Potential of Florida Natural Enemies to Control the Invasive Species Raoiella indica (Acari

MISSING IMAGE

Material Information

Title:
Potential of Florida Natural Enemies to Control the Invasive Species Raoiella indica (Acari Tenuipalpidae)
Physical Description:
1 online resource (151 p.)
Language:
english
Creator:
Carrillo,Daniel
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Entomology and Nematology
Committee Chair:
Pena, Jorge E
Committee Members:
Hoy, Marjorie A
Frank, J. Howard
Crane, Jonathan H

Subjects

Subjects / Keywords:
amblyseius -- arecaceae -- biological -- cocos -- control -- indica -- invasive -- largoensis -- mites -- neotropics -- palm -- phytoseiidae -- predatory -- raoiella -- red -- tenuipalpidae
Entomology and Nematology -- Dissertations, Academic -- UF
Genre:
Entomology and Nematology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The central issue of this dissertation is the red palm mite, Raoiella indica, a phytophagous tenuipalpid mite that recently invaded the Western Hemisphere. Efforts have been made to identify potential biological control agents of R. indica. A predatory mite Amblyseius largoensis (Acari: Phytoseiidae) has been found associated with R. indica in Florida. A series of studies evaluated A. largoensis for potential control of R. indica by determining (1) the development and reproduction of A. largoensis feeding on R. indica, other potential prey and pollen; (2) the prey-stage preferences and functional and numerical responses of A. largoensis to R. indica (3) the variability in response of four populations of A. largoensis with contrasting feeding histories to R. indica and Tetranychus gloveri Banks (Acari: Tetranychidae); (4) the effect of A. largoensis on R. indica using predator exclusion and predator release techniques. The results of these studies showed that: (1) Predators fed on a diet of R. indica had a faster development and a greater intrinsic rate increase than those fed on the other prey tested or pollen; (2) A. largoensis had a marked preference for R. indica eggs, followed by larvae, over other developmental stages. Predation and oviposition of A. largoensis increased as a function of increasing R. indica population densities. Amblyseius largoensis exhibited a type II functional response; (3) all predators, including those with no previous exposure to R. indica, had a high likelihood of consuming R. indica eggs over eggs and larvae of T. gloveri. However, populations of A. largoensis varied in their consumption of R. indica motile stages. Predators with previous exposure to the invasive species were more likely than na?ve predators to consume R. indica larvae and finally; (4) a predator release rate of 1 A. largoensis/ 10 R. indica reduced R. indica densities by 80% over the untreated control and contributed to reduce damage on coconut leaves caused by the phytophagous mite. Overall, results of this research suggest that A. largoensis is an important mortality factor of R. indica and should be considered as a key biological control agent in integrated pest management programs targeting R. indica.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Daniel Carrillo.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Pena, Jorge E.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043316:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 POTENTIAL OF FLORIDA NATURAL EN EMIES TO CONTROL THE INVASIVE SPECIES RAOIELLA INDICA (ACARI: TENUIPALPIDAE) By DANIEL CARRILLO 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 2011

PAGE 2

2 2011 Daniel Carrillo

PAGE 3

3 To my family and friends, whose patience a nd encouragement have he lped me throughout the process of writing this dissertation

PAGE 4

4 ACKNOWLEDGMENTS I owe thanks to many people whose assistance was indispensable in completing this work. First among these is Dr. Jorge E. Pea, advisor an d chair, whose trust and generosity motivated me throughout the process. I am thankful for his professional advice, scientific guidance, mentoring, financial support and frienship. I also thank the other members of my committee Dr. Marjorie A. Hoy, Dr. J. Howard Frank and Dr. Jonathan H. Crane for their continued support, scientific guidance, and for their multiple reviews to this dissertation and the articles derived from it. Additional recognition goes to Dr. Martha de Coss and Dr. Josep Jacas and who worked with me conducting some of the studies. Dr. Jacas played an important ro le in this dissertation providing advice and multiple reviews to manuscripts. I acknowledge Dr. Nalapang Sikavas and Dr. James Colee (University of Florida-Statistics Department) for providi ng statistical advice. I thank Drs. James A. McMurtry, Eric Palevsky, Ron Ochoa and Paul Kendra for helpful reviews to the manuscripts presented here. I am also thankful to Dr. Cal Welbourn, Dr. Raymond J. Gagn and Dr. Lionel A. Stange for the identifi cation of mite, midges, and lacewing specimens, respectively. I thank Dr. Er ic Riddick for providing Stethorus punctillum, and Drs. Amy Roda, Divina Amalin, Denise Navia, Jose Carlos Rodrigues, Ron Oc hoa and Francisco Ferragut for their cooperation. A special recognition goes for Rita E. Duncan, Katia Santos and David Long who were good friends and excellent coworkers throughout the project. I acknowledge the help received from the Faculty and Staff of the Tr opical Research and Edu cation Center, Homestead, especially from Ana Vargas, Jose Alegria, Manny Soto and Holly Glenn. I am grateful to the Fairchild Tropical Botanic Garden and the Royal Botanical Garden for providing plant material and access to their collections; the University of Florida. I thank Y. Ali, L. Bradshaw, R. Gonzales, K. Harradan, M. Kent, S. Maraj, P. Perez, N. Raj, and P. Siew, for their help. Finally, I acknowledge my family for their constant en couragement and my partner Paula for her

PAGE 5

5 dedication, patience and uncondition al support. This research was partially funded by the United States Depertment of Agriculture through a Tropi cal Subtropical Agriculture Research grant to Jorge E. Pea.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4 LIST OF TABLES ...........................................................................................................................9 LIST OF FIGURES .......................................................................................................................10 ABSTRACT ...................................................................................................................... .............12 CHAPTER 1 INTRODUCTORY REMARKS ............................................................................................14 Research Objectives ........................................................................................................... .....18 Structure of the Dissertation ...................................................................................................18 Literature Review ...................................................................................................................20 Taxonomy of Raoiella indica ..........................................................................................20 Biology of R. indica ........................................................................................................21 Mite and Host Plant Interactions ..................................................................................... 23 Influence of Abiotic Factors on R. indica .......................................................................25 2 A REVIEW OF THE NATURAL ENEM IES OF THE RED PALM MITE, RAOIELLA INDICA (ACARI: TENUIPALPIDAE) ................................................................................. 28 Introduction .................................................................................................................. ...........28 Natural Enemies of R. indica ..................................................................................................29 Predatory Mites ............................................................................................................... 29 Predatory Insects ............................................................................................................. 33 Pathogens ..................................................................................................................... ....37 Discussion .................................................................................................................... ...........38 3 DEVELOPMENT AND REPRODUCTION OF AMBLYSEIUS LARGOENSIS (ACARI: PHYTOSEIIDAE) FEEDING ON POLLEN, RA OIELLA INDICA (ACARI: TENUIPALPIDAE), AND OTHER MICROARTHROPODS INHABITING COCONUTS IN FLORIDA ...................................................................................................50 Introduction .................................................................................................................. ...........51 Materials and Methods ...........................................................................................................53 Source of A. largoensis R. indica Tetranychus gloveri Aonidiella orientalis Nipaecoccus nipae and Quercus virginiana Pollen. ....................................................53 Stock colony of A. largoensis ..................................................................................53 Stock colony of R. indica .........................................................................................54 Stock colony of T. gloveri ........................................................................................54 Stock colonies of A. orientalis and N. nipae ............................................................55 Source of pollen .......................................................................................................55

PAGE 7

7 Feeding Experiments with A. largoensis .........................................................................55 Experimental arenas ................................................................................................. 55 Experimental setup ...................................................................................................56 Statistical Analysis .......................................................................................................... 57 Results .....................................................................................................................................57 Discussion .................................................................................................................... ...........59 4 PREY-STAGE PREFERENCES AND FUNCTIONAL AND NUMERICAL RESPONSE S OF AMBLYSEIUS LARGOENSIS (ACARI: PHYTOSEIIDAE) TO RAOIELLA INDICA (ACARI: TENUIPALPIDAE) .............................................................. 66 Introduction .................................................................................................................. ...........67 Materials and Methods ...........................................................................................................69 Rearing and General Expe rimental Procedures ............................................................... 69 Raoiella indica Stage Preference by A. largoensis .........................................................70 Functional and Numerical Responses of Amblyseius largoensis to Varying Densities of R. indica Eggs .......................................................................................... 71 Results .....................................................................................................................................72 R. indica Stage Preference by A. largoensis ....................................................................72 Functional and Numerical Responses of A.largoensis to Varying Densities of R. indica Eggs...................................................................................................................73 Discussion .................................................................................................................... ...........74 5 VARIABILITY IN RESPONSE OF FOUR POPULATIONS OF AMBLYSEIUS LARGOENS IS (ACARI: PHYTOSEIIDAE) TO RAOIELLA INDICA (ACARI: TENUIPALPIDAE) AND TETRANYCHUS GLOVERI (ACARI: TETRANYCHIDAE) EGGS AND LARVA ............................................................................................................. 84 Introduction .................................................................................................................. ...........85 Materials and Methods ...........................................................................................................87 General Experimental Procedures ................................................................................... 87 Prey Mites ........................................................................................................................87 Predatory Mites ............................................................................................................... 88 No-choice Tests ...............................................................................................................89 Choice Tests ....................................................................................................................89 Data Analysis ...................................................................................................................90 Results .....................................................................................................................................91 No-choice Tests ...............................................................................................................91 Choice Tests ....................................................................................................................91 Discussion .................................................................................................................... ...........92 6 EFFECT OF AMBLYSEIUS LARGOENSIS (AC ARI: PHYTOSEIIDAE) ON RAOIELLA INDICA (ACARI: TENUIPALPIDAE) USING PREDATOR EXCLUSION AND PREDATOR RELE ASE TECHNIQUES ................................................................... 100 Introduction .................................................................................................................. .........100 Materials and Methods .........................................................................................................101

PAGE 8

8 Results ...................................................................................................................................103 Discussion .................................................................................................................... .........105 7 CONCLUSIONS .................................................................................................................. 111 APPENDIX: HOST PLANT RANGE OF R AOIELLA INDICA HIRST (ACARI: TENUIPALPIDAE) IN AREAS OF INVASION OF THE NEW WORLD ....................... 115 Introduction .................................................................................................................. .........116 Material and Methods ...........................................................................................................117 Fairchild Tropical Botanic Garden Survey, Miami, Florida ......................................... 117 Royal Botanical Gardens Survey, Port of Spain, Trinidad and Tobago ........................ 118 Evaluation of 3 Florida Native Pa lms as Reproductive Host of R. indica ....................119 Evaluation of Two Dicotyledon Plant Species, Phaseolus vulgaris and Ocimum basilicum, as Hosts of R. indica .................................................................................120 Results ...................................................................................................................................121 Fairchild Tropical Botanic Garden Survey, Miami, Florida ......................................... 121 Royal Botanical Garden Surve y, Port of Spain, Trinidad .............................................122 Evaluation of 3 Florida Native Pa lms as Reproductive Host of R. indica ....................123 Evaluation of Two Dicotyledon Plant Species, P. vulgaris and O. basilicum, as Hosts of R. indica .......................................................................................................123 Updated List of Reported Host Plants of R. indica .......................................................124 Discussion .................................................................................................................... .........125 LIST OF REFERENCES .............................................................................................................138 BIOGRAPHICAL SKETCH .......................................................................................................151

PAGE 9

9 LIST OF TABLES Table page 1-1 Species of the genus Raoiella ............................................................................................22 1-2 Developmental times of immature stages of Raoiella indica females ...............................23 2-1 Predatory arthropods repo rted in association with R. indica. ............................................44 3-1 Duration and survivorship of immature stages of Amblyseius largoensis fed on five diets and a no-food control. ............................................................................................... 63 3-2 Fertility life table parameters and other reproductive parameters of A. largoensis fed on three diets. .....................................................................................................................64 4-1 Parameters of the f unctional response of A. largoensis feeding on R. indica eggs ...........80 5-1 Predation of R. indica and Tetranychus gloveri eggs and larvae by four local populations of A. largoensis with disparate feeding history .............................................. 97 6-1 Initial mean R. indica infestations and A. largoensis release rates on coconut palms and mean R. indica infestations and number of A. largoensis recovered three months after predator release ........................................................................................................108 A-1 Mean densities of R. indica and phytoseiid mites found on various palms in the Fairchild Botanical Garden, Miami, Florida .................................................................... 130 A-2 Mean number of R. indica and phytoseiid mites found on palms located at the Royal Botanical Gardens, Trinidad. ...........................................................................................131 A-3 Average counts of R. indica and phytoseiid mites on basil, bean and coconut six weeks after being infested with 50 R. indica females ...................................................... 131 A-4 Reported host plant species of R. indica .........................................................................132

PAGE 10

10 LIST OF FIGURES Figure page 1-1 Reports of the presence of Raoiella indica ....................................................................... 16 1-2 Raoiella indica on coconut. ...............................................................................................26 1-3 Raoiella indica forming multigenerational colonies on the abaxial surface of coconut leaflets (pinnae). .................................................................................................................27 2-1 Amblyseius largoensis feeding upon R. indica and ovipositing on infested coconut leaves in Florida. ............................................................................................................ ....46 2-2 Life cycle of A. largoensis feeding upon R. indica in Florida ........................................... 47 2-3 Bdella distincta found feeding upon R. indica in Florida ................................................48 2-4 Ceraeochrysa claveri feeding upon R. indica and ovipositing on infested coconut leaves in Florida. ............................................................................................................ ....48 2-5 Arthrocnodax sp. observed feeding on R. indica in Florida. .............................................49 2-6 Aleurodothrips fasciapennis observed feeding upon R. indica in Florida.. ....................... 49 3-1 Daily oviposition rate of Amblyseius largoensis females fed on either R. indica, Tetranychus gloveri or Quercus virginiana pollen. ........................................................... 65 3-2 Survivorship of A. largoensis females fed on either R. indica T. gloveri or Q. virginiana pollen. ...............................................................................................................65 4-1 Daily consumption and oviposition by A. largoensis when offered R. indica stages in a no-choice condition. ........................................................................................................80 4-2 Preference of A. largoensis females for R. indica stages. .................................................. 81 4-3 Functional response of A. largoensis to increasing densities of R. indica eggs estimated through three curve fitting models. .................................................................... 82 4-4 Daily oviposition of A. largoensis females as a function of the number of R. indica eggs consumed/day. ........................................................................................................... 83 5-1 Percentage of individuals of R. indica and T. gloveri consumed by four populations of A. largoensis with disparate feeding histor ies (previous di et) under choice conditions. ................................................................................................................... .......98 5-2 Prey preference of four local populations of A. largoensis with disparate feeding history (previous diet) when preying on R. indica and T. gloveri eggs and larvae under choice conditions. ....................................................................................................99

