range of marketable fruit yields (5 kgm-2), assuming the average price for the period NovemberMay ($5.29/kg) (Figure 6). Total costs increased with increased fruit yields because of the need for additional labor for harvesting and packing, and of more pack ing supplies. The estimated linear responses indicate that returns to capita l and management were zero when a marketable fruit yield of 7.8 kgm-2 was obtained in the growing area. Unde r the assumptions of this budget analysis, break-even yield was similar to th e average fruit yield of 7.9 kgm-2 (1.62 lb/ft2) that was estimated from data collected by the 1998 Census of Specialty Crops (U.S. Dept. of Agriculture, 1998) (Table 4-4). Currently, fruit yields of 10 to 15 kgm-2 (2.0 to 3.1 lb/ft2) are realistic in fall to spring crops such as the one used in this analysis. A marketable fruit yield of 17 kgm-2 (3.5 lb/ft2), which is considered a high yield in st ructures with minimal climate control in subtropical regions, led to an estimated return of $30.84/m2 ($2.87/ft2) (Figure 6). 4.3.4 Sensitivity Analysis and Break-Even Analysis A sensitivity analysis was used to examine how returns to capital and management changed under favorable and unfa vorable situations of fruit pr oduction and market price (Table 4-11). Changes in returns from negative to positive values were calculated under various

PAGE 127

127 combinations of marketable yields (5 kgm-2) and fruit prices [$15.00.50 per 5-kg carton, as average values for the NovemberMay harv est season]. Costs of production (with sales transaction expenses dedu cted) ranged from $28.61/m2 to $34.96/m2 ($2.66/ft2 to $3.25/ft2) based on a range of possible marketable fruit yields of 5 to 17 kgm-2, respectively (Table 4-11). Returns for the 0.78 ha are included in Table 4-11. Break-even prices (in $/kg and $/carton) required to cover anticipated total production costs were calculated for various possible fruit yields (T able 4-12). For fruit yields ranging from 5 to 17 kgm-2, the break-even prices ranged fr om $39.73 to $13.90/5-kg carton ($3.60 to $1.26/lb). 4.4 Discussion Much of the U.S. demand for high quality colo red pepper is currently supplied by imports. The value and yield of colored fruits produced pe r unit area in greenhouses can be three or five times higher compared to colored and green pepp er fruits grown in the field, respectively. Imported greenhouse-grown peppers (as with tomatoes and cucumbers) compete with field-grown crops in the U.S. (Cantliffe and VanSickle, 2001). Moreover, in Florida, high quality colored pepper fruits produced in greenho uses compete with imports from Mexico, The Netherlands, Canada, Israel, and most recently, Spain. However, the high and steady market prices attracted by colored bell pe pper fruits in the past years, increases in consumer demand, and suitable environment for growing colored peppers under protected ag riculture in Florida have led to a small but expanding production area a nd have driven more growers to consider the economic viability of this crop. Green and colored peppers produced in the fiel d and greenhouse, respectively, are different types of commodities, which have different produc tion costs and attract different market prices. In our analysis, for the fruit yield of 13 kgm-2, the estimated total cost of producing colored

PAGE 128

128 peppers in the greenhouse was $41.09/m2 and the return to management and capital was $17.89/m2. Using estimates by Smith and Taylor ( 2004) as a reference, a field-grown pepper crop in Florida with a fruit yield of 31.4 tha-1 [14 tons/acre (1000 bu/acre at 28 lb/bu)] incurred a production cost of $2.53/m2 ($0.24/ft2) and a return to management and capital of $0.61/m2 ($0.06/ft2) based on a market price of $1.00/kg ($0.45/lb). Under the greenhouse crop management practices and market prices used in this study, fruit yields should be greater than 7.8 kgm-2 in order to generate positive returns to management. This break-even yield is a realistic goal in Flor ida because greater yields are currently obtained in greenhouse enterprises similar to the one consid ered in this analysis. In experimental and commercial crops, yields of 10 to 15 kgm-2 were possible when produc tion practices and length of the growing season were similar to the ones de scribed here. The estimation of probabilities for alternative yield outcomes will be completed in an upcoming research project. These estimates will make additional interpretati on of the sensitivity analyses possible. An investment would be profitable when the inte rnal rate of return for the investment is greater than the opportunity cost of capital (Kay and Edwards, 1994). The IRR of the 0.78-ha greenhouse enterprise was greater th an the average interest rate for bank loans to business for the past 10 years (7.3% as an average in the period 1993) (Federal Reserve, 2004). The estimated IRR was within the range of rates of return (8.5% to 19.4%) calculated by Hodges et al. (2001) for large greenhouse nurse ries that produced ornamental s in Florida in 1998. A cash flow analysis for a series of years, which woul d include interest on the capital investment, could supplement the budget analysis. Greenhouse enterprises, even when located in a small region, are variable in size, composition, and management. As such, grower s seeking to produce peppers in greenhouses

PAGE 129

129 should use this information only as a guide and ca lculate the budgets for their own enterprises. The initial capital investment will differ among greenhouse enterprise s; however, per unit area, it is always considerably high compared to inve stments in field vegetable production. For this budget analysis, we considered the states smalle st greenhouse area with bell pepper. It is expected that larger greenhouse enterprises will benefit from ec onomies of scale and will incur relatively lower costs of production per unit area. NaLampang et al. (2003), in their calculati ons of demand elastic ities for green bell peppers for four regional markets in the U.S., fo und an inelastic demand for green peppers, with elasticities ranging from -0.10 to -0.24. No compar able studies have been carried out for colored bell peppers. However, it is ex pected that the demand elasticity for colored bell peppers would be more elastic, indicating more room for re lative production growth as efficiencies are identified in the production and mark eting of colored bell peppers. Vegetable growers in Florida currently face production challenges in terms of marketing (demand for high quality products with year-r ound supply and foreign competition), land availability (increased urbani zation and development in warm weather agricultural areas), labor (reduced availability with the appearance of higher-paying j obs, such as in the construction business), water restrictions (stricter regulati ons to protect water qua lity and to minimize amounts used), and the upcoming banning of me thyl bromide (Montreal Protocol on ozone protection resolutions). For some growers, th e production of high value crops such as colored bell peppers in greenhouses usi ng soilless growing systems ma y represent an alternative production system that would overcome some of th ese critical challenges and would assist their continuation in the agricultural business.

PAGE 130

130 4.5 Summary The increase in U.S. demand for colored bell peppers ( Capsicum annuum ) has been satisfied with increased supplies from im ports and increased domestic production. Greenhouse-grown peppers of red, orange, and ye llow colors were imported during the period 1993 at wholesale fruit market pri ces that were three to five times greater than field-grown fruits. With high market prices and a suitabl e environment for growing colored peppers under inexpensive greenhouse structures [<$40/m2 ($3.7/ft2)], up to 14 ha (34.6 acres) of greenhouses produced bell peppers in Florida in the year 2002. To estimate the profitability of a bell pepper greenhouse enterprise, a budget analysis was used to calculate the re turns to capital and management. Production costs of greenhouse-grow n peppers were estimated assuming the use of current technology applied in commercial greenhouse crops in Florida and in experimental crops at the University of Florida. Producti on assumptions included a crop of nonpruned plants grown in soilless media in a high-roof polye thylene-covered greenhous e [0.78 ha (1.927 acres)] located in north-central Florid a. For a fruit yield of 13 kgm-2 (2.7 lb/ft2), the total cost of production was $41.09/m2 ($3.82/ft2), the estimated return was $17.89/m2 ($1.66/ft2), and the return over investment was 17.1%. A sensitivity an alysis indicated that fruit yields should be greater than 7.8 kgm-2 (1.60 lb/ft2) in order to generate positive returns based on a season average wholesale fruit price of $5.29/kg ($2.40/lb). For this price, a range of possible fruit yields [5 to 17 kgm-2 (1.0 to 3.5 lb/ft2)] led to returns ranging from $9.52 to 30.84/m2 ($0.88 to 2.87/ft2), respectively. The estimates indicate that production of greenhouse-grown peppers could represent a viable vegetable produc tion alternative for Florida growers.

PAGE 131

131 Table 4-1. Area with greenhouse-grown bell peppers in selected countries that export colored fruits to the U.S. and area with gr eenhouse-grown peppers in the U.S. Countr y Area ( ha ) aSource Mexico 165J.Z. Castellanos, personal communication Canada 144Agriculture and Agri-food Canada, 2003 Netherlands 1,200 Valero-Garca, 2003 Israelb 535P. Imas, personal communication Spain 10,000Snchez et. al, 2000; Valero-Garca, 2003 U.S.c 14U.S. Dept. of Agriculture, 1998 a1.0 ha = 2.47 acres. bIncludes high tunnels. cAn estimate for 2002 would be near 50 ha.

PAGE 132

132 Figure 4-1. Total value, volume, and unit value of bell pepper fruits (all colors) imported into the U.S. from major shipping countries throughout the ye ars 1996 (U.S. Dept. of Agriculture, 2004); 1.0 t = 1.10 tons; $1.00/kg = $0.454/lb.

PAGE 133

133 Table 4-2. Volumes, values, and origins of bell peppers imported into the U.S., and volumes and values of the production, exports, and use of bell peppers in the U.S. in the year 2002. Country Volume Value Unit value ( t ) a ( % ) ( $1 000 ) ( % ) ( $/k g) b Mexicoc 164,72467.8132,72745.7 0.81 Canadad 41,414 17.171,41724.6 1.72 Netherlandse 23,852 9.856,84419.6 2.38 Israele 6,563 2.715,6385.4 2.38 Spaine 3,694 1.510,1613.5 2.75 Dominican Republicc 1,787 0.71,7530.6 0.98 Belgiume 267 0.16230.2 2.33 Others 577 0.21,4270.5 2.47 Total imports 242,876 100.0290,589100.0 1.20 U.S. production 734,118---498,650--0.68 U.S. exports 73,247---73,421----U.S. domestic use 903,747---715,818----a1.0 t = 1.10 ton = 2,204.6 lb. Volume and price data extracted from the U.S. Dept. of Agriculture (2002; 2003a; 2003c). b$1.00/kg = $0.454/lb. cMostly field-grown green fruits and some colored. dField-grown green fruits and greenhouse-grown colored fruits. eGreenhouse-grown colored fruits. Figure 4-2. Total sales value of bell peppers im ported to the U.S. from selected countries throughout the years 1995 (U.S. De pt. of Agriculture, 2003a).