PAGE 11

11 6-1 Effect of four release rates of A. largoensis (0= control, 1:10, 1:20, 1:30 A. largoensis : R. indica ) on R. indica motile stage densities. .............................................. 109 6-2 Effect of four release rates of A. largoensis (0= control, 1:10, 1:20, 1:30 A. largoensis/R. indica ) on R. indica densities. .................................................................... 110 A-1 Monthly average counts of R. indica on the 3 native palms and coconut palm after being artificially infested .................................................................................................136 A-2 Average R. indica count on 3 native palms and coconut palms in Miami-Dade County. .............................................................................................................................137

PAGE 12

12 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 POTENTIAL OF FLORIDA NATURAL EN EMIES TO CONTROL THE INVASIVE SPECIES RAOIELLA INDICA (ACARI: TENUIPALPIDAE) By Daniel Carrillo August 2011 Chair: Jorge E. Pea Major: Entomology and Nematology The central issue of this disse rtation is the red palm mite, Raoiella indica a phytophagous tenuipalpid mite that recently invaded the We stern Hemisphere. Efforts have been made to identify potential biological control agents of R. indica. A predatory mite Amblyseius largoensis (Acari: Phytoseiidae) has b een found associated with R. indica in Florida. A series of studies evaluated A. largoensis for potential control of R. indica by determining (1) the development and reproduction of A. largoensis feeding on R. indica, other potential prey and pollen; (2) the preystage preferences and functiona l and numerical responses of A. largoensis to R. indica (3) the variability in response of four populations of A. largoensis with contrasting feeding histories to R. indica and Tetranychus gloveri Banks (Acari: Tetranychi dae); (4) the effect of A. largoensis on R. indica using predator exclusion and predator release techni ques. The results of these studies showed that: (1) Predators fed on a diet of R. indica had a faster development and a greater intrinsic rate increase than those fe d on the other prey tested or pollen; (2) A. largoensis had a marked preference for R. indica eggs, followed by larvae, over other developmental stages. Predation and oviposition of A. largoensis increased as a function of increasing R. indica population densities. Amblyseius largoensis exhibited a type II func tional response; (3) all predators, including those with no previous exposure to R. indica had a high likelihood of

PAGE 13

13 consuming R. indica eggs over eggs and larvae of T. gloveri However, populations of A. largoensis varied in their consumption of R. indica motile stages. Predators with previous exposure to the invasive speci es were more likely than nave predators to consume R. indica larvae and finally; (4) a pr edator release rate of 1 A. largoensis / 10 R. indica reduced R. indica densities by 80% over the untreated control and contributed to re duce damage on coconut leaves caused by the phytophagous mite. Overall, re sults of this research suggest that A. largoensis is an important mortality factor of R. indica and should be considered as a key biological control agent in integrated pest management programs targeting R. indica.

PAGE 14

14 CHAPTER 1 INTRODUCTORY REMARKS The introduction of non-indigenous species to an ecosystem sometimes results in serious harm to the environment. In such cases, immigr ant species are referred to as invasive species. Invasive species can cause serious alterations to plant and animal communities resulting in high economic losses, calculated in several billions of dollars in the USA (Pim entel et al. 2005) and trillions worldwide (Ricciardi et al. 2011). During the last few centuries, the opportunities for unintentional introductions of organisms (i.e., an imals, microbes, plants) to new areas have greatly increases due to an expansion in tr ade among nations, human movement and population growth (Elton 1958). Therefore, extensive e fforts have been made to prevent further introductions and to manage the species that have al ready established in various ecosystems. This dissertation focuses on a phytophagous mite species that has several traits shared by most invasive species. In ge neral, phytophagous mites are rstrategists, or species with small size, rapid development and high reproductive rates that enables a high instantaneous rate of increase. Most phytophagous mites can reproduce pa rthenogenetically; theref ore, a single female has the ability to start a new population. Mites ar e difficult to detect because they are often hidden on their host plants and the symptoms of mite infestation usually a ppear after populations are large and unmanageable. Moreover, severa l dispersal mechanisms allow phytophagous mites to colonize distant plants, incl uding aerial dispersal through sp ecific behaviors (silking or ballooning) or as aerial plankton, and phoresy (attaching themse lves to insects or other hosts). All these traits increase the likelihood of establishment into new areas and the probability they will become invasive species. Documented examples of phytophagous mites as invasive species are scarce compared to invasive insects or plants. However, many mite species ha ve been moved around the world

PAGE 15

15 inadvertently (Navia et al. 2010). The lack of mite taxonomic and biogeographical studies, and lack of experts in those areas, may have contribut ed to the failure of detecting and mitigating the effects of those invasions in a timely manne r. Perhaps the best documented case was the introduction and spread of the cassava green mite Mononychellus tanajoa (Bondar) (Acari: Tetranychidae) into Africa (Yan inek and Herren 1988). This mite of Neotropical origin, was discovered on cassava in Uganda in the 1970s and spread to othe r African countries thereafter causing significant damage to this important food crop. Several researchers, agencies and institutions made extensive efforts to identify mortality factors and practices for managing this species [Bellotti et al. 1999, Centro Internacional de Agricultura Tropical (CIAT-Colombia), International Institute of Biological Control (IIB C UK-Benin and Kenya), International Institute of Tropical Agriculture (IITA-Ni geria), Empresa Brasileira de Pesquisa Agropecuaria (EMBRAPA-Brazil)]. The central issue of this disse rtation is the red palm mite, Raoiella indica a tenuipalpid mite that recently invaded the Western Hemisphere. Raoiella indica is a polyphagous species with a wide host-plant range, mo stly within the Arecaceae (palms), but it also attacks some plants within the Pandanaceae, Musaceae, He liconiaceae, Zingibera ceae and Strelitziaceae (Appendix). Its major host is coconut Cocos nucifera L. Raoiella indica is highly destructive to coconut sometimes reaching 4,000 individuals/ le aflet (Pea et al. 2009) (Figure 1-2D). Infestations found on other host plants are lower in numbers but still problematic because of their importance as ornamental or native plants. Before its arrival in the New World R. indica was found only in tropical and subtropical areas of Africa and Asia (Figure 1-1). In 2004, this mite was detect ed in Martinique (Flechtmann and Etienne 2004) and subsequently spread to mu ltiple islands of the Caribbean. In 2007, it was

PAGE 16

16 found in West Palm Beach, Florida (USA), and in the state of Sucre, Venezuela. More recently, the mite was reported in the northern state of Rora ima in Brazil, in the north coast of Colombia and in Isla Mujeres and Cancun, Mexico (Figure 1-1). Figure 1-1. Reports of the presence of Raoiella indica 1. Native range in the Eastern Hemisphere [Egypt (Zayed 1942), India (P uttarudriah and ChannaBasavanna 1956), Mauritius (Moutia 1958), Pakistan (Chaudri 1974), Israel (Gerson et al. 1983), Oman (Elwan 2000), Iran (Askari et al. 2002), Phill ipines (Gallego et al. 2003), and Benin and Tanzania (Zannou et al. 2010)]. 2.Areas of recent invasion in the New World [Martinique (Flechtmann a nd Etienne 2004), St. Lucia and Dominica (Kane et al. 2005), Guadaloupe and Saint Martin (E tienne and Flechtmann 2006), Jamaica (RADA 2007), Puerto Rico and Culebra Is land (Rodrigues et al. 2007), Florida USA (FDACS 2007), Trinidad and Tobago (Roda et al. 2008), Cuba (de la Torre et al. 2010), Venezuela (Vsquez et al. 2008), Brazi l (Marsaro Jr. et al. 2009), Mexico (NAPPO 2009 and Colombia (Carrillo et al. 2011b)]. The establishment of R. indica in the Caribbean has caused serious economic harm to coconut production with over 50% yield reductions at some locatio ns (CARDI 2010). In Florida, infestations have spread to si x counties in the southern part of the peninsula (FDACS 2011). Quarantine-mandatory acaricide sprays before shipping R. indica hosts have been adopted in order to prevent R. indica s rapid dissemination (TDA 2008, FDAC S 2011). However, this mite

PAGE 17

17 continues to spread throughout the Neotropical region wh ere a great diversity of plants could potentially be affected (Appendix). The host-plant range and dispersal of R. indica throughout natural, agricultural, recreational and residential areas suggest that la rge-scale mitigation programs are required for managing this species. Chemical control, host-pl ant resistance and cultural-control tactics could be used to manage local populations; however, the most promising approach is to find a practical, long-lasting solution in the form of biological control. One of the factors contributing to the aggressiveness of invasive species is that they often arrive in new areas without their specific natural enemies (He rren and Neuenschwander 1991). Thus, one of the most common approaches to suppress invasive speci es is to search for natural enem ies in their site of origin in order to reunite pest and natural enemies thr ough importation of the latter (classical biological control) (Van Driesche and Bellows 1996). In many cases, indigenous natural enemies may provide some suppression of the invasive pest. In addition, biological control might be achieved with species that have not experienced close prior relationship with the target organism (Hokanen and Pimentel 1984). Raoiella indica became the target of classical biologica l control after its arrival in the New World (Hoy 2011; Taylor et al. 2011). In parallel, efforts were made to identify the natural enemy fauna in areas of invasion in the Wester n Hemisphere (Pea et al. 2009). Interestingly, surveys for natural enemies conducted in India, Ma uritius, and those made in the New World all found Amblyseius largoensis Muma (Acari: Phytoseiidae) as the most abundant predator of R. indica Bowman (2010) hypothesized that, as seen in other phytoseiids (Noronha and Moraes 2004), biotypes or cryptic speci es could exist within the ta xonomic group classified as A. largoensis However, the existence of A. largoensis in Florida would likely impede the release of

PAGE 18

18 predators under the same taxonomic identity unless it is proven that they w ill not interbreed with or have deleterious effects on the local populations (Hoy, pers. comm.). The actual lack of other ca ndidates for classical biologi cal control and the observed association of vari ous predators with R. indica (a complete review of the natural enemies of R.indica is presented in chapter 2) suggest that Florida native natural enemies merit a more thorough investigation. Moreover, the conspicuous association between A. largoensis and R. indica in Florida suggested the need for detailed studies to better understa nd their predator-prey interactions. These issues are the main focus of this dissertation. Research Objectives The aims of this dissertation were to identify the natural enemy complex of R. indica in Florida and to conduct analyses of th e predator-prey inte ractions between A. largoensis and R. indica Structure of the Dissertation This dissertation is presented as five separate papers (Chapt ers 2-6) in publication styles ready for submission to scientific journals. Ea ch paper is therefore a stand-alone document, addressing different aspects of biological control of R. indica Preceding these articles, Chapter 1 provides an introduction to the research problem and research objectives, and a brief literature review on the taxonomy and biology of R. indica including information on interactions with its major host. Following the articles, Chapter 7 pres ents the main conclusi ons of the dissertation and explores future directions for re search focused on biological control of R. indica In addition, an Appendix entitled Host plant range of Raoiella indica Hirst (Acari: Tenuipalpidae) in areas of invasion of the New World was included to complement the information on the biology and potential effects of R. indica in the Neotropics. This article was submitted for publication in the journal Experimental and Applied Acarology as the result of a coll aborative project between

PAGE 19

19 researchers from the Ministry of Agriculture, Land and Marine Resources of Trinidad and Tobago, USDA-APHIS, and the University of Florida TREC. Divina Am alin, Farzan Hosein, Amy Roda, Rita Duncan and Jo rge E. Pea are coauthors. The first article (Chapter 2) entitled A review of the natural enemies of the red palm mite, Raoiella indica (Acari: Tenuipalpidae) presents a compendium of all the available information about the natural enemies reporte d in association with the red palm mite both in the Eastern Hemisphere and in areas of recent invasion in the New World. This review outlines which natural enemies are more promising candidates for classical biological control programs of R. indica according to the available literature, a nd which have shown more potential among the natural enemies found in areas of recent invasion by R. indica, and thus, were selected for experimental evaluations in Florida. This article was submitted to Experimental and Applied Acarology. Drs. J. Howard Frank, Jose Carlos V. Rodrigues and Jo rge E. Pea are coauthors. The second article (Chapter 3) entitl ed Development and reproduction of Amblyseius largoensis (Acari: Phytoseiidae) feeding on pollen, Raoiella indica (Acari: Tenuipalpidae), and other micro-arthropods inhabiting Coconuts in Fl orida presents a study on the basic biology of A. largoensis The study was designed to determine how feeding on R. indica could affect the life-table parameters of A. largoensis when compared to feeding on other potential prey with a longer history of inhabiting coconuts in Florida. This article was published in Experimental and Applied Acarology (2010, 52: 119-129). Drs. Jorge E. Pea, Marjorie A. Hoy and J. Howard Frank are coauthors. The third article (Chapter 4) entitled Prey-stage preference s and functional and numerical responses of Amblyseius largoensis (Acari: Phytoseiidae) to Raoiella indica (Acari: Tenuipalpidae) presents a series of experiment s designed to determine whether the predator

PAGE 20

20 showed differences in consumption or preferences among R. indica stages, and how the predator could respond to changes in the population density of R. indica in terms of its predation and reproductive rates. This article was submitted to Experimental and Applied Acarology Dr. Jorge E. Pea is coauthor. The fourth article (Chapter 5) entitled Variabili ty in response of four populations of Amblyseius largoensis (Acari: Phytoseiidae) to Raoiella indica (Acari: Tenuipalpidae) and Tetranychus gloveri (Acari: Tetranychidae) eggs and larvae presents a study designed to determine how populations of predators with disparate previous exposure to R. indica could differ in their like lihood of feeding on R. indica This article was submitted to Biological Control Drs. Martha E. de Coss, Marjorie A. Hoy and Jorge E. Pea are coauthors. The fifth article (Chapter 6) entitled Effect of Amblyseius largoensis (Acari: Phytoseiidae) on Raoiella indica (Acari: Tenuipalpidae) us ing predator exclusion and predator release techniques presents quantitative evaluations of the efficacy of A. largoensis The target journal fo r this article has not been defined yet. Drs. Josep A. Jaca s and Jorge E. Pea are coauthors. Literature Review This section presents a brief revi ew on the taxonomy and biology of R. indica including information on mite-plant interactions with its major host plant, C. nucifera Taxonomy of Raoiella indica The superfamily Tetranychoidea groups severa l families of obligate plant-feeding mites (Krantz 2009). The Tetranychid ae (spider mites) are the most recognized and studied mites within the superfamily becau se this family includes many species considered as important pests around the world. The Tenuipalpidae, commonly known as flat mites or false spider mites, also has a worldwide distribution with several species considered as important pests. Most of the tenuipalpid species that cause economic damage to cu ltivated plants (by direct feeding or

PAGE 21

21 vectoring plant viruses, Rodrigues et al 2003) belong to the two largest genera, Brevipalpus Donnadieu and Tenuipalpus Donnadieu (Mesa et al. 2009). However, the genus Raoiella has gained importance after R. indica started to spread aggressive ly in the Neotropical region. The genus Raoiella was first described and placed in the Tetranychidae by Hirst (1924), but later transferred to the Tenuipalpidae by Sayed (1950). Tw elve species described from the Indomalayan (8), Afrotropical (2), Australasian (1) and Western Palearctic (1) zoogeographic regions compose the genus (Table 1-1); however, seven are suspected to be junior synonyms of the type species R. indica (Mesa et al. 2009). Moreover, the genera Rarosiella Rimando 1996 and Neoraoiella Mohanasundaram 1996 were synonymized with Raoiella Hirst 1924. Biology of Raoiella indica Raoiella indica is usually found forming large multigenerational colonies on the abaxial surface of the leaflets (pinnae) of their host plants (Figure 1-2C). Eggs are oblong, smooth, glossy, red in color, and measure 0.12 mm in length and 0.09 mm in breadth, with a stipe 0.15 long (Moutia 1958) (Figures 1-2A and 1-3). A single droplet of an unknown substance is sometimes found at the tip of the stipe. Upon ha tching, larvae are slow in their movements and start feeding on leaf tissue near the site of emergence. Development of R. indica passes through two nymphal stages with a quiescent period before each molt. The immature stages of R. indica are red with blackish marks in the dorsum and de velop near the site of oviposition. The duration of the immature development varies from a pproximately 22 to 34 d depending on the host plant and environmental conditions (Table 1-2) (Figure 1-3). Males are triangular in form, de velop faster and are smaller than females. They actively seek out females and settle closely behind a quiescent deutonymph or adult female in a guarding state until mating (Figure 1-2B). Th e preoviposition period ranges from 3 to 7 d

PAGE 22

22 (Moutia 1958, Flores et al. 2010) The type of reproduction in R. indica is unusual because mated females (sexual reproduction) produce progeny cons isting solely of females and unfertilized Table 1-1. Species of the genus Raoiella following the classificati on by Mesa et al. (2009). Species name and descriptor Country of collection Host plant australica Womersley Australia Eucalyptus spp. (Myrtaceae) eugenia (Mohanasundaram 1996) India Eugenia sp. (Myrtaceae) shimapana Meyer South Africa Cassine transvaalensis (Celastraceae) macfarlanei Pritchard & Baker Greece olive (Oleaceae) indica Hirst India Cocos nucifera (Arecaceae) camur Chaudhri & Akbar Pakistan Phoenix dactylifera (Arecaceae) empedos Chaudhri & Akbar Pakistan Phoenix dactylifera (Arecaceae) neotericus Chaudhri & Akbar Pakistan Phoenix dactylifera (Arecaceae) obelias Hasan & Akbar Pakistan Phoenix dactylifera (Arecaceae) rahii Akbar & Chaudhri Pakistan Phoenix dactylifera (Arecaceae) phoenica Meyer Sudan Phoenix dactylifera (Arecaceae) pandanae Mohanasundaram India Pandanus sp. (Pandanaceae) *Species suspected to be junior synonyms of the type species R. indica. females (asexual reproduction) produce only ma les (NageshaChandra and Channabasavanna 1984). The species is reported to have four and two chromosomes (females and males, respectively) which suggests that its reproduction is ar rhenotokous (Helle et al. 1980). According to NageshaChandra and Channabasavanna (1984) mated females produce at an average of 22 4.41 (females/female) and unmated females produ ce 18.4 1.95 (males/femal e) during their adult life on coconuts. Moutia (1958) re ported a similar average fec undity (~ 28.1 eggs/female) but other studies with other hosts reported substa ntially lower fecundities (12.6 .3 and 7.8 2.6 eggs/female on coconut and banana leaves, re spectively (Gonzales and Ramos 2010), and 7.0 3.5 eggs per female on Areca palms (Flores-Gala no et al. 2010). According to NageshaChandra and Channabasavanna (1984), longev ity is greater in females (50.9 11.4 days) than in males (21.6 1.95 days), but this can also be influenced by the host plant (Gonzales and Ramos 2010).

PAGE 23

23 Adults are more active and are responsible for disp ersion within the leaf, to colonize new leaves or new plants. Table 1-2. Developmental times of immature stages of R. indica females reported by various studies using variations of the leaf fl otation method under laboratory conditions. Host Plant Temp.( C) Rel. Hum. (%) Develop. time (d) Reference Cocos nucifera 24.2 22 Moutia 1958 Cocos nucifera 17.9 33 Moutia 1958 Phoenix dactylifera 26.1 57.90 21.4 Zaher et al. 1969 Cocos nucifera 23.9-25.7 59.85 24.5 1.92 NageshaChandra and Channabasavanna 1984 Cocos nucifera 26.3 1.26 74.8 4.3 31.4 3.31 Gonzales and Ramos 2010 Musa acuminata 26.3 1.26 74.8 4.3 33.4 4.76 Gonzales and Ramos 2010 Areca catechu 25.4 1.20 57.5 6.5 31.0 4.11 Flores-Galeano et al. 2010 *information not provided Mite and Host Plant Interactions Host plants could have an important influence in the population dynamics of R. indica However, other than reports on the suitab ility or not of plants as hosts for R. indica (a complete list of the host plants R. indica is presented in Appendix), very little is known about the influence of plant characteristics on their quality as hosts for R. indica Sakar and Somchoudhury (1989) studied possible resistance mechanisms to R. indica in 8 coconut varieties based on morphological (pinnae length and wi dth, leaf thickness, depth of midrib groove, and intervienal distance) and biochemical (crude protein, moisture nitrogen and mineral cont ent) characteristics. The study concluded that there was no relations hip between mite populations and the physical (morphological) characteristics of the coconut vari eties but the chemical characteristics had an influence on mite densities. Varieties which contained higher amounts of nitrogen and crude protein showed higher incidence an d higher mite population densities. More recently, R. indica was reported as the first mite species observed feeding through the stomata of their host plants (Kane et al. 2005). Later it was s hown that feeding via stomatal

PAGE 24

24 aperture occurred among several Raoiella species (Ochoa et al. 2011). Further studies are needed to understand the implications of the feeding habits of R. indica Through this specialized feeding habit, R. indica probably interferes with the photosyn thesis and respiration processes of their host plants. Moreover, the t ype of stomata and their densities could determine the suitability of a plant as host for R. indica For instance, coconut varietie s can be divided in two main groups, tall and dwarf varieties. Tall palms may attain heights of 30 m (approximately) while dwarfs have shorter internodes and thus are much shorter (Gomes and Prado 2007). Tall varieties are greatly influenced by wind currents that rem ove the humidity from the leaf surface and thus have more efficient osmoregulation mechanisms than dwarf varieties (Gomes and Prado 2007). Dwarf varieties have a greater number of stomata per leaf area and lower wax content on the leaf surface compared to the tall varieties (Rajagopa l et al. 1990). These differences among coconut palm varieties could be influential on their suitability as hosts for R. indica which merits investigation. Detailed studies are needed to determine the suitability of banana cultivars as hosts of R. indica Cocco and Hoy (2009) reported a poor establishment and reproduction of R. indica on leaf discks and young plants of seve ral banana and plantain cultivars. Raoiella indica failed to establish on seedelings and young banana plants of the Grand Nain cultivar despite multiple inoculations with mites from filed-infested bananas and coco nuts (Carrillo, unpublished data). However, high populations of R. indica were found infesting banana trees of unknown cultivars in Big Pine Key (Carrillo, pers. obs.) and Lake Worth (Cocco and Hoy 2009) Florida. Further studies are needed to determine why some ba nana trees can sustain large populations of R. indica but others seem to be nonhosts of this mite.

PAGE 25

25 Influence of Abiotic Factors on Raoiella indica Few studies have examined the effects of abiotic factors on R. indica Moutia (1958) reported that population build up was positively rela ted with temperature and sunshine hours, but negatively related with rainfall and relative humidity. In agreement, Nagesha and Channabasavanna (1983b) reported an increase of mite populations associ ated with periods of lower relative humidity, and Taylor et al. ( 2011) found that when conditions were hotter and drier, R. indica densities were significantly higher. Sakar and Somchoudh ury (1989) reported that mite density had a significant positive relation with temperature and that R. indica densities declined with the onset of the rainy season in India. The available litera ture suggests that this mite adapts well to tropical climatic conditi ons; however, prolonged dry conditions could favor R. indica population growth.

PAGE 26

26 Figure 1-2. Raoiella indica on coconut. A. Oviposition by R. indica B. R. indica colony, the arrow shows a male closely behind a female deutonymph in a guarding state. C. R. indica colonies on the underside of coconut leaves. D. Damage on coconut leaves caused by R. indica

PAGE 27

27 Figure 1-3. Raoiella indica forming multigenerational colonies on the abaxial surface of coconut leaflets (pinnae).

PAGE 28

28 CHAPTER 2 A REVIEW OF THE NATURAL ENEM IES OF THE RED PALM MITE, RAOIELLA INDICA (ACARI: TENUIPALPIDAE)1 Summary A review of all the available information about the natural enemies reported in association with the red palm mite, Raoiella indica is presented. Twenty-eight species of predatory arthropods, including mites and insects, ha ve been reported in association with R indica in Asia, Africa and the Neotropics. The available literature indicates that each site has a different natural enemy complex with only one predator species, Amblyseius largoensis (Acari: Phytoseiidae), present in all the geographi cal areas. The phytoseiids, Amblyseius caudatus Berlese, Amblyseius channabasavanni Gupta and A. largoensis, were regarded as impor tant natural enemies of R. indica, and their predatory efficiency studied in some detail. Among the predatory insects the coccinellids Stethorus keralicus Kapur and Telsimia ephippiger Chapin were reported as major predators of R. indica In addition, pathogenic fungi associated with R. indica in the Caribbean have been reported. The known distribution, abunda nce and relative importance of each species reported in association with R. indica are discussed. Key words: Raoiella indica natural enemies, predatory mites, predatory insects, Amblyseius largoensis. Introduction The red palm mite, Raoiella indica Hirst (Acari: Tenuipalp idae), is an invasive phytophagous pest with a large host range of plants within the Arecaceae, Musaceae, Strelitziaceae, Heliconiaceae and Zingiber aceae (Carrillo et al. 2011a). During 2004, R. indica was detected in Martinique, and in a few year s it spread through the Caribbean and reached North America (United States of America, Me xico) and northern South America (Venezuela, 1 Reprinted with permission from Exp Appl Acarol

PAGE 29

29 Colombia, Brazil), now threatening the entire Neotropical region. Th e host-plant range and dispersal of R. indica throughout natural, agricu ltural, recreational and residential areas suggest that large-scale mitigation programs are required for managing this species. Chemical control, host-plant resistance and cultural control tactics could be used to manage local populations; however, only biological control has the potential to re gulate populations of this species on a large scale. The objective of this review is to analyze all the available information about the natural enemies, within various taxonomic groups, that have been reported in association with R. indica Natural Enemies of Raoiella indica Altogether 28 species of predatory arthr opods, including mites and insects, have been reported in association with R indica in various parts of the world (Table 2-1). In addition, three species of pathogenic fungi found infecting R. indica in Puerto Rico have been reported. Predatory Mites Sixrteen predacious mite species belonging to six families in two orders have been reported. Phytoseiidae ( Mesostigmata) Among the families of predatory mites the Phytoseiidae has the most species reported in association with R. indica Moutia (1958) reported Amblyseius caudatus Berlese as the main predator of R. indica on coconut in Mauritius. Predation by A. caudatus was recorded both in the field and in th e laboratory showing a marked preference for R. indica eggs, a high numerical response, and populati on growth coinciding w ith peak populations of the pest. The species Amblyseius channabasavanni Gupta and Daniel wa s reported by Daniel (1981) feeding upon R. indica on Areca (Areca catechu L.) leaves in the field and the laboratory. The author determined the biology and habits of A. channabasavanni preying on R. indica infesting Areca leaves in Kerala, India. Daniel (1981) referred to this species as a potentially