PAGE 134

134 Table 4-3. Yields of bell pepper fruits produced in greenhouses, a nd in field crops in Florida. Marketable yielda (tha-1) (kgm-2) (kg/plant) Source Greenhouse crops The Netherlandsb 260 26 8.7 Costa and Heuvelink, 2000 Canadab 220260 226 7.38.7 Mirza, 2003 Israelc 7040 7 2.34.7 P. Imas, personal communication Spain (Almera)d 600 6 2.02.3 Costa and Heuvelink, 2000 U.S. (Florida)e 70 7 2.3.0 Jovicich, 200 1; Shaw and Cantliffe, 2002 Field crops U.S. (Florida)f 28.5 2.9 0.66 Maynard and Olson, 2003 aApproximate plant density in greenhouses is 30,000 pl ants/ha (12,141.0 plants/acre) while in FL field crops is 43,241.9 plants/ha (17,500 plants/acre). 1.0 tha-1 = 0.45 ton/acre. 1.0 kgm-2 = 0.21 lb/ft2. 1.0 kg/plant = 2.20 lb/plant. bIn The Netherlands, plants grown in soilless media, in glasshouses with climate control, including injection of carbon dioxide. Similar pr oduction systems are used in Canada. cIn Israel, 7 kgm-2 (1.4.0 lb/ft2) are yields in high plastic tunnels and unheated greenhouses. In heated greenhouses average yields are 14 kgm-2 (2.9 lb/ft2). dIn Spain (Almera), plants grown in sand, in lo w-roof low-cost greenhouse structures with no climate control. eMarketable fruit yields information obtained from grow ers and University of Florida experimental crops of colored peppers over the past 5 years. Plants grown in soil-less media and mostly in passively ventilated high-roof greenhouse structures covered with polyethylene, some of them with supplemental fuel-heating during winter months. fTen-year (1992002) state average of mostly green peppers [33,015.2 kgha-1 (1052 28-lb bu/acre)] (Maynard and Olson, 2003). Two rows per bed and 1. 80 m (5.9 ft) between centers of the rows. Yields for colored peppers grown in the field are lower. Yield data are not available for color peppers grown under field conditions in Florida.

PAGE 135

135 Table 4-4. Total harvested area, production, yield, and dollar valu es for field-grown bell peppers and greenhouse-grown peppers (season 1997) in Florida. aCrop season 1997; mostly green peppers (Florida Dept. of Agriculture and Consumer Service, 2002). Annual average fruit price calculated from total value divided by total production. bColored bell peppers produced in 1998 (U.S. Dept. of Agriculture, 1998). Area in the year 2003 was 14 ha (34.6 acres). c1.0 ha = 2.47 acres. d1.0 tha-1 = 0.45 ton/acre. eTotal production and average fruit yield in the tota l greenhouse area were estimated using the average fruit price for colored peppers in 1998 and total whol esale value and production area reported by the U.S. Dept. of Agriculture (1998). f$1.00/kg = $0.454/lb. tWholesale price at Miami terminal market, during th e harvest period Nov. 1997May 1998 (calculated from U.S. Dept. of Agriculture data, see Methods section in this manuscript). g$1.00/ha = $0.405/acre. Production and Value FieldaGreenhouseb Area (ha)c 7,611 9.5 Production (t)d 256,337 750e Yield (t/ha) 33.7 78.9 Value Fruit price ($/kg)f 1.085.09g Total (million $) 276.2 3.8 Average per ha ($/ha)h 36,294 401,684

PAGE 136

136 Table 4-5. Structure dimensions of one of th e two identical greenhouse units, and pepper plant arrangement used in the en terprise budget analysis. Structure dimensions and plant arrangement One greenhouse unit Structure dimensions Gutter base height (m)a 4 Front width (m) 70 Total no. of bays (no.) 10 Width of each bay (m) 6.40 Side length (m) 56 Length of each bay (m) 52 All-around distance from sides (m) 2 Area of a bay [6.40 52-m] (m2) 333 Total area [70 56-m] (m2) 3,920b Growing area [64 52-m] (m2) 3,328b Plant arrangement Distance between plant rows (m) 1.28 Within-row plant spacing (m) 0.25 Plant rows/bay (no.) 5 Plants/row (no.) 208 Plants/bay (no.) 1,040 Plant density in the growing area (no./m2) 3.13 Total no. of plants (no.) 10,400c a1.0 m = 3.28 ft. bThe total floor area for the two greenhouse units was 7,840 m2, 0.78 ha (84,391.8 ft2, 1.927 acre), and the total crop growing area was 6,656 m2, 0.67 ha (71,646.9 ft2, 1.655 acre). cThe total no. of plants in the two units was 20,800.

PAGE 137

137 Figure 4-3. Passively ventilated greenhouses at the Protected Agri culture Center in Gainesville, Fla. The greenhouse units used in the budget analysis presented in this report were similar but with a larger area. Figure 4-4. Pepper plants grown in 11.4-L (3 ga l) containers and trellised to the Spanish system in a passively ventilated greenhouse at the Protected Agri culture Center in Gainesville, Fla.

PAGE 138

138 Table 4-6. Monthly marketable fr uit yields, average wholesale ma rket prices, and gross revenues in a typical fall to spring greenhouse-grown bell pepper crop in Florida with a total estimated yield of 13 kgm (2.7 lb/ft2). Aug. Sep. Oct. Nov. Dec. Jan.Feb.Mar.Apr.May JuneJuly Nov. May Yielda (kgm-2) Tb 1.0 3.01.52.02.02.51.0 En d 13 Pricec ($/kg) 3.76 4.09 3.98 4.58 5.285.63 4.755.306.005.50 4.68 4.09 5.29dG. Revenuee ($/m2) 4.58 15.848.459.5010.6015.005.50 69.47fG. Revenue ($/0.78-ha) 30,484 105,431 56,243 63,232 70,55499,840 36,608 462,392 aMonthly fruit yields estimated from experimental crops at the University of Florida (Jovicich, 2001; Shaw and Cantliffe, 2002). 1.0 kgm-2 = 0.21 lb/ft2. bT: transplant. Seeding: 1 July, Transplant: Aug. 4, End of crop: 30 May. Harvesting period: Nov.May. cAverage wholesale prices (years 1993) for tran sactions of imported greenhouse-grown bell peppers at the Miami terminal market. dMean for the harvesting period Nov.May. $1.00/kg = $0.454/lb. The annual mean was $4.80/kg ($2.18/lb). eGross Revenue. $1.00/m2 = $0.093/ft2. fWith a greenhouse growing area that was 85% fro m the total area the gross revenue was $58.98/m2 ($5.49/ft2) of total area.

PAGE 139

139 Table 4-7. Summary of the investment costs and deprec iation of a 0.78-ha (1.927-acre) greenhouse operation planned fo r bell pepper production. Original cost Depreciation/year Invested item ($/0.78-ha)a ($/m2)b Projected life (years) ($/0.78-ha) ($/m2) Site preparation Labor leveling, compacting 20,000 2.55 Limerock and millings 6,000 0.77 Water piping to greenhouse complex 5,000 0.64 Site electrical/communications to complex 20,000 2.55 Total site work 51,000 6.51 Greenhouse permitting 1,500 0.19 Greenhouse structure and cover materials Columns, arch, gutters, polyethylene lo cking profiles 86,600 11.05 20 4,330 0.55 Access gates, 4 pavilions 3,400 0.43 10 340 0.04 Side-wall and roof-vent motors 14,900 1.90 10 1,490 0.19 Insect proof netting, 50-mesh (a ll openings) 3,860 0.49 10 386 0.05 Polyethylene cover 8,740 1.11 3 2,913 0.37 Thermal and shading screen 41,800 5.33 10 4,180 0.53 Freight overseaGainesville 10,000 1.28 10 1,000 0.13 White ground cover 5,280 0.67 7 754 0.10 Total greenhouse structure and cover materials 174,580 22.27 15,394 1.96 Greenhouse erection and concrete (by contractor) 160,000 20.41 10 16,000 2.04 Construction supervision 6,000 0.77 10 600 0.08 Head house structures (15 10 m)c 20,000 2.55 20 1,000 0.13 Fruit size grading machine 5,000 0.64 10 500 0.06 Refrigeration room 20,000 2.55 20 1,000 0.13 Backup generator 4,000 0.51 10 400 0.05 Heating and ventilation systems Floor mounted heating uni ts (diesel) 20 heating units 80,639 kcal e ach 51,620 6.58 10 5,162 0.66 Polyethylene convection tube (19 300 m per roll) 1,330 0.17 3 443 0.06 Diesel tank (11,340 L) with shading roof d 3,600 0.46 8 450 0.06 Site Diesel plumbing 3,000 0.38 10 300 0.04 Air circulation fans (60 units) 12,000 1.53 8 1,500 0.19 Total heating and ventilation systems 71,550 9.13 7,855 1.01 Irrigation and climate control systems Water well and pumps 10,000 1.28 15 667 0.09 Water tanks (2 x 56,700 L each) 26,000 3.32 15 1,733 0.22 Nutrient injector and climate control system 26,495 3.38 10 2,650 0.34 Nutrient solution tanks (6 2,000 L) 5,100 0.65 10 510 0.07 Weather station and temperature and humidity sensors 8,000 1.02 10 800 0.10 Computer and software 5,000 0.64 5 1,000 0.13 Training for using control systems 1,500 0.19 Water filters 700 0.09 10 70 0.01 Valves and pressure regulators 2,887 0.37 5 577 0.07 Irrigation emitters, stakes, and tu bing 22,575 2.88 5 4,515 0.58 Polyethylene pipe (5,700 m) 1,583 0.20 5 317 0.04 Pipe connectors and adaptors 550 0.07 5 110 0.01 Other irrigation parts and labor 5,000 0.64 5 1,000 0.13 11.4-L nursery pots 14,700 1.88 5 2,940 0.38 Total irrigation and cl imate control systems 130,190 16.61 16.889 2.17 Electrical 80,000 10.20 10 8,000 1.02 Drainage system (t roughs, pipes, pump) 3,120 0.40 5 624 0.08 Bulk storage tanks (3 tanks of 7600 L each) 12,300 1.57 10 1,230 0.16 Trellis accessories Cables for plant support (5400 m) and "U" clamps 5,600 0.71 10 560 0.07 Poles for plant support (13 per row) 6,500 0.83 10 650 0.08 Stem ring clips 1,050 0.13 2 525 0.07 Total Trellis accessories 13,150 1.68 1,735 0.22 Automotive (medium-duty delivery truck) 40,000 5.10 10 4,000 0.51

PAGE 140

140 Table 4-7. Continued Original cost Projected life Depreciation/year Invested item ($/0.78-ha)a ($/m2)b (years) ($/0.78-ha) ($/m2) Fork lift 12,000 1.53 10 1,200 0.15 Other durables Scales 1,500 0.19 5 300 0.04 Sprayer and fogger 2,000 0.26 5 400 0.05 pH meter 150 0.02 5 30 0.00 Electrical conductivity me ter 250 0.03 5 50 0.01 Ion meters for nitrate and pot assium 700 0.09 4 176 0.02 Harvest trolleys 1,500 0.19 6 250 0.03 Harvest bins 6,000 0.77 6 1,000 0.13 Tools 4,000 0.51 4 1,000 0.13 Total Other durables 16,100 2.05 3,205 0.41 Total investment 820,490 104.67 79,632 10.18 a$1.00/0.78 ha = $1.00/1.927 acre. b$1.00/m2 = $0.093/ft2. c1.0 m = 3.28 ft. 1.0 kcal = 3.97 Btu. d1.0 L = 0.26 gal. Table 4-8. Estimated fixed cost s to produce bell pepper in a 0.78-ha (1.927-acre) greenhouse in north central Florida. TOTAL Item ($/0.78-ha)a ($/m2)b Depreciation 79,632 10.18 Other fixed costs Repairs and maintenance 6,000 Taxes and licenses 1,600 Greenhouse insurance 4,000 Vehicle insurance 1,500 Telephone 4,500 Other expenses 2,500 Total other fixed costs 20,100 2.56 Total fixed costs 99,732 12.74 a$1.00/0.78 ha = $1.00/1.927 acre. b$1.00/m2 = $0.093/ft2.