PAGE 30

30 effective predator because of its short generation time relative to that of the pest. In the field A. channabasavanni densities peaked coinciding with R. indica highest populations. When R. indica was not present A. channabasavanni were able to survive on alternate food sources including Tetranychus fijiensis Hirst (Acari: Tetranychi dae), eggs and crawlers of scale insects and mealybugs. According to the available literature, A. caudatus and A. channabasavanni were promising natural enemies of R. indica in the Old World. After R. indica gained importance as an invasive pest in the New World, surveys for natural en emies were conducted in Mauritius, India, Tanzania and Benin (Zannou et al. 2010; Bowm an and Hoy 2011; Taylor et al. 2011). These surveys did not encounter A. channabasavanni nor A. caudatus, but rather predatory mites identified as Amblyseius largoensis Muma Amblyseius largoensis was also found on coconut palms infested with R. indica in the Phillipines (Gallego et al. 2003). In the New World, the association between R. indica and A. largoensis was noticed soon after th e pest was found in the Caribbean and Florida (USA) (Pe a et al. 2009). This species wa s the most abundant predator associated with R. indica on coconut leaves in Trinidad a nd Tobago and Puerto Rico. In both sites, populations of A. largoensis increased in respons e to the arrival of R. indica In Florida, A. largoensis was the most abundant predator associated with R. indica and the only one found continuously throughout the year. The interaction between A. largoensis and R. indica in Florida was further investigated (Carrillo et al. 2010; Carrillo and Pea 2011, see also Chapters 3 and 4, respectively). An initial study demonstrated that A. largoensis was able to feed, develop and reproduce on a diet c onsisting solely of R. indica (Figures 2-1 and 2-2). The intrinsic rate of natural increase of the predator was significantly higher when fed on R. indica than on other diets, including Tetranychus gloveri Banks (Acari: Tetranychidae), Aonidiella orientalis

PAGE 31

31 (Newstead) (Hemiptera: Diaspididae), Nipaecoccus nipae (Maskell) (Hemiptera: Pseudococcidae) and live oak ( Quercus virginiana Mill.) pollen. Further studies showed that A. largoensis had a marked preference for R. indica eggs over other stages of the pest. In addition, the predator showed a Type II functional re sponse and a positive numerical response to increasing densities of R. indica which could explain the populati on increase observed in areas of recent invasion. Amblyseius largoensis was also recorded as the only phytoseiid species in association with R. indica in Cuba and Colombia (Ramos et al. 2010, Carrillo et al. 2011b). The species Neoseiulus longispinosus Evans was regarded as a potential predator of a large number of mite pests (Gupta 1998, 2001). The aut hor reported an associ ation between this predator and R. indica, citing a study conducted by Nangia and ChannaBasavanna (1989) in Karnataka, India. However, that study was conducted with the species Typhlodromips tetranychivorus Gupta, not with N. longispinosus. It is unclear whether predation by N. longispinosus upon R. indica has ever been recorded in the Eastern Hemisphere. Nevertheless, N. longispinosus which has primarily an Asian distribution, was recorded in Mart inique (Moraes et al. 2000), the same region where R. indica was first found in the Caribbean a few years later. Interestingly, N. longispinosus was observed feeding on R. indica in the field in Saint Lucia (Ochoa, pers. comm.). The predator is found in other islands of the Ca ribbean (Hastie et al. 2010) and its potential as a bi ological control agent of R. indica should be further investigated. Other phytoseiid mites reported in association with R. indica include T. tetranychivorus, Amblysieus raoiellus, and two unspecified species in the genera Amblyseius and Phytoseius. The life history of T. tetranychivorus was studied in the laboratory using R. indica as a host (Jagadish and Nageshachandra 1981; Nangia and ChannaBa savanna 1989). The predator was able to develop and reproduce feeding exclusively on R. indica, showing a preference to feed on eggs.

PAGE 32

32 However, no records of association of the two species are available despite the fact that both are found in Karnataka, India (Chinnamade-Gowda a nd Mallik 2010) (Table 2-1). In contrast, A. raoiellus was reported preying on R. indica in the same region (Denmark and Muma 1989). Finally, populations of Amblyseius sp. and Phytoseius sp. were reported in West Bengal (India) having negative and posi tive correlations with the populations of R. indica respectively (Somchoudhury and Sarkar 1987). No further inform ation is available for either species about their association with R. indica Ascidae (Mesostigmata) The feeding potential of Lasioseius sp. upon all stages of R. indica was determined in the labora tory (Sheeja and Ramani 2009). These authors reported that all stages of Lasioseius sp. preyed on all stages of R. indica However, there are no reports of association of Lasioseius sp. and R. indica under natural conditions in Karnataka (India). This predator is naturally asso ciated with the species Aceria ailanthae Mohanasundaram (Acari: Eriophyidae), which feeds on Euodia lunu-ankenda (Rutaceae) plants (Sheeja and Ramani 2009). Further studies are necessary to substantiate the potential of this species as a biological control agent of R. indica. Bdellidae (Trombidiformes). Two snout mite species, Bdella sp. and Bdella distincta (Barker and Bullock) were reported in association with R. indica in Trinidad and in Florida, respectively (Pea et al. 2009). Furthe r investigations in Florida found B. distincta feeding on R. indica (Figure 2-3), but also upon A. largoensis eggs (Carrillo, pers. obs.). The bdellid from Trinidad was also observed feeding on A. largoensis (Roda, pers. comm.). The low abundance observed in the field, and the intraguild -predation upon A. largoensis makes these Bdellidae species unlikely to be prom ising biocontrol agents of R. indica

PAGE 33

33 Cheyletidae (Trombidiformes) Three Cheyletidae species have been observed associated with R. indica in areas of recent invasion by this pest. Cheletomimus sp. and Mexecheles sp. were found in association with R. indica in Trinidad (Pea et al. 2009; Welbourn, pers. comm.). The species Hemicheyletia bakeri Ehara was observed feeding upon all stages of R. indica and A. largoensis in Florida (Carrillo, pers. obs .). According to Muma (1975), H. bakeri feeds and reproduces readily on various tetranychid, phytose iid and acarid mites found on citrus plants in Florida, and is relatively co mmon during winter and spring. Cunaxidae (Trombidiformes). The species Armascirus taurus Kramer was reported as a predator of R. indica infesting coconut palms in Cami guin, northern Mindanao, Philippines (Gallego et al. 2003); however, no fu rther information is available regarding this species and its relationship with R. indica Xenocaligonellidae (Trombidiformes). Xenocaligonellidus sp. was observed once in association with R. indica in Trinidad and Tobago (Welbourn, pers. comm.). No additional information is available to substantiate its potential as biocontrol agent of R. indica. Predatory Insects Altogether 12 predacious insect species belongin g to five families in f our orders have been reported in association with R. indica (Table 2-1). Coccinellidae (Coleoptera): Several species be longing to the genus Stethorus Weise, composed of specialist mite predators (Biddinger et al. 2009), have been reported in association with R. indica The species Stethorus keralicus Kapur was described from specimens collected on areca palm leaves infested with R. indica (Kapur 1961). Laboratory studi es revealed that both their larval and adult stages fed in large quantities upon all stages of R. indica (Puttaswamy and Rangaswamy 1976). Predators also showed a hi gh reproductive potential and were present throughout the year on coconut and areca palms. Daniel (1981) considered S. keralicus the most

PAGE 34

34 important predator of R. indica in Kerala, India. The predat or was also reported feeding on Raoiella macfarlanei Pritchard and Baker (Acari: Tenui palpidae) infesting roseapple ( Syzygium jambos L.) (Nageshachandra and Channabasava nna 1983a). The authors suggested that S. keralicus could be specific on mites of the genus Raoiella Apart from S. keralicus, three other Stethorus species have been recorded associated with R. indica on Areca plants in the state of Karnataka. Stethorus tetranychi Kapur and Stethorus parcempunctatus Kapur were recorded in the region of Mysore, whereas Stethorus pauperculus Weise was found in Shimoga (Puttarudriah and ChannaBassavana 1956; Yadav-Babu and Manjunatha 2007). However, feeding upon R. indica was observed only in S. tetranychi and no other information is available for the other specie s indicating their potential as biological control agents. Because of the reports of Stethorus spp. preying on R. indica in India, the potential of a Florida native species, Stethorus utilis Horn, was addressed. Before R. indica arrived in Florida S. utilis was reported as a common pred ator found associated with Tetranychus gloveri Banks (Acari: Tetranychidae) and other spider mite species on coconut leaves (Pea et al. 2009). A simple bioassay using fieldcollected adult beetles was de signed to determine whether S. utilis could feed on R. indica (Carrillo, unpublished data). Twen ty adult beetle s were placed individually in Petri dishes and starved for 8 hrs before th e beginning of the bioassay. The feeding test was conducted under no-choice conditions by introduci ng coconut leaf rectangles infested with R. indica in half of the petri dishes and with T. gloveri in the other half. A few minutes after introducing the prey items, predators were actively preying upon T. gloveri but not on R. indica The situation did not change over time. While most predators offered T. gloveri consum ed all their prey and oviposited on the coconut leaves, those offered R. indica spent most

PAGE 35

35 of their time wandering on the petr i dish walls, refused to feed on R. indica and ultimately died of starvation after approximately 48 hr. A si milar assessment was used to test feeding on R. indica by Stethorus punctillum Weise, a species that is mass-produced and commercially available. Adult beetles provided by the United States Departement of Agriculture (USDA)National Biological Control Labo ratory (Stoneville, Mississi ppi, USA) showed a similar response when offered R. indica and T. gloveri The lack of feeding on R. indica by S. punctillum was surprising because this spec ies is known to feed upon various food items in absence of their common tetranychid mite prey (Biddinger et al 2009). However, results of feeding tests on S. utilis and S. punctillum suggest they are not promis ing biocontrol agents of R. indica Apart from Stethorus three Coccinellidae species have al so been recorded in association with R. indica Telsimia ephippiger Chapin was found preying on R. indica on coconut leaves in the Phillipines (Gallego et al. 2003). Further studies determined that this species could complete its life cycle and reproduce feeding exclusively on R. indica showing high consumption rates (Gallego and Batomalaque-Galazar 2004). Jauravia soror (Weise) and Chilorus cacti L. were collected from R. indica -infested areca palms in Mysore (Karnataka, India) and coconut palms in Florida, respectively (Puttarudr iah and ChannaBassavana 1956; Pea, pers. obs.). However, there is no further information available to subs tantiate their importa nce as predators of R. indica Staphylinidae (Coleoptera) : Rove beetles belonging to the genus Oligota have been reported in association with R. indica in India. Somchoudhury and Sarkar (1987) reported Oligota sp. as the dominant predator of R. indica on coconut in West Bengal, India, where predator-prey populations showed a positive correlation. In Shimoga, Karnataka, Oligota sp. was found associated with peak population of R. indica on Arecas (Yadav-Babu and Manjunatha

PAGE 36

36 2007). Additional sampling to identify the rove b eetle species observed in association with R. indica is required to explore their potentia l as biocontrol agents of this pest. Chrysopidae (Neuroptera): Pea et al. (2009) reported Chrysopidae species associated with R. indica in Trinidad and Tobago, Puerto Rico, and Florida on coconut palms. Follow-up studies in Florida that involved collecting a nd rearing lacewing larvae observed feeding on R. indica in coconut fields, identified two lacewing species, Ceraeochrysa claveri (Navs) and Chrysopodes collaris (Scheider) (Carrillo unpublished data). Ceraeochrysa claveri was more common and repeatedly found feeding upon R. indica and ovipositing on infested coconut leaves (Figure 2-4). Studies on the de velopment and reproduction of C. claveri feeding on three phytophagous arthropods commonly found i nhabiting coconuts in Florida [ R. indica Nipaecoccus nipae Maskell (Hemiptera: Pseudococcidae) and Aonidiella orientalis (Newstead) (Hemiptera: Diaspididae)] showed these lacewings active preyed upon R. indica during the first two instars and had similar deve lopmental times compared to th at observed with the two other diets (Figure 2-4 C and D). Ca rrillo (unpublished data) also observed that third-instar development, survivorship and reproduction by adults reared on the R. indica diet were poor compared to these attributes for lacewings reared on the other two diets. Results suggested that C. claveri can use R. indica as an alternate prey, especially during the first two instars, but it depends on the presence of larger prey to comp lete its life cycle and reproduce successfully. Cecidomyiidae (Diptera) : Predacious midge larvae ha ve occasionally been observed feeding on R. indica in Trinidad and Tobago and Florida. Fe w larvae were reared into adults on a R. indica diet in Florida (Figure 2-5). Two females were obtained and identified as Arthrocnodax sp. (by Dr. R. J. Gagn, USDA Systematic Entomology Laboratory, Washington D.C.), a cosmopolitan genus of about 50 species known to feed mainly on the Eriophyidae (Gagn 2004).