PAGE 141

141 Table 4-9. Estimated variable costs to produ ce bell pepper in a 0.78-ha (1.927-acre) greenhouse in north central Florida. Unit Quantity Price Amount TOTAL Item (no. units) ($/unit) ($/0.78 ha)a ($/0.78-ha) ($/m2)b Production costs Preharvest Fertilizer 13,978 1.78 480 g/plant used in 298 days kgc 9,984 1.40 13,978 Biologicals 3,610 0.46 A. colemani (1/m2) 2 releases 40 35.00 1,400 N. californicus (10/m2) 1 release 67 30.00 2,010 B. thuringiensis 2 drain applications 9.45-Ld 2 100.00 200 Pesticides 1,078 0.14 Fungicidese Spray 15 60.00 900 Miticidef Spray 1 120.00 178 Pollinators 880 0.11 Bumble bees 50 bees/hive 4 220.00 880 Other materials inputs 13,937 1.78 Twine spool 3,000 mg 8 13.00 104 Double hooks unit 16,000 0.01 160 Bleach L 154 0.26 40 Seedling trays unit 110 3.00 330 Media seedlings m3 1.2 70.00 84 Seed unit 21,840 0.35 7,644 Media for pots (perlite) m3 260 40.00 5,200 Sticky cards (insect pest monitoring) box 30 25.00 375 Energy 38,284 4.88 Diesel L 135,708 0.26 35,284 Electricityh kWh 30,000 0.10 3,000 Labor Times Total h 18,421 2.35 Seeding and seedling growing h 1 120 886 Preparation greenhouse h 1 363 2,679 Transplanting h 1 58 428 Plant support with twines and hooks h 6 333 2,458 Removal of cull fruits, old leaves and shoots h 2 150 1,107 Fertilizer preparation h 13 39 288 Solution monitoring and filter cleaning h 30 60 864 Scouting (pests, diseases and beneficials) h 35 140 2,520 Removal of plants and cleaning h 1 480 3,542 Polyethylene cover change (every 3 years) h 0.33 81 199 Pesticide application h 15 195 2,808 Empting and washing of pots (every 2 years) h 0.5 174 642 Total labor h 2,052 Total preharvest costs 90,188 11.50 Harvest Pick labor (75 h/harvest 30 harvests) h 2,250 7.38 16,605 Total Harvest costs 16,605 2.12

PAGE 142

142 Table 4-9. Continued Unit Quantity Price Amount TOTAL Item (no. units) ($/unit) ($/0.78 ha)a ($/0.78-ha) ($/m2)b Production costs Packing and Marketing Pack labor (1,442 h) kg 86,528 0.12 10,383 Cartons, dividers and labels kg 86,528 0.16 13,845 Marketing and miscellaneous packing kg 86,528 0.22 19,036 Vehicle operation kmi 16,000 0.19 3,000 Total packing and marketing costs 46,264 5.90 Total production costs 153,057 19.52 a$1.00/0.78 ha = $1.00/1.927 acre. b$1.00/m2 = $0.093/ft2. c1.0 kg = 2.20 lb. d1.0 L = 0.26 gal. eActive ingredient: azoxystrobin, myclobutanil, and Ampelomyces quisqualis. fActive ingredient: abamectin. g1.0 m = 3.28 ft. hElectric energy unit is kilowatthour (kWh). i1.0 km = 0.62 miles.

PAGE 143

143 Figure 4-5. Means and standard deviations for bell pepper fruit wholesal e prices from years 1993 to 2002 obtained from Thursdays transactions at the Miami terminal market: A) prices for yellow ( ), red ( ), and green ( ) fruits produced in the field (shipped from Florida, California, Georgia, and No rth Carolina), and B) prices for yellow ( ), red ( ), and orange ( ) fruits imported and produced in greenhouses (shipped from The Netherlands, Israel, a nd Spain). $1.00/kg = $0.454/lb. A B

PAGE 144

144 Table 4-10. Enterprise budget for greenhousegrown bell pepper [0.78 ha (1.927 acres)] in north-central Florida. Quantity Price Amount Total Item Unit (no. units) ($/unit) ($/0.78 ha)c ($/0.78 ha) ($/m2)d Gross revenue k g a 86,528 4.58 6.00 b 462,392 58.98 Variable costs Preharves t Fertilize r 13,978 Biolo g icals 3,610 Pesticides 1,078 Pollinators 880 Other material inputs 13,937 Diesel Le 135,708 0.26 35,284 Electricit y f kWh 30,000 0.10 3,000 Labo r h 2052 18,421 Total p reharvest costs 90,188 11.50 Harvest Labo r h 2,250 7.38 16,605 Total harvest costs 16,605 2.12 Packin g and marketin g Pack labor (1,442 h) k g 86,528 0.12 10,383 Cartons, dividers and labels k g 86,528 0.16 13,845 Marketin g and miscellaneous packin g k g 86,528 0.22 19,036 Vehicle k m g 16,000 0.19 3,000 Total p ackin g and marketin g costs 46,264 5.90 Total above variable costs 153,057 19.52 Sale transaction expenses (15%) 0.15 462,392 69,359 69,359 8.85 Total variable costs 222,416 28.37 Fixed costs Depreciation 79,632 Other fixed costs 20,100 Total fixed costs 99,732 12,74 Total cost 322,148 41.09 Return to capital and managementh 140,244 17.89 a1.0 kg = 2.20 lb. bRange of average wholesale prices for greenhouse-gr own bell peppers during the harvest period. Gross revenue was the sum of estimated monthly revenues. Revenues in each month were calculated as the product of the average market price by the estimated monthly yield [total estimated fruit yield: 13 kgm-2 (2.7 lb/ft2) of growing area]. c1.0 ha = 2.47 acres. d$1.00/m2 = $0.093/ft2. e1.0 L = 0.26 g fElectric energy unit is kilowatthour (kWh). g1.0 km = 0.62 miles. hLand, opportunity costs (interest on investment capital and interest on operating capital), and the growers salary were not included in the budget.

PAGE 145

145 Figure 4-6. Estimated responses of total costs, returns to capital and management (excluding cost of land), and gross revenues to in creased marketable fruit yields of greenhouse-grown bell peppers in north-centra l Florida. Costs, revenues, and returns based on unit area of a 0.78-ha (1.927-acre) greenhouse. Fruit yields expressed per unit of growing area. Average fruit whol esale price for the production season was estimated at $5.29/kg ($2.40/lb). Returns to capital and to management were 0 (zero) at fruit yield of 7.8 kgm (1.60 lb/ft2); $1.00/m2 = $0.093/ft2; 1.0 kgm-2 = 0.21 lb/ft2.

PAGE 146

146 Table 4-11. Sensitivity analysis for bell pepper re turns to capital and management per unit area and in the 0.78-ha ( 1.927-acre) greenhouse. Yielda Costb Market pricec ($/5-kg carton) (kgm-2) ($/m2) 15.00 17.50 20.00 22.50 25.00 26.45d 27.50 30.00 32.50 Return ---------------------------------------------------$/m2 ($/0.78-ha) ----------------------------------------------------------------5 28.61 -17.79 (-139,439) -15.98 (-125,296) -14.18 (-111,153) -12.37 (-97,010) -10.57 (-82,868) -9.52 (-74,665) -8.77 (-68,725) -6.96 (-54,582) -5.16 (-40,440) 7 29.52 -14.36 (-112,612) -11.84 (-92,812) -9.31 (-73,012) -6.79 (-53,212) -4.26 (-33,413) -2.80 (-21,929) -1.74 (-13,613) 0.79 (6,187) 3.31 (25,987) 9 30.42 -10.94 (-85,785) -7.69 (-60,328) -4.45 (-34,871) -1.20 (-9,414) 2.05 (16,042) 3.93 (30,807) 5.29 (41,499) 8.54 (66,956) 11.79 (92,413) 11 31.33 -7.52 (-58,958) -3.55 (-27,844) 0.42 (3,270) 4.39 (34,384) 8.35 (65,498) 10.66 (83,544) 12.32 (96,611) 16.29 (127,725) 20.26 (158,839) 13 32.24 -4.10 (-32,131) 0.59 (4,640) 5.28 (41,411) 9.97 (78,182) 14.66 (114,953) 17.38 (136,280) 19.35 (151,724) 24.04 (188,495) 28.73 (225,265) 15 33.15 -0.68 (-5,304) 4.74 (37,124) 10.15 (79,552) 15.56 (121,980) 20.97 (164,408) 24.11 (189,016) 26.38 (206,836) 32.79 (249,264) 37.21 (291,692) 17 34.96 2.75 (21,522) 8.88 (69,608) 15.01 (117,693) 21.15 (165,778) 27.28 (213,863) 30.84 (241,752) 33.41 (261,948) 39.55 (310,033) 45.68 (358,118 aFruit yields are expressed on a growing floor area ba sis but returns are based on unit area of 0.78 ha (1.927 acres). 1.0 kgm-2 = 0.21 lb/ft2. Fruit yields on a growing floor area basis multiplied by 0.85 give the yield on a unit area of the total 0.78 ha (1.927 acres). bTotal production costs without the sales transaction cost included. The transaction cost (15%) was deducted from the gross revenues for each yieldprice combination. $1.00/m2 = $0.093/ft2. $/0.78 ha = $/1.927 acres. cAverage market price for the harvest period Nov.May. $/5-kg carton = $/11.0-lb carton. dAverage wholesale price for transactions of importe d greenhouse-grown peppers at the Miami terminal market during Nov.May and over the year period 19932002 [$5.29/kg ($2.40/lb)]. Table 4-12. Estimated break-even prices for a rang e of marketable bell pepper fruit yields of 5 17 kgm-2 (1.0.5 lb/ft2). Yield in growing areaa Break-even priceb (kgm-2) (carton/m2) ($/kg) ($/5-kg carton) 5.0 1.0 7.9539.73 7.0 1.4 5.8429.22 7.8 1.6 5.29c26.45 9.0 1.8 4.6823.40 11.0 2.2 3.9419.71 13.0 2.6 3.4317.15 15.0 3.0 3.0715.35 17.0 3.4 2.7813.90 aFruit yields are expressed on a growing floor area basis. 1.0 kgm-2 = 0.21 lb/ft2. 1.0 carton = 5 kg (11.0 lb). 1.0 carton/m2 = 0.09 carton/ft2. b$1.00/kg = $0.454/lb. cAverage wholesale price for transactions of importe d greenhouse-grown peppers at the Miami terminal market during Nov.May and over the year period 19932002.

PAGE 147

147 APPENDIX A ADDITIONAL FIGURES AND TA BLES FOR CHAPTER 2 Figure A-1. Pest-predator scenar ios evaluated on peppe r seedlings. Seedlings were infested with two P. latus at three growth developmental stages (unfolded cotyledons, two leaves, or four leaves) and two N. californicus were released before, at, or after the day of infestation. Additional seedling groups had only one mite species introduced at one of each developmental growth stage, or no mite s introduced. COTYLEDONS 13 DAS Non-infested 2 LEAVES 28 DAS 4 LEAVES 37 DASPC P P2 P P4 P 7-8 LEAVES 42 DAS 9 d 15 d 6 d N P PC-N2 P N PC-N4 P N P2-N4 P + N PC-NC P + N P2-N2 P + N P4-N4 N P P2-NC P N P4-N2 P N P2-N4 N NC N N2 N N4

PAGE 148

148 C B D E FNC P P P P P A E C B D E FNC P P P P P A E C B D E FNC P P P P P A E Figure A-2. Procedure used for recovering mites from seedlings or leaves. Whole seedlings were sampled in the experiment described in Chapter 2 and terminal leaves were sampled in both greenhouse experiments desc ribed in Chapter 3. (A) Whole seedling, (B) terminal leaves, (C) et hanol-wash and centrifugation, (D ) concentration of mites, (E) counting of mites, and (F) measurement of growth variables on seedlings and leaf area on sampled leaves. P: P. latus ; NC: N. californicus

PAGE 149

149 P. latus cumulative mite days (mite-days / seedling) 1 10 100 1000 1 0000 Transplant stem length (m m) or leaf area (cm 2 ) 0 20 40 60 80 100 120 140 160 180 Transplant dry weight (mg) 0 50 100 150 200 250 300 Transplant leaf number (no.) 0 1 2 3 4 5 6 7 8 Damage index 0 1 2 3 4 Damage index Leaf number Stem length Leaf area Dry weight Figure A-3. Transplant growth variable s and damage index as affected by P. latus cumulative mite-days (in a log.-scale) in 42-days old peppers. Leaf area ( ; purple), stem length ( ; blue), dry weight ( ; green), leaf number ( ; red), and damage index ( ; black). Data from all treatments and replications were used in the analyses (n = 48). Regression equations are pr esented in Table A-1. Table A-1. Regression equa tions for growth variables and for visual damage index as affected by increased P. latus cumulative mite-days, in 42-days old pepper transplants. Growth parameter and visual damagea Equation R2 SE of the estimate F P Dry weight (mg) = 281.20 e 0.0003 x 0.75 42.6 136.8 0.0001 Leaf area (cm2) = 73.59 e0.0003 x 0.74 12.1 129.9 0.0001 Leaf number (no.) = 6.8 e0.0001 x 0.68 0.79 97.6 0.0001 Stem length (mm) = 136.46 e0.0002 x 0.74 15.9 77.8 0.0001 Damage index [from 0 (undamaged) to 4 (serious damage)] = 4 e0.0004 x 0.86 0.54 290.9 0.0001 aAnalyses used data of all treatments and replications (n = 48).

PAGE 150

150 P. latus cumulative mite da ys (mite-days / seedling) 1 1 0 100 1000 10000 Relative reduction to non-infested transplant (%) -20 0 20 40 60 80 100 Damage index 0 1 2 3 4 Leaf number Damage index Stem length Dry weight Leaf area Figure A-4. Regression curves fo r seedling relative gr owth reduction and visual damage index as affected by increased P. latus cumulative mite-days (in a log.-scale), in 42-days old pepper seedlings. Leaf area ( ; purple), stem length ( ; blue), dry weight ( ; green), leaf number ( ; red), and damage index ( ; black). Data from all treatments that had P. latus and three replications were used in the analyses (n = 39). For the zero percent growth reduction, data from seedli ngs that were mite-free and that only had N. californicus were averaged. Regression equa tions are presented in Table A-2. Table A-2. Regression equations for seedling relative growth re duction and for visual damage index as affected by increased P. latus cumulative mite-days, in 42-days old pepper seedlings. Relative growth reduction and visual damagea Equation R2 SE of the estimate F P Dry weight (%)b = 68.24 (1 e0.0008 x) 0.78 14.0 128.2 0.0001 Leaf area (%) = 83.90 (1 e0.0004 x) 0.74 16.1 107.7 0.0001 Leaf number (%) = 0.0085x 0.62 13.6 62.4 0.0001 Stem length (%) = 63.11 (1 e0.0005 x) 0.69 13.0 84.0 0.0001 Damage index [from 0 (undamaged) to 4 (serious damage)] = 4 e0.0004 x 0.860.54 290.9 0.0001 aData of all treatments with P. latus infestations and replications were combined (n = 39). bFor the zero percent growth reduction, data from se edlings that were mite-free and that only had N. californicus were averaged.

PAGE 151

151 Figure A-5. Pepper transplants with P. latus infestation and no N. californicus release (42 DAS). Seedlings were infested with two P. latus at (A) unfolded cotyledons, PC (B) unfolded two leaves, P2 and (C) unfolded four leaves, P4 New leaves that were damaged are indicated with an arrow. Figure A-6. Pepper transplants with N. californicus released on the same day P. latus infested seedlings (42 DAS). Seedlings that were infested with two P. latus and had two N. californicus released at (A) unfolded cotyledons, PC-NC (B) unfolded two leaves, P2-N2 and (C) unfolded four leaves, P4-N4 New leaves that were damaged are indicated with an arrow. A BC A BC

PAGE 152

152 Figure A-7. Pepper transplants with N. californicus released before P. latus infested seedlings (42 DAS). Seedlings had (A) two N. californicus released on cotyledons and two P. latus infested 15 d later, PC-N2 (B) two N. californicus released on two unfolded leaves and two P. latus infested 24 d later PC-N4 and (C) two N. californicus released on two unfolded leaves and two P. latus infested 9 d later P4-N2 Figure A-8. Pepper transplants with N. californicus released after P. latus infested seedlings (42 DAS). Seedlings were infested with (A) two P. latus at unfolded cotyledons and had two N. californicus released 15 d later, PC-N2 (B) two P. latus at unfolded cotyledons and had two N. californicus released 24 d later, PC-N4 and (C) two P. latus at unfolded two leaves and had two N. californicus released 9 d later, P2-N4 Damaged leaves (white) and new undamaged leaves (blue) indicated with arrows. A BC A BC A BC

PAGE 153

153 Figure A-9. Pepper transplants with a preventive release of N. californicus and no P. latus infestation (42 DAS). Seedlings that had two N. californicus released at (A) unfolded cotyledons, NC (B) unfolded two leaves, N2 and (C) unfolded four leaves, N4 Figure A-10. Pepper transplants with no mites (42 DAS). A BC

PAGE 154

154 APPENDIX B ADDITIONAL FIGURES AND TA BLES FOR CHAPTER 3 Figure B-1. Pepper seedli ng infested with two P. latus that were transplanted in the greenhouse three days later. No mite damage sy mptoms were visibl e on the seedling at transplanting. Figure B-2. Passively ventilated greenhouse a nd arrangement of plot s in Spring-2005. (A) Passively ventilated greenhouse at the Prot ected Agriculture Center (Plant Science Research and Education Unit, IFAS, University of Florida, Citra, Fla.), (B and C) plots with bell pepper plants growing in pots with pine bark (transplanting: 9 March and first fruit harvest: 5 June, 88 DAT). A C B

PAGE 155

155 Figure B-3. Daily air temperatures (maximum mean, and minimum) inside the passively ventilated greenhouse and over canopie s of pepper plants in Fall-2004 and Spring-2005. Fall transplanting on 17 Sept and first fruit harvest on 1 Jan. 2005, 106 DAT. Spring transplanting on 9 March and first fruit harvest on 5 June, 88 DAT.

PAGE 156

156 Table B-1. Cost estimates for one sulfur spray on transp lanted bell pepper plants or seedlings for a 1-hectare greenhouse crop. Cost for one sulfur spray Spray time and inputs $/h h/ha $/ha $/m2 $/plant After transplantinga Laborb 23.684.5106.56000.0106560 0.00355200 Equipmentc 0.043.30.13200.0000132 0.00000440 Combustibled 0.763.32.50800.0002508 0.00008360 Pesticidee ----10.03000.0010030 0.00033433 Total ---119.23000.0119230 0.00397433 Before transplanting Laborf 23.680.0300.71040.0000710 0.00002378 Equipment 0.040.0220.00090.0000001 0.00000003 Combustible 0.760.0220.01670.0000017 0.00000056 Pesticide ----10.03000.0010030 0.00033433 Total --10.75800.0010758 0.00035860 aIn a greenhouse area of 1 ha with 30,000 plants. bPermanent employee assumed, with salary ($13.93/h) and benefits ($9.75/h). Estimate for standard occupational classification # 37-3 012, U.S. Dept. of Labor (http://www.bls.gov/ ). Labor time includes preparation, water and pesticide recharge, cleaning of equipment, and break. Walking distance for spraying is 8,250 m (150 plant rows 55 m/plant row) at a walking speed of 0.7 m/s. cMotorized sprayer backpack (e.g. Stihl SR420, cost $600, 5 useful years) depreciation ($120/year) and maintenance ($80/year), and protective equipment (e.g. respirator, boots, clothing, and gloves; $166/year). dGas and oil (1:25 v/v mix), consumed at a rate of 0.95 L/h. eMicronized sulfur 80% w/w formulation ($2.9/kg). Appli cation rate is 3.4 kg/ha of formulated product. Volumes applied to runoff and increase with increase d plant size, from 0.01 L/seedling to 0.20 L/plant. fOn transplant plugs ( seedling density 113 per m2), labor time for spraying is estimated at 1/150 of that on transplanted plants.

PAGE 157

157 Table B-2. Cost estimates for one release N. californicus on transplanted bell pepper plants or seedlings for a 1-hectare greenhouse crop. Cost of one release of 2 N. californicus /plant Cost of one release of 4 N. californicus /plant Release time and inputs $/h h/ha $/ha $/m2 $/plant $/ha $/m2 $/plant After transplantinga Laborb 23.68 5.0 118.400.01180.0039 118.40 0.0118 0.0039 Predatory mitec --330.000.03300.0110 660.00 0.0660 0.0220 Shippingd --455.000.04550.0152 910.00 0.0910 0.0303 Total --903.400.09030.0301 1688.40 0.1688 0.0563 Before transplanting Labore 23.68 2.0 47.360.00470.0016 47.36 0.0047 0.0016 Predatory mite --330.000.03300.0110 660.00 0.0660 0.0220 Shipping --455.000.04550.0152 910.00 0.0910 0.0303 Total --832.360.08320.0277 1617.36 0.1617 0.0539 aIn a 1-hectare greenhouse with 30,000 plants. bPermanent employee assumed, with salary ($13.93/h) and benefits ($9.75/h). Estimate for standard occupational classification # 37-3 012, U.S. Dept. of Labor (http://www.bls.gov/ ). Labor time includes calibration for release density and release by gently sh aking vials over top plant leaves. Walking distance during release is 8,250 m (150 plant rows 50 m/plant row plus turning points) at a walking speed 0.5 m/s. cN. californicus in vials containing approx. 1000 predatory m ites and an inert carrier. Estimated cost for a large order: $5.50/vial. A release density of 2 and 4 mites per plant requires 60 and 120 vials/ha, respectively. dPacking ($4.5/6-vials-box) added to air and ground overnight shipping (CA to FL) is $41/6-vials-box. eCalibration, walking and release rate (0.2 s/seedli ng) used for a release on transplant plugs (seedling density 113 per m2).