PAGE 37

37 Efforts will continue to complete the identifica tion of the species for which males are needed; however, their relatively low abundance in the field makes them unlikely to be promising mortality factors of R. indica Phlaeothripidae (Thysanoptera) : The predatory thrips Aleurodothrips fasciapennis (Franklin) was reported in association with R. indica in Trinidad and Tobago and Florida (Pea et al. 2009). The available litera ture on this species (Watson et al. 1998, 2000), and results from the surveys in Trinidad, Puerto Rico and Florida suggest that the life history of A. fasciapennis depends on diaspidid insect prey (Figure 2-6). This predator has been occasionally observed feeding upon R. indica but so far has not shown potential as a biological contro l agent of this pest. Pathogens There are no studies known of pa thogens associated or infecting R. indica in the literature. Pronounced reductions of the pest were observed in various sites in Puerto Rico (Rodrigues and Colon, unpublished). The reduction coincided with rainfall increases an d consequent higher moisture, which leads to the suggestion that rainfa ll could directly play an important role on the mite population density. Similar understandings were reported in Mauritius (Moutia 1958). However, a close look on the mites showed that great numbers were dying from an infection. In order to isolate the pot ential infectious microorganisms asso ciated with the mite populations, individual mites showing symptoms of infection were placed in Petri dishes with selective media for isolation of bacteria, fungi and actinomycetes Isolated fungi were morphologically identified and had a fragment of ITS gene sequenced (Rod rigues and Colon, unpublished). Four isolates of fungi were found to be pathogenic to the mites. Three species, Simplicillium sp., Lecanicillium lecanii, and Hirsutella thompsonii, were isolated and identified. St udies were carried out in leaf

PAGE 38

38 arenas and in greenhouse conditions to demonstrate infection of R. indica (Rodrigues and Colon, unpublished). Discussion Altogether 28 species of predators have been reported in association with R. indica in various regions of Asia, Africa, the Caribbean basin, and North and South America (Table 2-1). Most records were made in India, with 13 species from the southwestern states of Karnataka, Kerala, and fewer from West Bengal. Apart from India, the only other reports from Asia were from the Philippines. In Africa, two species we re reported in Mauritius and one in Benin and Tanzania. In the New World five predatory sp ecies have been recorded in the Caribbean (Trinidad, Puerto Rico, Cuba a nd St. Lucia), nine species in Florida (USA), and one both in Colombia and Mexico. According to the available literature, each site has a different natural enemy complex with only one predator species, A. largoensis present in all the geographical areas. Of the sixrteen predatory mites spec ies recorded in association with R. indica in various parts of the world, only nine were observed feeding on R. indica in the field. Within those eight species, only three Phytoseiidae species have been studied in so me detail, including A. caudatus, A. channabasavanni and A. largoensis Bioassays conducted with thes e species determined that they were able to complete developm ent and reproduce when feeding solely on R. indica (Moutia 1958; Daniel 1981; Carrillo et al. 2010, chapter 3). The three Amblyseius species showed preference for eggs over other devel opmental stages of the pest; in addition, A. channabasavanni also preferred quiescent larvae and nymphs. The three species have shorter developmental times (approximately one week) than R. indica and a positive numerical response to increasing populations of the pest. Until now th ere is no information suggesting that the other Phytoseiidae species found associated with R. indica could play an important role regulating this

PAGE 39

39 pest. The predatory mites of the families Bdellidae, Cheyletidae and Cunixidae were found at significantly lower densities and sometimes pr eying on phytoseiids. There is no information suggesting that they could be of import ance as biological control agents of R. indica The available evidence suggests that within the Phytoseiidae A. caudatus, A. channabassavani and A. largoensis have the most potential as biological control agents of R. indica The first two species were regarded as major predators of R. indica in Mauritius and the st ate of Kerala (India), respectively. However, no other records are availa ble of the association between these predators and R. indica since they were reported in 1958 and 1981, respectiv ely. Since then, R. indica gained importance as an invasive pest in the Neotropics and surveys for natural enemies were conducted in several places of the world (Roda et al. 2008; Pea et al. 2009; Ramos et al. 2010; Zannou et al. 2010; Carrillo et al. 2011b) incl uding Mauritius and Ke rala (Bowman and Hoy 2011; Taylor et al. 2011). Through these surveys, A. largoensis was identified as the most abundant predator and often as the only phytoseiid species associated with R. indica With the exception of a single observation of N. longispinosus feeding on R. indica recorded by Ochoa in Saint Lucia, to this day no other phytoseiid mite has been reported feeding upon R. indica in the multiple surveys. This particular situation ra ises the question of whether A. largoensis is truly a single species or a group of morphologically similar species. The species A. largoensis was first described from specimens collected in Key Largo, Florida by Muma (1955) under the combination Amblyseiopsis largoensis and later moved to the genus Amblyseius. However, A. largoensis has a cosmopolitan distribution (McMurtry and Moraes 1984; Moraes et al. 2004). This species is part of a group of nine clos ely related species refe rred collectively as the Largoensis group (McMurtry a nd Moraes 1984). Except for A. largoensis A. herbicolus and A.

PAGE 40

40 eharai the other six species of the group have been recorded only in the Australian biogeographic realm, suggesting that the largoensis group originated there (McMurtry and Moraes 1984). The authors hypothesized that A. largoensis was dispersed through movement of plant material. Observations ma de in Florida suggest that A. largoensis has an important ability to colonize various plants. For instance, in surv eys conducted at the Fairchild Tropical Botanic Garden in Florida R. indica was found on 36 palm species and A. largoensis was found on half of them (Appendix). This predator was also repeated ly reported on citrus and other plants (Muma 1955; Muma 1975; Daneshvar 1980; Galvao et al. 2007). It has been hypothesized that the classification of A. largoensis based on morphological characters could obscure the fact that these are multiple cryptic species. Using mo lecular techniques Bowman (2010) compared populations of A. largoensis from Florida and Mauritius. Their analysis found differences between the populations but was not conclusive as to whethe r the Florida and Mauritius populations were biotypes or cryptic species of A. largoensis Therefore, it should be determined whether all the species identified as A. largoensis are equally efficient natural enemies of R. indica Only two studies have a ddressed the efficiency of A. largoensis In an initial approach Pea et al. (2009) concluded that A. largoensis was unable to suppress th e large populations of R. indica observed in Trinidad and Puerto Rico during th e early stages of the invasion by this pest. In Florida, the functional and numerical responses of the A. largoensis to R. indica were determined in the laboratory and their efficacy te sted using exclusion and release techniques (see chapter 6). Results of thos e studies suggested that A. largoensis could be effici ent at regulating low prey population densities. It will be important to design strategies that can allow realistic comparisons of the efficiency of predators of R. indica in various parts of the world. While these questions remain unresolved, it is clea r that predatory mites identified as A. largoensis have

PAGE 41

41 shown a conspicuous association with R. indica and represent the most important biotic factor known to be acting over R. indica in the different places where this pest is present. Among the 12 predatory insects re ported in association with R. indica the family Coccinellidae is represented by 7 species. Four Stethorus species were reported in the regions of Karnataka and Kerala in Indi a. Among these the species S. keralicus showed great potential and was reported as a voracious predator of all stages of R. indica was found throughout the year and showed a degree of specificity towards Raoiella species. However, the latest record of the association between the two species was made in 1976 despite the increasing interest in identifying possible classical bi ological control agents for R. indica There is no information available regarding the other three Stethorus spp. reported in association with R. indica Simple bioassays, such as those conducted in Florida with S. utilis and S. punctillum could be useful to determine whether they feed on R. indica Studies on S. keralicus and other Stethorus species previously reported in association with R. indica are needed as they could represent an important tool in managing this invasive species in the Neotropics and other areas of the world. Apart from Stethorus the other ladybeetle species showing some potential as a biological control agent of R. indica is T. ephippiger The species was regarded as a voracious predator of R. indica in the Philippines (Gallego et al. 2003). The authors reported T. ephippiger as predator of Rarosiella cocosae Rimando, a synonym of R. indica (Mesa et al. 2009), which could create confusion for researchers interested in the lite rature about natural enemies of R. indica There is no information suggesting that the other two coccinellids ( J. soror and C. cacti ) could be importa nt as biological control agents of R. indica The predacious beetle Oligota sp. was regarded as an important predator of R. indica in West Bengal (India) (Somc houdhury and Sakar 1987). The genus Oligota contains multiple species that are specialist mite predators (Frank et al. 1992). It would

PAGE 42

42 be useful to investigate the occurrence of Oligota species in West Bengal and in other areas where R. indica is present. The other insect predators include predacious lacewings, thrips, and midges. All of them were repor ted in areas of recent invasi on in the Neotropics and were observed feeding on R. indica However, the available evidence s uggests that their life history is linked to other prey. Their prey preferences and low abundance obs erved in the field make these species unlikely to be prom ising biocontrol agents of R. indica The observation and isolation of ar thropod-pathogens associated with R. indica is quite recent (Rodrigues and Colon, unpublished). Four acar opathogenic fungi were reported associated with epizootics in R. indica populations in Puerto Rico. Fungi c ould be particularly important in reducing mite densities in humid regions, wh ich are mostly found in coconut and bananagrowing areas in Central America, the Caribbean and parts of South Am erica. These pathogens, rather than direct impact of the rain as s uggested by Moutia (1958), could be the cause of significant reductions in mite population densitie s observed in some sites in Puerto Rico (Rodrigues et al. unpublished). Furt her studies to evaluate the e fficiency of single or complex species of pathogens to control R. indica should be undertaken to determ ine their potential use in commercial settings. In conclusion, the available literature indicates that A. largoensis is the most abundant predator of R. indica where this pest is present. It will be important to determine which populations of A. largoensis are more efficient in preying upon R. indica and why other phytoseiid species, such as A. caudatus and A. channabasavanni have not been recorded in recent surveys for natural enemies. Due to the marked preference that A. largoensis exhibits for R. indica eggs, it would be desirable to find natural en emies that could prey on or parasitize other stages of the pest. For instance, the recen t finding of acaropathogenic fungi attacking R. indica in

PAGE 43

43 Puerto Rico could be very useful for managing this pest (Rodrigues et al unpublished). At a local level some predatory species ( S. keralicus and T. ephippiger) were reported as important predators but remain unexplored as po tential biological control agents of R. indica. It is likely that other mortality factors will be required to effectively suppress the populations of this invasive pest. Search for effective natural enemie s should be intensified and ways to improve the levels of control by the existing natural enemie s (i.e. provision of a lternative food sources) should be further investigated.

PAGE 44

44 Table 2-1. Predatory arthropods re ported in association with R. indica. Scientific name (synonyms) Order: Family Place of report Reference Amblyseius caudatus Berlese 1914 (= Typhlodromus caudatus) (Acari: Phytoseiidae) Mauritius Moutia 1958 Amblyseius channabasavanni Gupta & Daniel (= Amblyseius channabasavannai ) (Acari: Phytoseiidae) Karnataka and Kerala, India Daniel 1981, Gupta 2001 Amblyseius largoensis (Muma 1955) (Acari: Phytoseiidae) Benin; Tanzania Mauritius; Kerala, India; Philippines; Trinidad; Puerto Rico; Florida, USA; Colombia;Cuba Zannou et al. 2010 Bowman and Hoy 2011 Taylor et al. 2011 Gallego et al. 2003 Pea et al. 2009; Roda et al. 2008 Pea et al. 2009 Carrillo et al. 2011b; Ramos et al. 2010 Amblysieus raoiellus Denmark & Muma 1989 (Acari: Phytoseiidae) Karnataka, India Denmark and Muma 1989 Neoseiulus longispinosus (Evans 1952) (= A.. longispinosus ) (Acari: Phytoseiidae) St. Lucia Roda et al. 2008 Typhlodromips tetranychivorus Gupta 1978 (= Amblyseius tetranychivorus = Transeius tetranychivorus ) (Acari: Phytoseiidae) Bangalore and Karnataka, India Jagadish and Nageshachandra 1981, Nangia and Channabasavanna 1989 Amblyseius sp. (Acari: Phytoseiidae) West Bengal, India Somchoudhury and Sakar 1987 Phytoseius sp. (Acari: Phytoseiidae) West Bengal, India Somchoudhury and Sakar 1987 Lasioseius sp. (Acari: Ascidae) Karnataka, India Sheeja and Ramani 2009 Bdella distincta (Barker and Bullock) (Acari: Bdellidae) Florida, USA Pea et al. 2009 Bdella sp. (Acari: Bdellidae) Trinidad Roda et al. 2008; Pea et al. 2009 Cheletomimus sp. (Acari: Cheyletidae) Trinidad R oda et al. 2008; Pea et al. 2009 Hemicheyletia bakeri (Ehara) (Acari: Cheyletidae) Florida, USA Carrillo, pers. obs. Mexecheles sp. (Acari: Cheyletidae) Trinidad Welbourn, pers. obs. Armascirus taurus Kramer (Acari: Cunaxidae) Philippines Gallego et al. 2003 Xenocaligonellidus sp. (Acari: Xenocaligonellidae) Tr inidad Welbourn, pers. obs. Stethorus keralicus Kapur (Coleoptera: Coccinellidae) Kerala, India Kapur 1961; Puttaswamy and Rangaswamy. 1976 Stethorus parcempunctatus Kapur (Coleoptera: Coccinellidae) Karnataka, India Puttarudriah and ChannaBassavana 1956 Stethorus pauperculus Weise (Coleoptera: Coccinellidae) Karnataka, India Yadav-Babu and Manjunatha 2007 Stethorus tetranychi Kapur (Coleoptera: Coccinellidae) Kerala, India Puttarudriah and ChannaBassavana 1956

PAGE 45

45 Table 2-1. Continuation Scientific name Order: Family Place of report Reference Jauravia soror (Weise) (Coleoptera: Coccinellidae) Karnataka, India Puttarudriah and ChannaBassavana 1956 Telsimia ephippiger Chapin (Coleoptera: Coccinellidae) Phillipines Gallego et al. 2003 Chilorus cacti L (Coleoptera: Coccinellidae) Florida, USA Pea, pers. obs. Oligota sp. (Coleoptera: Staphylin idae) Karnataka and West Bengal, India Yadav-Babu and Manjunatha 2007; Somchoudhury and Sakar 1987 Ceraeochrysa claveri (Navs) (Neuroptera: Chrysopiidae ) Florida, USA Carrillo, pers. obs. Chrysopodes collaris (Scheider) (Neuroptera: Chrysopiid ae) Florida, USA Carrillo, pers. obs. Aleurodothrips fasciapennis (Franklin 1908) (Thysanoptera: Phlaeothripidae) Florida, USA; Trinidad Pea et al. 2009; Roda et al. 2008 Arthrocnodax sp. (Diptera: Cecidomyiidae) Florida, USA Carrillo, pers. obs.

PAGE 46

46 Figure 2-1. Amblyseius largoensis feeding upon R. indica and ovipositing on infested coconut leaves in Florida. A. A. largoensis female feeding on R. indica eggs. B. A. largoensis female feeding on R. indica larva, dorsal view. C. A. largoensis female feeding on R. indica lateral view. D. A. largoensis eggs on the midvein of a coconut pinna. E. A. largoensis oviposition near R. indica eggs. F. A. largoensis and R. indica eggs.

PAGE 47

47 Figure 2-2. Life cycle of A. largoensis feeding upon R. indica in Florida. A. Egg. B. Larva C. Protonymph D. Deutonymph. E. Adult female

PAGE 48

48 Figure 2-3 Bdella distincta found feeding upon R. indica in Florida. Figure 2-4. Ceraeochrysa claveri feeding upon R. indica and ovipositing on infested coconut leaves in Florida. A. C. claveri eggs B. C. claveri neonate larvae C. C. claveri first instar larva feeding on R. indica nymphs D. C. claveri second instar larvae feeding on R. indica

PAGE 49

49 Figure 2-5. Arthrocnodax sp. observed feeding on R. indica in Florida. A. Predacious midge larvae B. Adult female. Figure 2-6. Aleurodothrips fasciapennis observed feeding upon R. indica in Florida. A. Adult in lateral view. B. A. fasciapennis feeding upon R. indica C. A. fasciapennis feeding upon on diaspidid insect prey.

PAGE 50

50 CHAPTER 3 DEVELOPMENT AND REPRODUCTION OF AMBLYSEIUS LARGOENSIS (ACARI: PHYTOSEIIDAE) FEEDING ON POLLEN, RAOIELLA INDICA (ACARI: TENUIPALPIDAE), AND OTHER MICROAR THROPODS INHABI TING COCONUTS IN FLORIDA2 Summary The red palm mite, Raoiella indica (Acari: Tenuipalpidae), is an important pest of palms (Arecaceae) and other species within the Zingi beraceae, Musaceae and Strelitziaceae families. Raoiella indica was discovered in the United States of America (Palm Beach and Broward counties, Florida) late in 2007, and it subsequently spread to other Florida counties. The predatory mite Amblyseius largoensis (Acari: Phytoseiidae) has b een found associated with R. indica in Florida. In order to verify whether A. largoensis can develop and reproduce when feeding exclusively on R. indica the biology of this predator was evaluated on contrasting food sources, including R. indica Five diets [ R. indica Tetranychus gloveri Aonidiella orientalis Nipaecoccus nipae oak ( Quercus virginiana ) pollen] and a no-food control were tested to determine the predators development, survivor ship, oviposition rate, sex ratio and longevity at 26.5 1C, RH 70 5 % under a 12:12 L:D photophase. Amblyseius largoensis was able to complete its life cycle and re produce when fed exclusively on R. indica The development of immature stages of A. largoensis was faster and fecundity and su rvivorship were higher when fed on R. indica or T. gloveri compared to the other food sources. The intrinsic rate of natural increase of A. largoensis was significantly higher when fed on R. indica than on other diets. These results suggest that, de spite earlier assessments, A. largoensis can play a role in controlling R. indica Key words: mites, invasive species, biological control, Cocos nucifera Raoiella indica Amblyseius largoensis. 2 Reprinted with permission from Exp Appl Acarol

PAGE 51

51 Introduction Raoiella indica Hirst (Acari: Tenuipalpidae) is an important pest of coconut, Cocos nucifera L. (Hirst 1924), areca palm, Dypsis lutescens (H.Wendland) (Kane et al. 2005), date palm, Phoenix dactylifera L. (Sayed 1942; Zaher et al. 1969), canary palm, Phoenix canariensis hort. ex Chabaud (Etienne and Flechtmann 2006), other species within the Arecaceae, bananas and plantains ( Musa spp. ) (Kane et al. 2005; Etienne a nd Flechtmann 2006) within the Musaceae, gingers Zingiber spp. (Pea et al. 2006) within th e Zingiberaceae, and bird of paradise, Strelitzia spp. (Etienne and Flechtmann 2006), within the Strelitziaceae. Until recently, this mite was found only in the Eastern Hemisp here, probably widespread throughout tropical and subtropical regions (Nageshachandra and Ch annabasavanna 1983b) of India, the Philippines, Mauritius, Reunion, Malaysia, Israel and Egypt. During 2004, R. indica was detected in Martinique and St. Lucia in the Western Hemi sphere, and rapidly expanded its geographical range to other Caribbean islands, where high popul ations have severely affected the coconut, ornamental palm and banana industries (Kan e et al. 2005; Etienne and Flechtmann 2006; Rodrigues et al. 2007). More recently, R. indica was discovered in North and South America; an infestation was reported in the state of Sucr e in Venezuela in 2007 (Gutirrez et al. 2007; Vsquez et al. 2008), and in December of the same year it was detected in the West Palm Beach area of south Florida (F DACS 2007). Since then, R. indica has spread to six counties in Florida, being detected in at least 439 individual sites on 25 host plant species (Welbourn 2007). Populations of this tenuipalpid have incr eased dramatically, sometimes reaching 4,000 individuals per leaflet on coconut s (Pea et al. 2009), the most co mmon host plant in Florida. On coconuts, R. indica is found in very large numbers on palm fronds with all stages located on the abaxial surface of the leaflets (p innae). Recent studies indicate th at some Florida native palms, such as the saw palmetto, cabbage palm, dwarf palmetto, and sable palmetto, are unsuitable hosts

PAGE 52

52 for R. indica (Cocco and Hoy 2009; Amalin, pers. comm.). Nevertheless, R. indica has been found attacking the endangered na tive Florida thatch palm ( Thrinax radiata Lodd. ex J. A. and J. H.) and several ornamental palms (Ward et al. 20 03; Amalin, pers. comm.). Furthermore, this is a quarantined pest for some countries (EPPO 2009) A management plan to mitigate the damage that this invasive species is causing to Floridas palms and comm ercial ornamental palm industry is urgently needed. Biological control is one of the most importa nt components of integrated pest management with the potential to regulate populations of invasive species. Several predators have been found associated with R. indica in the Old World. Mou tia (1958) reported that Amblyseius caudatus Berlese (= Typhlodromus caudatus Chant) (Acari: Phytoseidae) was the main predator of R. indica in Mauritius. Daniel (1981) re ported that the phytoseiid mite Amblyseius channabasavanni Gupta and the lady beetle Stethorus keralicus Kapur (Coleoptera: Coccinellidae) were the most important predators of R. indica in India. In Florida, preand postinfestation surveys identified some predatory natural enemies. On coconuts, Amblyseius largoensis (Muma) (Acari: Phytoseiidae), Stethorus utilis (Horn) (Coleoptera: Coccinellidae), and Ceraeochrysa claveri Navas (Neuroptera: Chrysopidae) were the most common predators followed by Bdella distincta (Barker and Bullock) (Acari: Bdellidae) and Aleurodothrips fasciapennis (Franklin) (Thysanoptera: Phlaeothripid ae) (Pea et al. 2009). Among these, A. largoensis was the most abundant predator and its popul ations increased in numbers after the arrival of R. indica in south Florida (Pea et al. 2009). Amblyseius largoensis was also found in Trinidad and Tobago, Puerto Rico and Dominica associated with R. indica on coconut (Rodrigues et al. 2007; Pea et al. 2009; Hoy, pers. comm.; Roda, pers. comm.). Based on these

PAGE 53

53 observations it became apparent that a detaile d study on the biological control potential of A. largoensis against R. indica was needed. This study was designed to determine whether A. largoensis can: 1) develop and reproduce when feeding solely on R. indica 2) develop and reproduce on othe r prey or food sources, and 3) compare developmental and reproductive parameters of A. largoensis when feeding on R. indica, other potential prey or pollen. Materials and Methods The development and reproduction of A. largoensis was evaluated on five diets. The diets were selected based on two surveys made in south Florida to determine the most common phytophagous arthropods inhabiting coconuts (Pea et al. 2009). In addition, pollen was included as a food source that might play a role on sustaining the populations of A. largoensis (Yue and Tsai 1996). These diets were R. indica Tetranychus gloveri Banks (Acari: Tetranychidae), Aonidiella orientalis (Newstead) (Hemiptera: Diaspididae), Nipaecoccus nipae (Maskell) (Hemiptera: Psudococcidae) and live oak ( Quercus virginiana Mill) pollen. A no-food treatment was included as a negative control. Source of A. largoensis R. indica T. gloveri A. orientalis N. nipae and Q. virginiana Pollen. Stock colony of A. largoensis Foliage samples containing large numbers of R. indica and A. largoensis were collected from a pesticide-free Malayan dwarf coconut palm plantation locat ed in Broward County (26o02.925 N 080 o09.822 W) in January of 2009, and taken to the Tropical Fruit Entomology Laboratory at the Tropical Research and Education Center (TREC) of the University of Florida, Homestead. The samples were caref ully inspected under a dissecting microscope (50) to isolate and transfer 150 A. largoensis females into 3 rearing arenas (50 fe males per arena) that were kept at 26.5 1C, RH 70 5 % and 12:12 L:D photophase The rearing arenas were square (8 8

PAGE 54

54 cm) black card stock (acid and lignin free) co ated with paraffin and placed on water-soaked cotton (10 10 3 cm); paper strips (Kimwipe, KimberlyClark Corporation, Roswell, GA) were placed along the edges of th e cardboard to keep the mites w ithin the arena. Each rearing unit was placed in a plastic tray (12 12 cm) and water was added daily to keep the cotton wet. The predatory mites were provided R. indica (all stages brushed onto the arena), oak ( Quercus virginiana ) pollen and honey-water so lution (10 %) as food sources three times a week. The colonies were subcultured by transferring a mi nimum of 50 adult female predators to a new arena every three weeks. Stock colony of R. indica Additional foliage samples from the same site as above were used to collect about 500 R. indica females to infest 1-year-old potted Malaya n dwarf coconut palms (about 1.5 m in height). Five palms were carefully cleaned with a br ush to remove all other arthropods. A yellow tagging tape coated with Tanglefoot was tied around the base of the palms to exclude any crawling arthropods. On each palm, 100 R. indica females were placed on the abaxial surface of the frond and the palms were placed inside an isol ated rearing room kept at 30 1C, RH 60 5 % under a 12:12 L:D photophase. The palms were left undisturbed for 45 days; then R. indica females were collected from 1 palm and used as stock to infest a new palm and also to feed the A. largoensis colonies. The procedure was repeated every week. Stock colony of T. gloveri Tetranychus gloveri was collected and reared similarly to R. indica The T. gloveriinfested palms were placed in a separate gl ass-greenhouse (30 10C, 60 30 % RH and natural photophase), and regularly inspected with a 10 hand lens to remove all other arthropods. Tetranychus gloveri and A. largoensis were identified by Dr. Cal Welbourn, Florida Department of Agriculture and Consumer Se rvices, Division of Plant Indu stry, Gainesville, Florida.

PAGE 55

55 Stock colonies of A. orientalis and N. nipae The oriental scale ( A. orientalis ) and the coconut mealybug ( N. nipae ) were obtained from existing colonies reared for several generations at TREC. Aonidiella orientalis was reared on butternut squash, Cucurbita moschata (Duchesne ex Lam.), as described by Elder and Smith (1995). Nipaecoccus nipae was reared on sprouted potatoes ( Solanum tuberosum sub. tuberosum ) inside an incubator kept at 25 1C, RH 60 5 % in complete darkness, as described by Meyerdirk et al. (1988). Coconut palms were cleane d as described previously and infested by transferring either scale or mealybug crawlers to the abaxia l surface of the fronds. Infested palms were transferred to separate greenhouses (30 10C, 60 30 % RH and natural photophase) and inspected regularly to remove all other arthropods. Source of pollen The pollen used in this study was co llected from flowering live oak ( Quercus virginiana Mill) trees in Homestead, FL (25o32.39 N 80 o28.26 W) during February 2009. Inflorescences were cut and placed in paper bags and sealed for a week until the pollen was released. The pollen was then sifted through a 3 sieve set (U.S.A. Standard testing sieves, Nos. 50, 100 and 325; 0.297, 0.149 and 0.043 mm openings respectively; Fisher Scientific Company) to remove any impurities, collected and placed in a sealed co ntainer and kept frozen at 0C until used. Feeding Experiments with A. largoensis Experimental arenas The experimental arenas cons isted of squares (4 4 cm) cu t from mature coconut leaves that were previously infested with the different arthropod species considered. Each leaf square was inspected under a dissecting microscope to re move any other arthrop od and to standardize prey density in the treatments that had prey as a food source. Treat ments containing either R. indica or T. gloveri had at least 30 individuals including a minimum of 10 eggs, 10 nymphs, and

PAGE 56

56 10 adults. Treatments containing either A. orientalis or N. nipae had 3 gravid adult females and at least 30 crawlers (mobile first instars). Th e treatment containing pollen was constructed by placing approximately 0.2 g of Q. virginiana pollen on a clean leaf square, and the no-food treatment consisted of a clean leaf square. The coconut le af squares corresponding to the different treatments were placed with the ab axial surface up on cotton squares (6 6 2 cm) saturated with water in a plastic tray (hex agonal polystyrene weighi ng dishes 12.7/ 8.9 cm, Fisherbrand cat. No. 02-202-103) Paper strips (Kimwipe, Kimberly Clark Corporation, Roswell, GA) were placed along th e edges of the leaf squares to prevent the arthropods used in the assay from escaping. A small plastic square (0.5 0.5 cm) cut from Ziplock bags was placed on top of each arena as a shelter for A. largoensis Experimental setup About 300 A. largoensis females from the colonies were transferred to 3 experimental arenas (100 individuals per aren a) with no food for 24 h at 26.5 1C, RH 70 5 % and 12:12 L:D photophase. Afterwards females were remove d from the arenas and the newly deposited eggs kept at the same envir onmental conditions. The arenas we re observed every 2 h for egg eclosion. Emerged larvae were transferred to ex perimental arenas individually and observed every 12 h to determine the survivorship and duration of the immature stages on each diet. The diets in each arena were replenished daily and arenas were replaced every three days. After completing development, sex was determined and a male from the colony was transferred into each arena where an adult female had emerged. Introduced males were replaced if they died. Adults were observed every 24 h to determine their preoviposition and oviposition periods, and the longevity of both sexes. Eggs were removed and transferred to new e xperimental arenas to determine the sex ratio of the offspring in each treatment.