PAGE 158

158Table B-3. P. latus cumulative mite-days, plant growth variables, and plan t visual damage index in gr eenhouse-grown pepper plants in Fall-2004 (106 DAT). Plants were treated to manage a P. latus infestation initiated three days before transplanting. Leaves Stems Treatment codea Cumulative mite-days (mite-days/cm2) Fresh weight (g/plant) Dry weight (g/plant) Number (no./plant ) Area (cm2/plant) Fresh weight (g/plant) Dry weight (g/plant) Node number (no./plant) Diameter (mm) Plant height (cm) Damage index (0) N. californicus 2N+4 735 ab 452.1 ab 69.8 ab 254ab 9,217ab 301.7 bc 50.3b 61ab 16.2a 103bc4.0ab 2N+4,21 566 ab 406.2 ab 64.4 ab 226ab 10,395ab 264.5 bc 45.2bc68ab 16.0a 97bc2.6ab 2N+4,21,30 581 ab 447.6 ab 67.4 ab 276ab 12,118a 393.9 ab 54.8ab72ab 15.5ab111b 2.9ab Sulfur 4S+16 390 b 541.4 a 69.2 ab 279ab 14,763a 468.7 ab 62.0ab103a 16.8a 153a 0.3b Non-treated 1529 a 258.6 b 36.9 b 105b 3,812b 155.2 c 17.0c 38b 12.9b 71c 6.0a Non-infested 0 c 645.8 a 79.1 a 317a 15,914a 567.8 a 82.3a 100a 17.8a 176a 0.0c Significance F P 12.2 0.0001 6.7 0.0054 4.5 0.0208 7.6 0.0034 7.1 0.0045 10.3 0.0011 12.6 0.0005 5.6 0.0105 9.1 0.0017 31.7 0.0001 5.1 0.0143 Means within the same column followed by different lette rs are significantly different based on Tukey-Kramer test, P < 0.05. aTreatments were: two N. californicus released 4 DAT (2N+4), a second release 21 DAT (2N+4,21), or a third release 30 DAT (2N+4,21,30); four sulfur sprays initiated 16 d after transplanting (4S+16); plants infested with P. latus and no pest management (Non-treated); plants with no P. latus (Non-infested).

PAGE 159

159Table B-4. P. latus cumulative mite-days, plant growth variables, and plan t visual damage index in gr eenhouse-grown pepper plants in Spring-2005 (88 DAT). Plants were treated to manage a P. latus infestation initiated three days before transplanting. Leaves Stems Treatment codea Cumulative mite-days (mite-days/cm2) Fresh weight (g/plant) Dry weight (g/plant) Number (no./plant) Area (cm2/plant) Fresh weight (g/plant) Dry weight (g/plant) Node number (no./plant) Diameter (mm) Plant height (cm) Damage index (0) N. californicus 2N 105 bcd 622.6 ab88.0ab 424ab 15,879ab 431.2ab 70.9a 138ab 17.5ab 106a 0.0b 2N0 321 abc 595.1 bc80.9ab 384bc 14,101ab 386.6ab 61.8ab127ab 16.8ab 104a 1.3b 2N+4 524 ab 461.0 c 64.0bc 319c 9,448cd 255.8cd 43.2bc93bc 16.8ab 93a 0.2b 4N 18 d 731.6 ab102.5a 476ab 19,192a 497.2ab 77.9a 157a 19.0a 111a 0.0b 4N0 28 cd 702.9 ab97.6a 431ab 18,057ab 396.3ab 60.1ab151a 17.5ab 100a 0.0b 4N+4 117 bcd 617.1 ab91.9a 421ab 15,979ab 371.6bc 58.8ab145ab 17.5ab 106a 0.0b Sulfur 384 abcd 633.7 ab92.8a 416abc16,704ab 431.7ab 65.8ab149ab 18.2a 108a 1.7b 5S0 339 abc 692.9 ab102.3a 468ab 18,244ab 454.8ab 78.2a 150a 18.5a 106a 1.0b Non-treated 1468 a 302.0 d 44.7c 156d 4,966d 176.2d 25.4c 55c 14.2b 63b 5.0a Non-infested 0 e 748.7 a 104.5a 491a 19,493a 512.3a 80.5a 174a 18.3a 113a 0.0b Significance F9, 18 P 7.2 0.0001 23.4 0.0001 14.4 0.0001 25.9 0.0001 23.5 0.0001 17.5 0.0001 11.6 0.0001 9.8 0.0001 4.2 0.0048 10.0 0.0001 9.7 0.00 01 Means within the same column followed by different lette rs are significantly different based on Tukey-Kramer test, P < 0.05. a Treatments were: one-time releases of N. californicus at a per plant predator density of either two (2N; 2N0; 2N+4) or four (4N; 4N0; 4N+4), with release time either at 0, or 4 DAT, r espectively; five sulfur sprays initiated either DAT (5S) or 0 DAT (5S0); plants infested with P. latus and no pest management (Non-treated); plants with no P. latus (Non-infested).

PAGE 160

160 Time from an infestation initiated with 2 P. latus / seedling (d) -4-3-2-1012345678 Fruit yield relative to non-infested plants (%) 0 10 20 30 40 50 60 70 80 90 100 2 N. californicus / seedling 4 N. californicus / seedling abc c d ab abc abc PestPredatorPredator Predator Transplanting Figure B-4. Relative marketable fruit yi eld in pepper plants infested with P. latus and that had one release of N. californicus at two release densities a nd at three different release times. Per plant predator density was either two or four, with release time either at 6, 0, or 4 DAT, respectively (at 3, or 7 days after seedlings were infested with two P. latus respectively). Spring-2005; transplanti ng in greenhouse: 9 March; first fruit harvest: 5 June, 88 DAT. Letters for comparis ons are based on Tukey-Kramer test for mean separation (see Table 3-5).

PAGE 161

161 Figure B-5. Relationships between selected variables measured in pepper plants and P. latus cumulative mite-days in infested plants treat ed with or without sulfur in Spring-2005. Plant damage index (A) relative fruit yield (B), leaf area, (B) and fruit set (C). Regressions were calculated for data of thr ee replications (two sulfur treatments, n = 6; no sulfur, n = 24).

PAGE 162

162 Figure B-6. Regression curves for variables measured in bell pepper plants and P. latus cumulative mite-days (in a log.-scale) at the time of first fruit harvest in experiments Fall-2004 and Spring-2005. Visual plant damage index (A). Percentages relative to plants not infested with P. latus (100%): fruit yield (B), fruit set (C), and plant growth variables (DL). Data form treatments were pooled (n = 18 in 2004 and n = 30 in 2005). Equations and values for r2, F, and Pvalues are presented in Table B-5.

PAGE 163

163 Table B-5. Regression equa tions for plant visual damage inde x, fruit yield, fruit set, and plant growth variables, as affected by P. latus cumulative mite-days (in a log.-scale) at the time of first fruit harvest on top canopy leaves of gree nhouse-grown pepper plants in Fall-2004 and Spring-2005. Data from treatments were pooled. Estimated variable Equation r2 F df P P. latus -free plant (100% value) Fall-2004 (n = 18) Plant injury indexa = 6 (1 e.0011 x) 0.52 17.6 1, 17 0.0007 0 Plant measurements (%)b Marketable yield = 100 e.0010 x 0.79 59.7 1, 17 0.0001 4.6 kgm-2 Marketable fruit set = 100 0.015 x 2.50 10-0.005 x2 0.51 7.8 2, 17 0.0001 22 % Leaf area = 100 e.0007 x 0.71 39.2 1, 17 0.0001 15,914 cm2 Leaf fresh weight = 100 e.0006 x 0.76 51.3 1, 17 0.0001 645.8 g Leaf dry weight = 100 e.0004 x 0.61 24.8 1, 17 0.0001 79.1 g Leaf node number = 100 e.0006 x 0.59 23.2 1, 17 0.0002 100 Leaf number = 100 e.0005 x 0.60 23.7 1, 17 0.0002 317 Stem diameter = 100 e.0002 x 0.72 41.9 1, 17 0.0001 17.8 mm Plant height = 100 e.0007 x 0.78 57.2 1, 17 0.0001 176 cm Stem fresh weight = 100 e.0008 x 0.68 33.9 1, 17 0.0001 567.8 g Stem dry weight = 100 e.0009 x 0.82 74.0 1, 17 0.0001 82.3 g Spring-2005 (n = 30) Plant injury index = 6 (1 e.0007 x) 0.75 83.9 1, 29 0.0001 0 Plant measurements (%) Marketable yield = 100 e.0013 x 0.73 74.8 1, 29 0.0001 5.0 kgm-2 Marketable fruit set = 100 0.014 x 3.21 10-0.005 x2 0.63 23.4 2, 29 0.0001 14 % Leaf area = 100 e.0009 x 0.65 52.1 1, 29 0.0001 19,493 cm2 Leaf fresh weight = 100 e.0006 x 0.65 51.5 1, 29 0.0001 748.7 g Leaf dry weight = 100 e.0006 x 0.65 51.4 1, 29 0.0001 104.5 g Leaf node number = 100 e.0008 x 0.50 28.1 1, 29 0.0001 174 no. Leaf number = 100 e.0006 x 0.65 50.9 1, 29 0.0001 491 no. Stem diameter = 100 e.0001 x 0.44 21.9 1, 29 0.0001 18.3 mm Plant height = 100 e.0003 x 0.51 28.7 1, 29 0.0001 113 cm Stem fresh weight = 100 e.0008 x 0.41 19.1 1, 29 0.0002 512.3 g Stem dry weight = 100 e.0007 x 0.38 17.2 1, 29 0.0003 80.5 g aScale indices ranged from zero (no injury) to six (extreme severe injury). bEstimate percentage relative to plants not infested with P. latus.

PAGE 164

164 Figure B-7. Plant canopies of gr eenhouse-grown peppers in Fall-2004 (75 DAT). Plants were infested with two P. latus three days before transpla nting and were treated with N. californicus releases and sulfur spra ys. Treatments were: two N. californicus released 4 DAT (2 N +4), a second release 21 DAT (2 N +4,21), or a third release 30 DAT (2 N +4,21,30); four sulfur spra ys initiated 16 DAT (4 S +16); plants infested with P. latus and no pest management ( Non-treated ); plants with no P. latus ( Non-infested ). Not-infested 2 N +4 2 N +4,21 2 N +4,21,30 Not-treated 4 S +16 1m

PAGE 165

165 Figure B-8. Plant canopies of gr eenhouse-grown peppers in Spri ng-2005 (45 DAT). Plants were infested with two P. latus three days before transpla nting and were treated with N. californicus releases and sulfur sprays. Tr eatments were: one-time releases of N. californicus at a per plant predator density of either two (2 N ; 2 N 0; 2 N +4) or four (4 N ; 4 N 0; 4 N +4), with release time either at 0, or 4 DAT, respectively; five sulfur sprays initiated either DAT (5 S ) or 0 DAT (5 S 0); plants infested with P. latus and no pest management (Not-treated). Plants not-infested are not shown; they had the same appearance as plants in with four predators released or with sulfur sprays. Not-treated 5 S 0 5 S 2 N 2 N +4 2 N 0 4 N 4 N +4 4 N 0 1m

PAGE 166

166 Figure B-9. Greenhouse-grown peppe r plants with leaves or fruit removed in Spring-2005 (88 DAT). Plants were infested with two P. latus three days before transplanting and were treated with N. californicus releases and sulfur sp rays. Treatments were: one-time releases of N. californicus at a per plant predator density of either two (2 N 6; 2 N 0; 2 N +4) or four (4 N ; 4 N 0; 4 N +4), with release time either at 0, or 4 DAT, respectively; five sulfur sp rays initiated either DAT (5 S ) or 0 DAT (5 S 0); plants infested with P. latus and no pest management (Not-treated). Plants not-infested are not shown; they had the sa me appearance as plants in with four predators released or with sulfur sprays. 0.5 m 5 S 0 5 S 2 N 2 N 0 4 N 4 N 0 Not-treated 2 N +4 4 N +4