PAGE 57

57 Statistical Analysis The experimental design was completely randomi zed with six treatments (5 diets and a nofood control) and 40 replicates per treatment. AN OVAs (SAS Institute, Inc. 2003) were used to detect the effect of diet on the duration of the immature stages, preoviposition and oviposition period, daily and total oviposition, and the longevity of A. largoensis Survivorship of immature stages and the sex ratios of the proge ny were compared using chi-square ( 2) tests (SAS Institute, Inc. 2003). Fertility life table parameters including the net reproductive rate ( Ro), generation time (T), and intrinsic rate of increase ( rm) were estimated using the Jackknife procedure as described by Maia et al. (2000). We tested differences in the fe rtility life table parameters between diets using Student Newman-Keuls multi ple range test (SAS Institute, Inc. 2003). Results The development and survivorship of immature stages of A. largoensis were influenced by the type of diet consumed. Larvae of A. largoensis developed into protonymphs feeding on R. indica T. gloveri, A. orientalis N. nipae or Q. virginiana pollen, but also without food. Larval developmental times of A. largoensis were not influenced by the type or presence of food ( F = 1.21; df = 5, 212; P = 0.300); however, larval survival wa s significantly higher with food than without [ 2 (5, N=240) = 55.96, P < 0.001] (Table 3-1). In the absence of food, the protonymphs of A. largoensis failed to develop into the deutonymph. In the presence of food, protonymphal development was influenced by the type of diet ( F = 26.35; df = 5, 133; P < 0.001). Protonymphal development of A. largoensis was significantly fast er when feeding on R. indica or T. gloveri than on A. orientalis N. nipae or Q. virginiana pollen (Table 3-1). Survival rates were similar on protonymphs fed with R. indica T. gloveri A. orientalis or Q. virginiana pollen, but significantly lower on those fed with N. nipae [ 2 (5,N=213)= 89.93, P < 0.001]. Deutonymphal development of A. largoensis was also influenced by the type of diet ( F = 25.4; df

PAGE 58

58 = 4, 104; P < 0.001), being significa ntly shorter on the R. indica or T. gloveri than on A. orientalis N. nipae or Q. virginiana pollen diets (Table 3-1). Surviv al rates of this stage were high in individuals fed with R. indica T. gloveri or Q. virginiana pollen but significantly lower on those fed with A. orientalis or N. nipae [ 2 (4, N=134) = 36.23, P < 0.001]. In summary, the immature development of A. largoensis was significantly faster when fed either R. indica or T. gloveri than on other food sources ( F = 44.02; df = 5, 104; P < 0.001) (Table 3-1). In addition, more A. largoensis individuals reached the adult stage when fed on R. indica T. gloveri or Q. virginiana pollen [ 2 (5, N=240) = 103.94, P < 0.001]. Survival rates of A. largoensis were significantly lower when fed on A. orientalis and lowest when fed on N. nipae The type of diet influenced the reproductive rates of A. largoensis For instance, females fed with A. orientalis or N. nipae did not reproduce. For this reason, A. largoensis females fed with these two diets were excluded from th e fertility and life table analyses. The diet significantly influenced the durati on of the pre-oviposition period (F = 23.91; df = 5, 52; P < 0.001). Females fed with R. indica or T. gloveri had a significantly shorter preoviposition periods compared to those fed with Q. virginiana pollen (Table 3-2). However, the oviposition period was similar among those three diets ( F = 1.34; df = 5, 52; P =0.267). Female longevity was not influenced by the type of diet, but males fed with R. indica and T. gloveri lived significantly longer compared to those fed with Q. virginiana pollen ( F = 2.72; df = 5, 52; P = 0.276 and F = 7.36 df = 5, 52; P < 0.001 for female and male longevities respectively) (Figure 3-2). The daily oviposition rate of A. largoensis females fed exclusively with R. indica was significantly higher than those fed with T. gloveri or Q. virginiana pollen ( F = 25.88; df = 5, 52; P < 0.001) (Figure 3-1). In addition, the total numbe r of eggs deposited was signifi cantly higher in females fed R. indica or T. gloveri compared to those fed Q. virginiana pollen ( F = 9.54; df = 5, 52; P < 0.001)

PAGE 59

59 (Table 3-2, Figure 3-1). The proportion of females in the progeny ranged from 0.63 to 0.73 among the three diets. No effects of the type of diet on the sex ratio of the offspring were detected [ 2 (2, N=682) = 5.64 P = 0.059] (Table 3-2). The net reproductive rate ( Ro) of A. largoensis fed with R. indica was significantly higher than those fed with T. gloveri or Q. virginiana pollen ( F = 18.32; df = 2, 58; P < 0.001). The mean generation time ( T ) was significantly shorter in A. largoensis fed on R. indica or T. gloveri compared to those fed with Q. virginiana pollen ( F = 12.36; df = 2, 58; P < 0.001). Finally, the intr insic rate of increase ( rm) of A. largoensis fed with R. indica was significantly highe r than those fed with T. gloveri or Q. virginiana pollen ( F = 34.18; df = 2, 58; P < 0.001) (Table 3-2). Discussion The results of this study suggest that larvae of A. largoensis are facultative feeders. We found no differences among developmental times of larvae with or without food. These results contrast with those reported by Schausberger and Croft (1999) w ho found that larvae of 13 other phytoseiid mite species developed slower withou t food than those with food, regardless of the feeding type (non, facultative, or obligatory feeding types). However, we found that survival of A. largoensis larvae was higher with food than without, regardless of the type of diet. Direct observations of feeding and indirect indications of feeding (color ation of the alimentary tract) suggested that when food is present A. largoensis larvae tend to feed. The type of prey influenced the developmen t and survivorship of succeeding immature stages and the reproduction of adult A. largoensis In this study we tested four arthropod species that are commonly found inhabiting coconut leav es in Florida (Pea et al. 2009). Our results suggest that hemipteran prey alone ( A. orientalis and N. nipae ) do not provide enough resources to sustain A. largoensis populations. We found high mortalit y rates and longer developmental times of immature stages, and no reproduction of A. largoensis when feeding on either of these

PAGE 60

60 prey species. In contrast, A. largoensis had shorter developmental tim es and higher survival rates of immature stages, and the highest reproducti ve parameters, when feeding on the two mite species (R. indica and T. gloveri ). Our results suggest that mite s are better prey-resources for population growth of A. largoensis than the other prey used in this experiment. However, it is important to point out that under natural conditions, A. largoensis could potentially feed on various prey and alternative food s ources. Based on studies conducted using T. urticae McMurtry and Croft (1997) considered A. largoensis a type III generalist predator. This group of predators usually has a broad prey range including several mite and insect species. Potential mite prey includes tetranychids, eriophyids, tydeids, tenuipalpids and acaridids. Potential insect prey includes thrips, whiteflies, mealybugs, and scal e crawlers (McMurtry and Croft 1997). LawsonBalagbo et al. (2008) reported that A. largoensis was the most abundant predator found on coconut leaflets in several regions of Brazil, wh ere it was found associated with varied potential prey including several prostigmatid ( Tetranychus mexicanus (McGregor), T. neocaledonicus Andr, Oligonychus sp., Tenuipalpus sp., Brevipalpus phoenicis (Geijkes), Retracrus johnstonii (Keifer), Notostrix nasutiformes Gondim Jr, Flechtmann, Moraes, and Lorryia formosa Cooreman) and oribatid mites. Mo reover, Kamburov (1971) found that A. largoensis can develop and reproduce readily f eeding on three mite species ( B. phoenicis Eutetranychus orientalis (Klein) and Tetranychus cinnabarinus (Boisduval)) showing developmental times and oviposition rates similar to those found in our st udy. That same study reported poor development and reproduction of A. largoensis when fed on scale crawlers of Aonidiella aurantii (Maskell) and Chrysomphalus aunidum (L.) (Hemiptera Diaspididae), sim ilar to the results obtained here. A study conducted by Rodriguez and Ramos (2004) showed that A. largoensis can develop and reproduce feeding solely on the broad mite Polyphagotarsonemus latus (Banks) (Acari:

PAGE 61

61 Tenuipalpidae). In addition to having a broad prey ra nge it is possible that A. largoensis as seen in other phytoseiid species (Kamburov 1971; Van Rijn and Tanigoshi 1999), could use non-prey food sources. Some phytoseiids may use plan t exudates as well as honeydew as food supplements, which may increase their reproductive potential in the presence of prey (McMurtry 1992). Results of this study, and those reporte d by Yue and Tsai (1996), suggest that A. largoensis is not an obligatory predator because it can develop and reproduce on a pollen diet. Galvao et al. (2007) evaluated the survival of A. largoensis feeding on Aceria guerreronis Keifer (Acari: Eriophyidae) and other nonprey food sources (pollen and honey) and determined that a mixed diet of A. guerreronis or T. urticae plus either pollen or honey, improved its demographic parameters as compared to a diet of A. guerreronis alone. The broad food range of A. largoensis could be considered a desi rable attribute in terms of its potential control of R. indica because it suggests that A. largoensis can persist when R. indica densities are low or absent. The importance of this study lies in addressi ng the biological cont rol potential of an invasive species by A. largoensis Our results show that A. largoensis has some desirable attributes in terms of its potenti al as biological control agent of R. indica Amblyseius largoensis showed high survival rates, shorter developmen tal times and high reproductive rates when fed solely on R. indica compared with single other food sources However, it is important to point out that the colony used as source of A. largoensis was fed on R. indica and pollen, which could have resulted in lab selection affecting the ability of this predator to us e other food. Nevertheless, A. largoensis developed and reproduced feeding on T. gloveri showing some demographic parameters similar to those found when feeding on R. indica This suggests that A. largoensis retained the ability to use T. gloveri as prey despite any possibl e lab selection. Our findings suggest that A. largoensis has a high numerical response when feeding on R. indica which could

PAGE 62

62 explain previous reports s howing that populations of A. largoensis increased in numbers on coconut leaves afte r the arrival of R. indica to south Florida (Pea et al. 2009). A similar situation was observed in Puerto Rico and Trin idad and Tobago (Pea et al. 2009; Roda, pers. comm.). Another consideration, in te rms of the potential control of R. indica is that A. largoensis has a shorter developmental period than R. indica Our results suggest that A. largoensis can complete its immature development in less th an a week (5.92 0.67 d) at 26.5 1C, a much shorter developmental period than R. indica (24.5 d) at 23.9-25.7 C (Nageshachandra and Channabasavanna 1984). However, despite exhibiti ng attributes that are vital in terms of population growth, further studies are need ed to determine th e effectiveness of A. largoensis We are currently investigating the functio nal response and prey preference of A. largoensis It is possible that A. largoensis will be effective only at certain prey population densities, playing a complementary role to other mortality factors (other predators, acaropathogens, weather) that could act as regulators of R. indica populations. Moreover, A. largoensis could increase its reproductive potential in the presence of alternative foods. These aspects of A. largoensis biology are part of an on-going investigation because they might furnish a key to improving the control of R. indica

PAGE 63

63 Table 3-1. Duration and survivorsh ip of immature stages of A. largoensis fed on five diets and a no-f ood control treatment (26.5 1C, 70 5 % RH and 12:12 L:D photophase). Immature stage R. indica T. gloveri A. orientalis N. nipae Pollen No food Larva Duration (d)* 1.56 0.40 a 1.73 0.51 a 1.73 0.48 a 1.82 0.56 a 1.77 0.48 a 1.77 0.55 a Survival (%)** 97.5 a 97.5 a 98 a 92.5 a 92.5 a 55 b Protonymph Duration (d) 2.30 0.65 a 2.98 0.88 a 4.65 1.58 b 4.62 1.50 b 4.61 1.37 b 0 Survival (%) 90 a 82 a 75 a 22 b 91 a 0 Deutonymph Duration (d) 2.04 0.43 a 2.33 0.51 a 5.2 2.65 b 4 0.00 b 5.07 1.74 b 0 Survival (%) 97a 85a 52 b 25 b 94 a 0 Egg-Adult Duration (d) 5.92 0.67 a 7.11 0.94 a 11.43 3.27 b 13.5 0.00 b 11.51 2.11 b0 Survival (%) 85a 68a 38b 5c 70a 0 *(means SEM), means times (d) within a row followed by the sa me letter are not significantly different ( P< 0.05; ANOVA, Tukeys studentized range test). ** Percentage of surviving individuals within a row followed by the same lett er are not significantly different, ( P < 0.05; 2 test).

PAGE 64

64 Table 3-2. Fertility life table parameters a nd other reproductive parameters (duration of preoviposition and oviposition periods, total number of ovipositions, daily rate of oviposition, female and male longevity, sex ratios) of A. largoensis fed on three diets (26.5 1C, 70 5 % RH and 12:12 L:D photophase). R. indica T. gloveri Pollen Pre-oviposition period (d)* 3.18 1.74 a 3.38 1.50 a 9.12 2.62 b Oviposition period (d)* 14.09 8.30 a 8.69 5.05 a 10.12 9.00 a Total number of eggs deposited 19.96 14.6 a 6.82 6.05 a 3.00 2.86 b Daily rate of oviposition (eggs / d)* 1.63 0.27 a 1.13 0.39 b 0.54 0.32 c Female longevity (d)* 23.70 13.4 a 15.67 8.20 a 22.76 11.9 a Male longevity (d)* 16.44 6.65 a 9.00 4.18 a 8.72 4.52 b Female proportion of the progeny ** 0.73 0.19 a 0.63 0.23 a 0.67 0.22 a Net reproductive rate (females per female) Ro*** 12.59 1.46 a 5.66 0.78 b 3.90 0.30 b Generation time (d) T *** 19.95 1.55 b 17.01 1.19 b 27.74 1.54 a Intrinsic rate of increase (day-1) rm*** 0.127 0.008 a 0.102 0.06 b 0.049 0.002 c Means SEM within a row followed by the sa me letter are not significantly different *( P< 0.05; Tukeys studenti zed range test). **( P < 0.05;2-test) ***( P < 0.05; Student Newman-Keuls multiple range test)

PAGE 65

65 Figure 3-1 Daily oviposition rate of A. largoensis females fed on either R. indica, T. gloveri or Q. virginiana pollen at 26.5 1C, 70 5 % RH and 12:12 L:D photophase. Error bars represent the standard error of the mean. Figure 3-2. Survivorship of A. largoensis females fed on either R. indica, T. gloveri or Q. virginiana pollen at 26.5 1C 70 % 5 % RH and 12:12 L:D photophase. Error bars represent standard error of the mean.