PAGE 167

167 Figure B-10. Fruit on pepper plants that were not infested with P. latus and that were infested before transplanting (Fall2004). Plants (here 90 DAT; harvest of red fruit was 106 DAT) with fruit attached and leaves rem oved. (LEFT) Plant not infested with P. latus (RIGHT) Plant was infested with two P. latus 3 d before transplanting and the pest was left unchecked (all fruit were damaged or there were no fruit). Reference line is 60-cm tall. 0.5 m

PAGE 168

168 Figure B-11. Fruit on pepper plan ts treated with and without N. californicus for an early P. latus infestation (Spring-2005). Plants (88 DAT) with fruit attached and leaves removed. (LEFT) Plant was infested with two P. latus 3 d before transplanting but had two N. californicus released 6 d before transplanting, and (RIGHT) Plant was infested with two P. latus 3 d before transplanting and the pest was left unchecked (all fruit were damaged). Reference line is 60-cm tall. 0.5 m

PAGE 169

169 LIST OF REFERENCES Agriculture and Agri-food Canada, 2003. Introduc tion to the greenhouse industry. Agriculture and Agri-food Canada, Ontario, Canada. . Almaguel, L., Prez, R., Ramos, M., 1984. Ciclo de vida y fecundidad del caro Polyphagotarsonemus latus en pimiento. Cienc. Tec. Ag ric., Proteccin de Plantas 7, 93 114. Auger, P., Tixier, M.-S., Kreiter, S., Fauvel, G., 1999. Factors affecting ambulatory dispersal in the predaceous mite Neoseiulus californicus (Acari: Phytoseiidae). Exp. Appl. Acarol. 23, 235. Badii, M.H., McMurtry, J.A., 1984. Feeding behavior of some phytoseiid predators on the broad mite Polyphagotarsonemus latus [Acari: Phytoseiidae, Tarsonemidae]. Entomophaga 29, 49. Basset, P., 1985. Tarsonemid mites. In: Hussey, N.W ., Scopes, N. (Eds.), Bi ological pest control: the glasshouse experience. Cornell Univ. Press, Ithaca, N.Y., pp. 93. Blockmans, K.J.F., 1999. Commercial aspect of biological control in greenhouses. In: R. Albajes, M.L. Gullino, J.C. Van Lenteren, and Y. Elad (Eds.), Integrated pest and disease management in greenhouse crops. Kluwer Academ. Pub., The Netherlands, pp. 310. Brown, R.D., Jones, V.P., 1983. The broad mite on le mons in southern California. Calif. Agric., July-Aug., 21. Bureau of Economic and Business Research, 2000. Florida statistical ab stract 2000. Warrington College of Business Administration, 34th ed., Univ. of Florida, Gainesville, Fla. Cantliffe, D.J., Jovicich E., Hochmuth, G.J ., 1999. Where has all the good land gone? Protected vegetable cultureOur future, p. 21. In: Greenhouse techniques towards the 3rd millennium. Intl. Conf. and Brit.Israe li Wkshp., Haifa, Israel, 5 Sept. 1999. Cantliffe, D.J., VanSickle J.J., 2001. Competitiv eness of the Spanish and Dutch greenhouse industries with the Florida fresh vegetable industry. Proc. Fla. St ate Hort. Soc. 114, 283 287. Cantliffe, D.J., VanSickle J.J., 2003. Mexican co mpetition: Now from the greenhouse, p. 2. In: P. Gilreath and W.H. Stall (Eds.). Proc. Florida Tomato Inst., Naples. Castagnoli, M., Falchini L., 1993. Suitability of Polyphagotarsonemus latus (Banks) (Acari: Tarsonemidae) as a prey for Amblyseius californicus (McGregor) (Acari Phytoseiidae). Redia 75, 273. Castagnoli, M., Simoni, S., 1991. Influence of temperature on population increase of Amblyseius californicus (McGregor) (Acari: Phyt oseiidae). Redia 74, 621.

PAGE 170

170 Castagnoli, M., Simoni, S., 1994. The effect of di fferent constant humidities on eggs and larvae of Amblyseius californicus (McGregor) (Acari: Phyt oseiidae). Redia 77, 349. Castagnoli, M., Simoni, S., 1995. The effect of high temperature on development and survival rate of Amblyseius californicus (McGregor) eggs (Acari: Phyt oseiidae). In: O. Kropcynska, J. Boczek, and A. Tomczyk (Eds.), Physiologica l and Ecological Aspects of Acari Host Relationships, Oficyna DABOR, Warsaw, pp. 357. Castagnoli, M., Simoni, S., 1999. Effect of long-term feeding history on functional and numerical response of Neoseiulus californicus (Acari: Phytoseiidae) Exp. Appl. Acarol. 23, 217. Castineiras, A., Baranowski, R. M., Glenn, H., 1997. Distribution of Neoseiulus cucumeris (Acarina: Phytoseiid ae) and its prey, Thrips palmi (Thysanoptera: Th ripidae) within eggplants in south Florida. Fla. Entomol. 80, 211. Cho, M.R., Jeon, H.Y., La, S.Y., Kim, D.S., Yiem, M.S., 1996. Damage of broad mite Polyphagotarsonemus latus (Banks), on pepper growth and yield and its chemical control. Korean J. Appl. Entom. 35, 326. Collier K.F.S., Albuquerque, G.S., Eiras, A.E., Bl ackmer, J.L., Arajo, M.C., Monteiro, L.B., 2001. Estmulos olfativos envolvidos na loca lizao de presas pelo caro predator Neoseiulus californicus (McGregor) (Acari: Phytoseiidae) em macieiras e hospedeiras alternativas. Neotrop. Entomol. 30, 631. Costa, J.M., Heuvelink E., 2000. Greenhouse horticultu re in Almera (Spain): Rpt. on a study tour, 24 Jan. 2000. Grafisch Serv. Centrum Van Gils, Wageningen, The Netherlands. Cross, J.V., 1979. Broad mite, Polyphagotarsonemus latus (Blanks) causing damage to peppers. Plant Pathol., 28, 55. Davies, J.C., 1963. Observation of the effect of DDT formulation on the yellow tea mite, Hemitarsonemus latus (Banks) in Uganda. Bull. Entomol. Res. 54, 425. de Coss-Romero, M., Pea, J.E., 1998. Relationship of broad mite (Acari: Tarsonemidae) to host phenology and injury levels in Capsicum annuum Fla. Entomol. 81, 515. de Courcy Williams, M.E., Kravar-Garde, L., Fenlon, J.S., Sunderland, K.D., 2004. Phytoseiid mites in protected crops: the effect of humi dity and food availability on egg hatch and adult life span of Iphiseius degenerans Neoseiulus cucumeris N. californicus and Phytoseiulus persimilis (Acari: Phytoseiidae). Ex p. Appl. Acarol. 32, 1. de Holanda, A.C., Oliveira, J.V., Silva, I.P., 1992. Loss caused by Polyphagotarsonemus latus (Banks, 1904) (Acari: Tarsonemidae) and Tetranychus desertorum Banks, 1900 (Acari: Tetranychidae) in the beans Phaseolus vulgaris L. and Vigna ungiculata (L.) Walp. Ann. Soc. Entomol. Brasil 21, 115.

PAGE 171

171 Dik, A.J., Ceglarska, E., Ilovai, Z., 1999. Sweet pe ppers. In: R. Albajes, M.L. Gullino, J.C. Van Lenteren, and Y. Elad (Eds.), Integrated pe st and disease management in greenhouse crops. Kluwer Academ. Pub., The Netherlands, pp. 473. Echer, M.M., Fernandes, M.C.A., Ribeiro, R.L. D., Peracchi, A.L., 2002. Avaliao de gentipos de Capsicum para resistncia ao caro branco. Hort. Brasileira 20, 217. Fan, Y., Petitt, F.L., 1994. Biological control of broad mite, Polyphagotarsonemus latus (Banks), by Neoseiulus barkeri Hughes on pepper. Biol. Control 4, 390. Fan, Y., Petitt, F.L., 1998. Dispersal of the broad mite, Polyphagotarsonemus latus (Banks) (Acari: Tarsonemidae) on Bemisia argentifolii (Homoptera: Aleyrodidae). Exp. Appl. Acarol. 22, 411. Faraji, F., Janssen, A., Sabelis, M.W., 2002. Ovi position patterns in a predatory mite reduce the risk of egg predation caused by prey. Ecol. Entomol. 27, 660. Fasulo, T.R., 2005. Broad Mite, Polyphagotarsonemus latus (Banks) (Arachnida: Acari: Tarsonemidae). Florida Coop. Ext. Electronic Da ta Info. Source (EDIS), Univ. of Florida, Gainesville. 5 Jan. 2007. . Federal Reserve, 2004. Bank prime loanAnnual Fede ral Reserve Bank of Atlanta, Consumer Serv., Stat. Release. Board of Governors of the Federal Re serve System, Washington, DC. 6 July 2004. . Fintrac, 2003. Peppers fresh or chilled: U.S. imports for consumption. Agribusiness online. Fintrac Inc., Washington, DC. 21 Aug. 2003. . Fitch, G.P., Margolies, D.C., Nechols, R.J ., Williams, K.A., 2006. Overhead and drip-tube irrigation affect twospotted spider mites and their biological control by a predatory mite on impatiens. HortScience 41, 691. Flechtmann, C.H., Guerrero, B.J.M., Arroyave J.A., Constantino C.L.M., 1990. A little known mode of dispersal of Polyphagotarsonemus latus (Banks). Int. J. Acarol. 16, 181. Florida Agriculture Statistics Service, 2003. Florid a facts: Bell peppers. Florida agricultural statistics vegetable summary, 2001. Fla. Dept Agr. and Consumer Serv., Tallahassee, Fla. 13 June 2003. . Florida Dept. of Agriculture and Consumer Se rvice, 2002. Florida agricu lture facts directory. 2002 Directory. Fla. Dept. Agr. a nd Consumer Serv., Tallahassee. Frank, C.A., Nelson, R.G., Simonne, E.H., Behe, B.K., Simonne, A.H., 2001. Consumer preferences for color, price, and vitamin C content of be ll peppers. HortScience 36, 795 800.

PAGE 172

172 Gerson, U., 1992. Biology and control of the broad mite, Polyphagotarsonemus latus (Banks) (Acari: Tarsonemidae). Exp. Appl. Acarol. 13, 163. Gillespie, D.R., Raworth, D.A., 2005. Biological control of two-spotted spider mites on greenhouse vegetable crops. In: Heinz K.M., Va n Driesche R.G., Parrella, M.P. (Eds.), Biocontrol in protected culture. Ball Pub., Batavia, Illinois, pp. 201. Gotoh T., Yamaguchi, K., Mori, K., 2004. Effect of the temperature on the life history of the predatory mite Amblyseius ( Neoseiulus ) californicus (Acari: Phytoseiidae). Exp. Appl. Acarol. 32, 15. Griffiths, D.A., 1999. Biological control of mite s. In: R. Albajes, M.L. Gullino, J.C. Van Lenteren, and Y. Elad (Eds.), Integrated pe st and disease management in greenhouse crops. Kluwer Academ. Pub., The Netherlands, pp. 217. Grinberg, M., Perl-Treves, R., Palevsky, E., S homer, I., Soroker, V., 2005. Interaction between cucumber plants and the broad mite, Polyphagotarsonemus latus : from damage to defense gene expression. Entomol. Exp. Appl. 115, 135. Gui, L., Gong, X., Meng, G., Gui, L.Y., G ong, X.W., Meng, G.L., 2001. On the relationship between eggplant leaf structure and its resi stance to broad mite. Acta Phytophylacica Sinica 28, 213. Gullino, M.L., Albajes, R., van Lenteren, J.C ., 1999. Setting the stage: characteristics of protected cultivation and tools for sustainabl e crop production. In: Albajes, R., Gullino, M.L., van Lenteren, J.C., Elad, Y. (Eds.), In tegrated pest and disease management in greenhouse crops. Kluwer Academ. P ub., The Netherlands, pp. 1. Hassan, S.A., Van de Veire, M., 2004. Compatibility of pesticidas with biological control agents. In: Heinz K.M., Van Driesche R.G., Parrella, M. P. (Eds.), Biocontrol in protected culture. Ball Pub., Batavia, Illinois, pp. 129. Ho, C., 1991. Life history of Polyphagotarsonemus latus (Banks) feeding on lemon, tea and pepper. J. Agric. Res. China, 40, 439. Hochmuth, R.C., Belibasis, E., 1991. Costs of production for greenhouse tomatoes in North Florida. Coop. Ext. Serv. Univ. of Florida, In st. Food Agr. Sci., Gainesville. Rpt. SVAREC 91-20. Hochmuth, G.J., 1997. Fertilizer management fo r greenhouse vegetables. In: G.J. Hochmuth (Ed.). Florida greenhouse vegetable produc tion handbook, vol. 3. Inst. Food Agr. Sci., Univ. of Florida, Gainesville. Ci rc. SP 48, Gainesville, Fla, pp. 13. Hodges, A.W., Satterthwaite, L.N., Haydu, J.J., 2001. Business analysis of ornamental plant nurseries in Florida, 1998. Econ. Info. Repor t EI 00-5r. Food and Resource Economics Dept., Institute of Food and Agricultural Scienc es, Univ. of Florida, Gainesville, Fla.

PAGE 173

173 Hussey, N.W., 1985. Peppers. In: Hussey, N.W., Scopes, N. (Eds.), Biological pest control: the glasshouse experience. Cornell Uni v. Press, Ithaca, N.Y., p. 189. Ibrahim, Y.B. and Low, T.H, 1998. Life cycle of the broad mite Polyphagotarsonemus latus (Banks) (Acari: Prostigmata; Tarsonemidae) a nd its innate prolifera ting capacity on chilli. Malays. Appl. Biol. 27, 51. Jeppson, L.R., Keifer, H.H, Baker, E.W., 1975. M ites injurious to econo mic plants. Univ. of Calif. Press, Los Angeles, Calif. Jones, V.P., Brown, R.D., 1983. Reproduc tive responses of the broad mite, Polyphagotarsonemus latus (Acari: Tarsonemidae), to constant temperature-humidity regimes. Ann. Entom. Soc. Amer. 76, 466. Jovicich, E., 2001. Hydroponic greenhouse pepper in Florida: Practices of plant trellising, population, transplant depth, soilless media, a nd irrigation. MS Thesis, Univ. of Florida, Gainesville, Fla. Jovicich, E., Cantliffe, D.J., 2001. Trellis system, plant density, and container type can improve fruit yield and plant growth of a gr eenhouse hydroponic crop. HortScience 36, 434. Jovicich, E., Cantliffe, D.J., 2004. Salts deposited on the lower stem of bell pepper contribute to a basal stem disorder in soilless, gr eenhouse-grown plants. HortScience 39, 36. Jovicich, E., Cantliffe, D.J., Hochmuth, G.J., 1999. Elephants Foot, a plant disorder in hydroponic greenhouse sweet pepper. Proc Fla. Soc. Hort. Sci. 112, 310. Jovicich, E., Cantliffe, D.J., Sargent, S.A., Osborne L.S., 2004a. Production of greenhouse-grown peppers in Florida. Florid a Coop. Ext. Electronic Data Info. Source (EDIS), Univ. of Florida, Gainesville. 28 July 2004. . Jovicich, E., Cantliffe, D.J., Stoffella, P.J., 2004b. Fruit yield and quality of greenhouse-grown bell pepper as influenced by density, contai ner, and trellis syst em. HortTechnology 14, 507. Jovicich, E., Cantliffe, D.J., Stoffella, P.J., Haman, D.Z., 2007. Bell pepper fruit yield and quality as influenced by solar radiation-based irrigation and c ontainer media in a passively ventilated greenhouse. HortScience 42 (in print). Jovicich, E., VanSickle, J.J., Cantliffe, D.J., Stoffella, P. J., 2005. Greenhouse-grown colored peppers: a profitable al ternative to vegetabl e production in Florid a? HortTechnology 15, 355. Jung, C., Croft, B.A., 2001. Aerial dispersal of phytoseiid mites (Acari: Phytoseiidae): estimating falling speed and dispersal distance of adult females. Oikos 94, 182. Kay, D.R., Edwards, W.M., 1994. Farm mana gement. 3rd ed., McGraw-Hill, New York.

PAGE 174

174 Koppert Biological Systems, 2006. Side Effect s Database. Koppert BV, the Netherlands. 10 Sept. 2006 . Kramer, C.Y., 1957. Extension of multiple range tests to group correlated adjusted means. Biometrics 13, 13. Laffi, F., 1982. Presenza di Polyphagotarsonemus latus (Banks) su semenzai di pepperone in Emilia-Romagna. Inf. Fitopatol. 11, 55. Lee, S.H., Park, C.G., Choi, K.M., 1992. Tars onemid mite: morphology, damage symptom, development and occurrence in Korea. Re s. Rept. R.D.A. (Crop Protection) 34, 55. Li, L., Li, Y., 1986. Studies on the fluctuation of the broad mite. Acta Entomol. Sinica 29, 41 48. Li, L., Li, Y., Bu, G., 1985. The effect of temperature and humidity on the growth and development of the broad mite, Polyphagotarsonemus latus Acta Entomol. Sinica 28, 181. Littell, R.C., Milliken, G.A., Stroup, W.W. a nd Wolfinger, R.D., 1996. SAS System for Mixed Models, SAS Institute, Inc., Cary, N. Carolina. Lopez, J., Shwedel, K., 2001. The Mexican gr eenhouse vegetable industry. Food Agribusiness Res. Issue 032-2001, Oct. 2001, Rabobank Intl., Utrecht, The Netherlands. Martnez, P.F., 1999. An overview of the sout hern European greenhouse industry. Proc. 21st Annu. Can. Greenhouse Conf., 6 Oct. 1999, Uni v. of Guelph, Guelph, Ont., Canada. Maynard, D.N., Olson, S.M. (Eds.), 2003. Ve getable production guide for Florida 2002. Inst. Food Agr. Sci., Univ. of Florida, Gainesville. SP 170. McMurtry, J.A., Badii, M.H., Johnson, H.G., 1984. The broad mite, Polyphagotarsonemus latus as a potential prey for phytoseiid m ites in California. BioControl 29, 83. McMurtry, J.A., Croft, B.A., 1997. Life-styles of phytoseiid mites and thei r roles in biological control. Annu. Rev. Entomol. 42, 291. Mirza, M., 2003. Greenhouse industr y profile 2003. Agriculture, food and rural development of Alberta. Government of Al berta, Canada. 25 July 2003. . Mitchell, J.M., Cantliffe, D.J., 2005. The Eff ects of the 2004 hurricane season on greenhouse vegetable production in Florida. Florida Coop. Ext. Electronic Data Info. Source (EDIS), Univ. of Florida, Gainesville. 5 Feb. 2007 . Mizobe, M., Tamura, I., 2004. Biologi cal control of the broad mite Polyphagotarsonemus latus (Banks) by Amblyseius cucumeris (Oudemans) on greenhouse sweet pepper. Proc. Assoc. Plant Prot. Kyushu 50, 62.

PAGE 175

175 Monteiro, A.A., 1994. Outlook on growing techni ques of greenhouse solanacea in mildwinter climates. Acta Hort. 366, 21. Morgan, L., Lennard, S., 2000. Hydroponic capsicum production, a comprehensive, practical and scientific guide to commerc ial capsicum production. Casper Pub., Narrabeen, Australia. NaLampang, S., VanSickle, J.J., Evans, E.A., 2003. An application of demand systems on selected vegetables and fruit in the U.S. Selected paper Amer. Agr. Econ. Assn. Annu. Mtg., 5 Aug. 2003, Montreal. Nuez, F., Gil, R., Costa, J., 1996. El cultivo de pimientos chiles y ajes. Mundi-Prensa, Madrid, Spain. Olson, S.M., Simonne, E.H., (Eds.) 2006. Ve getable production guide for Florida 2005. Vance Publishing, Lenexa, Kans. Olson, S.M., Simonne, E.H., Maynard, D.N., Hochmuth, G.J., Varina, C.S., Stall, W.M., Pernezny, K.L., Webb, S.E., Taylor, T.G ., Smith, S.A., 2005. Pepper Production in Florida. Hort. Dept. Doc. HS-732 Fla. Coop. Ext. Serv., Inst. of Food and Agric. Sci., Univ. of Florida, Gainesville, Fla. Osborne, L., Pea, J.E., Ridgway R.L., Kl assen W., 1998. Predacious mites for mite management on ornamentals in protected cu ltures. In: Ridgway, R.L., Hoffmann M.P., Inscoe M.N., Glenister C.S. (Eds.), Mass-rear ed natural enemies: application, regulation, and needs. Proc. Thomas Say Publications in Entomology, Entom. Soc. Amer., pp. 116 138. Osborne, L.S., Bolckmans, K., Landa, Z., Pea, J.E., 2005. Kinds of natural enemies. In: Heinz K.M., Van Driesche R.G., Parrella, M.P. (Eds .), Biocontrol in protec ted culture. Ball Pub., Batavia, Illinois, pp. 95. Palevsky, E., Reuveny, H., Okonis, O., Gerson, U ., 1999. Comparative beha vioural studies of larval and adult stages of phyt oseiids (Acari: Mesostigmata) Typlodromus athiasae and Neoseiulus californicus Exp. Appl. Acarol. 23, 467. Palevsky, E., Soroker, V., Weintraub, P., Mansour, F., Abo-Moch, F., Gerson, U., 2001. How species-specific is the phoretic relationship between the broad mite, Polyphagotarsonemus latus (Acari: Tarsonemidae), and its in sect hosts? Exp. A ppl. Acarol. 25, 217. Pea, J.E., 1990. Relationships of broad mite (Acar i: Tarsonemidae) density to lime damage. J. Econ. Entomol. 83, 2008. Pea, J.E., 1992. Predator-prey interactions between Typhlodromalus peregrinus and Polyphagotarsonemus latus : effects of alternative prey and other food resources. Fla. Entomol. 75, 241

PAGE 176

176 Pea, J.E., Bullock, R.C., 1994. Effects of feed ing of broad mite (A cari: Tarsonemidae) on vegetative plant growth. Fla. Entomol. 77, 180. Pea, J.E., Campbell, C.W., 2005. Broad mite. Flor ida Coop. Ext. Electroni c Data Info. Source (EDIS), Univ. of Florida, Gainesville. 28 Jan. 2007. . Pea, J.E., Osborne, L., 1996. Biological control of Polyphagotarsonemus latus (Acarina: Tarsonemidae) in greenhouse and field trials using introductions of predacious mites (Acarina: Phytoseiidae ). Entomophaga 41, 279. Pea, J.E., Osborne, L., Duncan, R.E., 1996. Potencia l of fungi as biologi cal control agents of Polyphagotarsonemus latus (Acari: Tarsonemidae). Entomophaga 41, 27. Polack, A., Mitidieri, M., 2005. Produccin de pimiento diferenciado protocolo preliminar de manejo integrado de plagas y enfermedades. EEA San Pedro, Ed. Instituto Nacional de Tecnologa Agropecuaria, Bu enos Aires, Argentina. Portree, J., 1996. Greenhouse vegetable production guide for commercial growers. Ministry of Agr., Fisheries and Food, Province of British Columbia, Ext. Systems Branch, Victoria, B.C., Canada. Ramakers, P.M.J., 2005. IPM program for sweet pe ppers. In: Heinz K.M., Van Driesche R.G., Parrella, M.P. (Eds.), Biocontrol in protecte d culture. Ball Pub., Ba tavia, Illinois, pp. 439 455. Resh, H.M., 1996. Hydroponic food production, 5th ed. W oodbridge Press, Sant a Barbara, Calif. Rylski, I. 1986. Pepper ( Capsicum ), p. 341. In: S.P. Monselise (Ed.). CRC handbook of fruit set and development. CRC Press, Boca Raton, Fla. Snchez, J.A., Alczar, A., Lacasa, A., Llamas, A ., Bielza, P., 2000. Integrated pest management strategies in sweet pepper plas tic houses in the sout heast of Spain. Inte grated control in protected crops, Mediterranean climate. Intl. Org. Biolog. Control of Noxious Animals and Plants, West Palaearctic Reg. Section, Bul. 23, 21. SAS Institute, 2001. SAS/STAT Release 8.02. Cary, NC, USA. Saxton, A.M., 1998. A macro for converting mean se paration output to lette r groupings in Proc Mixed. In: Proceedings of 23rd SAS Users Group International, SAS Institute, Cary, NC, pp. 1243. Shaw, N.L., Cantliffe, D.J., 2002. Brightly colo red pepper cultivars fo r greenhouse production in Florida. Proc. Fla. State Hort. Soc. 115, 236. Silva, E.A., Oliveira, J.V., Gondim Jr ., M.G.C., Menezes, D., 1998. Biologia de Polyphagotarsonemus latus (Banks) (Acari: Tarsonemid ae) em pimento. An. Soc. Entomol. Brasil 27, 223.

PAGE 177

177 Smith, F.F., 1933. The cyclamen mite and the broad mite and their control. U.S. Dept. Agric. Circ. 301. 13 p. Smith, J., Hewitt, T., Hochmuth, R., Hochmu th, G., 2003. Greenhouse production: Profitability and cash flow analysis. C itrus Veg. Mag., Aug. 2003, p. 20. Smith, S.A., Taylor, T.G., 2004. Green Peppers: Estimated production costs in the Southwest Florida area 2001. Food and Resource Econom ics Department, Univ. of Florida, Gainesville, Fla. 20 June 2004. . Soroker, V., Nelson, D.R., Bahar, O., Reneh, S ., Yablonski, S., Palevsky, E., 2003. Whitefly wax as a cue for phoresy in the broad mite, Polyphagotarsonemus latus (Banks) (Acari: Tarsonemidae). Chemoecology 13, 163. SPSS Science, 2002. SigmaPlot Release 8.02. Chicago, Illinois, USA. Steiner M.Y., Goodwin S., Wellham T.M., Barchi a I.M., Spohr L.J., 2003a. Biological studies of the Australian predatory mite Typhlodromalus lailae (Schicha) (Acari: Ph ytoseiidae). Aust. J. Entomol. 42, 131. Steiner M.Y., Goodwin S., Wellham T.M., Barc hia I.M., Spohr L.J., 2003b. Biological studies of the Australian predatory mite Typhlodromips montdorensis (Schicha) (Acari: Phytoseiidae), a potential biocontrol agent for western flower thrips, Frankliniella occidentalis (Pergande) (Thysanoptera: Thripi dae). Aust. J. Entomol. 42, 124. Stevens III, T.J., Kilmer, R.L., Glenn, S.J., 2000. An economic comparison of biological control strategies for whiteflies (Homoptera: Aleyr odidae) in greenhouse poinsettias. Biol. Microb. Control 93, 623. Trumble, J.T., Kolodny-Hirsh, D.M., Ting, I. P., 1993. Plant compensation for arthropod hervibory. Ann. Rev. 38, 93. Tyson, R., Hochmuth, B.C., Hochmuth, G.J., Lamb, E.M., Sweat, M.S., 2001. A decade of change in Floridas greenhouse vegetable i ndustry: 1991. Proc. Fla. State Hort. Soc. 114, 280. Tyson, R., Hochmuth, R., Lamb, E., McAvoy, E., Olczyk, T., Lamberts, M., 2004. Greenhouse vegetables in Florida's mild winter climate2004 Update. Acta Hort. 659, 37. U.S. Dept. of Agriculture, 1998. Greenhouse produced food crops by state: 1998, pp. 254, http://www.nass.usda.gov/census/census97/hor ticulture/table37.pdf, In: Census of horticultural specialties, 1998, Vol. 3, Spec. studi es, Census 1997. Natl. Agric. Stat. Serv., U.S. Dept. of Agr., Washington, DC. 7 June 2003. .

PAGE 178

178 U.S. Dept. of Agriculture, 2001. Sweet peppers: Saved by the bell. Econ. Res. Serv., USDA. Agricultural outlook (by G. Lucier and B.H. Lin), p. 12, Dec. 2001. U.S. Dept. of Agric., Washington, DC. 10 June 2003. . U.S. Dept. of Agriculture, 2002. Foreign agricultu ral service, U.S. trade imports, FAS online. U.S. Dept. of Agric., Washington, DC. 5 Nov. 2002. . U.S. Dept. of Agriculture, 2003a. Vegetable and melon trade, p. 15. In: Per capita consumption expected to rise in 2003. Vegetables and melons outl ook/VGS-297/June 20, Econ. Res. Serv., U.S. Dept. Agric., Washington, DC. 23 June 2003. . U.S. Dept. of Agriculture, 2003b. Per capita consum ption expected to rise in 2003. Vegetables and melons outlook/VGS-297/June 20, Econ. Res. Serv., U.S. Dept. Agric. Washington, DC. 23 June 2003. . U.S. Dept. of Agriculture, 2003c. Fresh vegetabl es and melon, p. 14, In: Per capita consumption expected to rise in 2003. Vegetables and melons outl ook/VGS-297/June 20, Econ. Res. Serv., U.S. Dept. Agric. Washington, DC. 23 June 2003. . U.S. Dept. of Agriculture, 2004. Foreign Agricult ural Services U.S. Trade Internet System (FASonline). U.S. Dept. of Agric., Washington, DC., 30 June 2004. . U.S. Dept. of Agriculture, 2007a. Florida Ve getables: Bell peppers. 2005. National Agric. Stat. Serv., U.S. Dept. of Ag ric., Washington, DC. 1 March 2007. U.S. Dept. of Agriculture, 2007b. The National Organic Program. U.S. Dept. of Agric., Washington, DC. 1 March 2007 . Vaissayre, M., 1986. Chemical control of the mite Polyphagotarsonemus latus (Banks) in cotton fields. Coton Fibres Trop. 41, 3144. Valero-Garca, L.M., 2003. La importancia de la gama en el comercio del pimiento. Surinver Coop. in FITECH 2003. Horticom News. 23 July 2003. . Van Driesche, R.G., Heinz, K.M., 2004. An overvie w of biological control in protected culture. In: Heinz K.M., Van Driesche R.G., Parrella, M. P. (Eds.), Biocontrol in protected culture. Ball Publishing, Batavia, Illinois, pp. 1. van Lenteren, J.C., 2006. IOBC Internet Book of Bi ological Control. (Ed.) J.C. van Lenteren. 5 Jan. 2007. .

PAGE 179

179 VanSickle, J.J., 2003. Marketing strategies fo r vegetable growers, p. 103. In: D.N. Maynard and S.M. Olson (Eds.). Vegetable production guide for Florida 2003. Vance Publishing, Lenexa, Kans. Walter, D.E., 1996. Living on leaves: mites, tome nta, and leaf domatia. Ann. Rev. Entomol. 41, 101. Walter, D.E., ODowd, D.J., 1992. Leaf morphology a nd predators: effect of leaf domatia on the abundance of predatory mites (Acari: P hytoseiidae). Environ. Entomol. 21, 478. Walzer, A., Blmel, S., Schausberger, P., 2001. P opulation dynamics of interacting predatory mites, Phytoseiulus persimilis and Neoseiulus californicus held on detached bean leaves. Exp. Appl. Acarol. 25, 731. Weintraub P.G., Kleitman, S., Mori, R., Shapir a, N., Palevsky, E., 2003. Control of the broad mite ( Polyphagotarsonemus latus (Banks)) on organic greenhouse sweet peppers ( Capsicum annuum L.) with the predatory mite, Neoseiulus cucumeris (Oudemans). Biol. Control 27, 300. Weintraub, P.G., Alchanatis, V., Palevsky, E ., 2004. Distribution of the predatory mite, Neoseiulus cucumeris in greenhouse pepper. Acta Hort. 659, 281. Winsberg, M.D., 1990. Florida weather. Univ. of Central Florida Press, Orlando, Fla. Wittwer, S.H., Castilla, N., 1995. Protected cult ivation of horticultural crops worldwide. HortTechnology 5, 6. Zhang Z.-Q., 2003. Mites of greenhouses, identifi cation, biology and control. CABI Pub. Oxon, UK.

PAGE 180

180 BIOGRAPHICAL SKETCH Elio Jovicich was born on August 23, 1969 in C rdoba, Argentina. In 1987, he graduated as a mechanical technician from the Renault Inst itute. He then went on to study agronomy at the Catholic University of Crdoba, where he graduate d as an engineer in ag ricultural sciences in 1993. While attending university, he spent a semester as an intern at the National Institute for Agricultural Technology (INTA) in Manfredi, Cr doba. After graduating, he worked for a year in a small private agricultural se rvice enterprise and in an organi c farm in Jess Mara, Crdoba. In 1994, he joined the horticultura l research team of IN TA in Pocito, San Juan, where he carried out research on irrigation management and earl y-season production of vegetable crops. In 1997, he was awarded a research and te aching assistantship at the Hortic ultural Sciences Department at the University of Florida to pursue a Master of Science. He studied practices for managing plant canopy and fertigation in peppers grown in soi lless media in the greenhouse. In 2001, he continued a doctoral program in horticultural science, researchi ng biological control as an alternative to pesticide use in greenhouse-grown peppers.