PAGE 66

66 CHAPTER 4 PREY-STAGE PREFERENCES AND FUNCTIONAL AND NUMERICAL RESPONSES OF AMBLYSEIUS LARGOENSIS (ACARI: PHYTOSEIIDAE) TO RAOIELLA INDICA (ACARI: TENUIPALPIDAE) 3 Summary Raoiella indica Hirst (Acari: Tenuipalpidae) is a phyt ophagous mite that recently invaded the Western Hemisphere. This mite is a multivolti ne and gregarious species that can reach very high population densities and cause significant damage to various palm species (Arecaceae). The predatory mite Amblyseius largoensis (Muma) (Acari: Phytoseiidae) has been found associated with R. indica in Florida. This study evaluated A. largoensis for potential to control R. indica by (1) determining predator preferen ces among developmental stages of R. indica, and (2) estimating predator functional and numerical responses to varying de nsities of its most preferred prey-stage. Under no-choice conditions A. largoensis consumed significantly more eggs than other stages of R. indica In choice tests A. largoensis showed a preference for R. indica eggs over all other prey stages ( P <0.01). Amblyseius largoensis displayed a type II functional response showing an increase in number of prey killed with an increase in prey population density. Consumption of prey stabilized at approximately 45 eggs/day, the level at which oviposition by the predator was maximized (2.36 0.11 eggs /day; mean SEM). Results of this study suggest that A. largoensis can play a role in controlling R. indica populations, particularly when prey densities are low. Key words: mites, invasive species, Cocos nucifera Raoiella indica Amblyseius largoensis, prey-stage preference, functiona l response, numerical response. 3 Reprinted with permission from Exp Appl Acarol

PAGE 67

67 Introduction Raoiella indica Hirst (Acari: Tenuipalpidae), the re d palm mite, is a phytophagous mite that recently invaded the Western Hemisphere. This mite is a multivoltine and gregarious species that can reach very high population densities a nd cause significant damage to various plant species, especially palms (Arecaceae). Raoiella indica was detected in Martinique and St. Lucia in 2004, and rapidly expanded its geographical range throughout the Caribbean ( Kane et al. 2005; Etienne and Flechtmann 2006;Rodr igues et al. 2007). In December 2007, R. indica was detected in Florida in Palm Beach County (F DACS 2007) and since has spread to six other Florida counties. Raoiella indica also has spread to South Amer ica, reported from Venezuela (Gutirrez et al. 2007; Vsquez et al. 2008), Brazil (Marsaro et al. 2009) a nd Colombia (Carrillo et al. 2011b). More recently, the mite was found in Mexico (NAPPO 2009). Establishment of this exotic species in the Neotr opical region has given rise to con cerns about its potential effect on a vast number of economically and ecologically important plants. Efforts have been made to identify and evaluate the potential of nativ e predators to control R. indica in areas of invasion in th e Western Hemisphere (Pea et al. 2009; chapter 2). A predatory mite, Amblyseius largoensis Muma (Acari: Phytoseiidae), has been found as the most abundant predator associated with R. indica infesting coconuts in Fl orida, Puerto Rico, and Trinidad and Tobago. Preand postinfestation su rveys in Florida revealed that populations of A. largoensis increased after establishment R. indica (Pea et al. 2009). Furt her studies revealed that A. largoensis showed high survival rates, short developmental times and high reproductive rates when fed solely on R. indica compared with other single f ood sources (Carrillo et al. 2010, chapter 3). However, it is stil l unclear whether predation by A. largoensis results in reduction of R. indica populations. Data addressing that issue may be obtained th rough studies of functional response (response of individual predators to ch anges in prey density) and numerical response

PAGE 68

68 (change in density of the predat or population in response to changes in prey density) (Solomon 1949). The functional response of a predator as a function of prey density generally follows one of three mathematical models (Holling 1959a, 1959b, 1961) With a Type I functional response, the number of prey killed in creases linearly at constant rate as a function of prey density. With a Type II response the number of prey killed increases up to a maximum (predator saturation) but the proportion of prey killed declines with prey density. With a type III response, predation results in a sigmoidal curve. At low prey densit ies prey killed is positiv ely density-dependent as a result of an increase in the searching activity of the predator with increasing prey densities. However, at high prey densities predator saturation also occurs. Predators with Type III response are regarded as capable of re gulating prey populations (Holli ng 1965). Predators with type II response have proven to be efficient especially at low prey densities (Krebs 1978; Koehler 1999). The functional response of a predator is not fixed, but can vary depending on many factors, including the developmental stage of its prey (Santos 1975) the plant species upon which the interaction occurs (Skirvin and Fenlon 2001), and the spatial distribution of the prey (Ryoo 1996). This study was designed to assess the potential of A. largoensis to serve as a biocontrol agent for R. indica. The specific objectives of the inve stigation were to: (a) determine the preferences of A. largoensis for the different life stages of R. indica and (b) to estimate the functional and numerical responses of this predatory mite to varying densities of its most preferred stage of R. indica

PAGE 69

69 Materials and Methods Rearing and General Experimental Procedures Amblyseius largoensis and R. indica were obtained from laborat ory colonies maintained for approximately two years at the Tropical Res earch and Education Center (TREC), University of Florida, Homestead. Both mites were re ared at 26.5 1C, RH 70 5 % and 12:12 L:D photoperiod. Sources of colonies an d rearing procedures were previ ously described in Carrillo et al. (2010, chapter 3). All A. largoensis used in experiments were mated females in their peak oviposition period (3-6 days posteclosion), reared on R. indica (all stages), and starved for 24 hr prior to each experiment. Only replicates in whic h the female remained within the test arena and produced at least one egg within 48 hr were considered for analysis. Experiments were conducted at the same envi ronmental conditions used for rearing. The experimental arenas consisted of rectangl es (4 2.5 cm) cut from mature coconut ( Cocos nucifera L.) leaves without any damage or presen ce of any arthropod. Leaf rectangles were placed with the abaxial surface f acing up on cotton squares (6 6 2 cm) saturated with water in a plastic tray (hexagonal polystyrene weighi ng dishes 12.7/ 8.9 cm, Fisherbrand cat. No. 02202-103). Paper strips (Kimwipe, Kimberly Clark Corporation, Roswell, GA) were placed along the edges of the le af squares to minimize escape of m ites. Arenas were prepared over a period of 7 days. Eggs were obtained by placing 100 R. indica ovipositing females in each arena and allowing oviposition for 5 days. After this period females were removed and the desired number of eggs adjusted by removing excess of eggs with a fine brush. Arenas with only R. indica larvae, nymphs or adults were obtained by tr ansferring individuals from the stock colony into the arenas. Arenas with combinations of different stages were obtained by first allowing oviposition (on those treatments in cluding eggs), then transferring R. indica larvae, nymphs

PAGE 70

70 and/or adults, to adjust the desired ratios betw een the stages. A small plastic square (0.5 0.5 cm) was placed on top of each arena as a shelter for A. largoensis Raoiella indica Stage Preference by A. largoensis Prey preferences of A. largoensis among R. indica developmental stages were determined by choice and no-choice tests. Single A. largoensis females were transferred into arenas with specific densities of R. indica -stages and the numbers of each stage killed during a period of 24 h was recorded by counting the number of shriveled corpses and by subtracting the number of eggs remaining from the number of eggs provided. In no-choice tests, 30 R. indica specimens (eggs, larvae, protonymphs or adult females) per day were offered individually to A. largoensis females during two days. Females were transferred to new arenas with the corresponding R. indica densities for the second day of evaluation. The no-choice experiment was repl icated 15 times and analyzed through 1-way ANOVAs and means separated th rough Tukeys test (SAS 2003). Under choice conditions, the following R.indica -stage combinations were offered in separate arenas: 30 Eggs:30 Larvae, 30 Eggs :30 Nymphs, 30 Eggs:30 Adults, 30 Larvae:30 Nymphs, 30 Larvae:30 Adults, a nd 30 Nymphs:30 Adults. Choice e xperiments were replicated 15 times and the evaluation period for each test was 24 h. Preference was quantified using the index proposed by Manly et al. (1972): where N and N are the numbers of each prey-stage provided and Nc and N are the numbers of each prey-stage killed. The index assigns prefer ence values from 0 to 1, where 0.5 represents no

PAGE 71

71 preference. Mean -values were considered significant wh en 95% confidence intervals based on the t-distribution did not overlap with =0.5. Functional and Numerical Responses of A. largoensis to Varying Densities of R. indica Eggs Eight R. indica egg densities (5, 10, 20, 30, 40, 60, 80, and 100 eggs/arena) were offered to A. largoensis females on the arenas described above. A single A. largoensis female was transferred into an arena with a fixed R. indica egg-density and the numbers of eggs consumed during a 24-h period was recorded by subtracting the number eggs remaining from the number of eggs provided. Eggs were replenished ever y day for 4 consecutive days and oviposition by A. largoensis was recorded daily. The experiment was re plicated 15 times. Data from the first day were excluded from analysis to minimize potential effects from the prior diet of mixed stages. This allowed predators time to adapt physiolo gically to the new hom ogeneous diet and its relative level (Castagni oli and Simoni 1999). The type of functional response of A. largoensis preying on R. indica eggs was determined using a polynomial logistic regression (Systat Software Inc. 2006) as described by Juliano (2001). This procedure uses the signs of the coefficients in the equation Ne/N = a + bN + cN2 + dN3 + e where Ne is the proportion of killed prey and N the prey offered, as indicators of the shape of the functional response. In Type I responses the intercept ( a ) and the linear coefficient ( b) are >0; in type II responses the linear coefficient ( b) is <0 and the propor tion of prey killed declines with an increase in the number of prey offered. In type III responses the linear coefficient ( b) is >0 and the quadratic coefficient (d) is <0 (Juliano 2001). After determining the type of the functional response, three models were used to estimate the functional response parameters:

PAGE 72

72 Holling (1959) Fan and Petitt (1994) Rogers (1972) where Na is the number of prey consumed by each predator during time T (in our experiment T= 1 day), N is the initial prey density, is the attack rate coefficien t (proportion of prey captured by each predator per unit of searching time) and the handling time (time that the predator spends identifying, pursuing, killing, consuming and digesting a prey). The functional response models were fitte d using the non-linear regression procedure (NLIN; SAS Institute 2003). The numerical re sponse (variation in ra te of oviposition by A. largoensis as a function of prey density) was calcul ated through regressi on analysis using a hyperbolic model (Systat Software Inc. 2006). The hyperbolic model is described by the equation: where (y ) is the daily oviposition by A. largoensis at the various prey densities ( x ), ( ) is the maximum daily oviposition (i.e. plateau), and ( b) is the prey density needed to elicit half the maximum response. Results Raoiella indica Stage Preference by A. largoensis The no-choice experiments showed a significantly higher consumption of eggs than of other developmental stages of R. indica in the two days of evaluation (F = 111.2; df = 3, 57; P < 0.001, Figure 4-1). In addition, consumption of la rvae was significantly higher than consumption of nymphs or adults.

PAGE 73

73 The choice experiment indicated that A. largoensis showed a significant preference for R. indica eggs (95% confidence in tervals did not include =0.5 ) when compared to larvae, nymphs or adults. Preference for larvae over adults was al so significant (Figure 42). No preference was observed when larvae and nymphs or nymphs and adults were offered simultaneously (Figure 42). Overall, the results of the choice and no-choice experiments indicate that A. largoensis prefers R. indica eggs followed by larvae, over other developmental stages. Functional and Numerical Responses of A. largoensis to Varying Densities of R. indica Eggs The daily consumption rate of R. indica eggs gradually increased w ith an increase in prey density (Figure 4-3). Initially, th e response curve increased almost linearly with increasing prey density, and then began to level off as consump tion rate decreased. Nearly 100% of the prey was consumed at the lowest densities: (5, 10, 20 and 30 eggs/arena/day). Egg consumption by A. largoensis females tended to stabilize when more than 40 R. indica eggs/day were offered, reaching a plateau at approximately 45 eggs/day. The binomial logistic regression analysis used to determine the shape of the functional response resulted in the following co efficient estimates: intercept ( a ) = 0.22 0.005 (SE), t=40.2 P< 0.001; linear (b ) = -0.02 0.001, t=-22.41, P< 0.001; quadratic ( c ) =0.002 0.005, t=15.75, P< 0.001 and cubic ( d ) = -9.6 10 -7 7.64 10 -8, t=-12.6, P< 0.001. The linear coefficient ( b) was <0 indicating that A. largoensis exhibited a Type II functi onal response. The functional response data of A. largoensis preying on R. indica eggs successfully fitted to the three tested models (Table 4-1, Figure 4-3). The va lues of the attack rate coefficient and the handling time produced by each model are shown in Table 4-1. The numbers of eggs laid per day per predator (y ) and the prey density ( x ) were highly correlated ( R = 0.98, P < 0.001, Figure 4-4). Daily oviposition of A. largoensis females

PAGE 74

74 increased with an increase in the predation rate and tended to stabilized when prey consumption was greater than 30 R. indica eggs/day. The maximum daily oviposition ( ) of A. largoensis was 2.36 0.11 eggs (mean SEM) and the prey density needed to elicit half the maximum response ( b) was 15.2 0.11 R. indica eggs (mean SEM). Discussion The results of our study showed that A. largoensis females consumed more eggs than other stages when they were offered se parately, and preferred eggs of R. indica over other developmental stages when offered the choice. Amblyseius largoensis usually required a single attack attempt to successfully penetrate the chor ion and suck the egg contents. In contrast, the predator often needed several attack attempts to penetrate the mite cutic le and showed a longer handling time while preying on nym phs or adults. While feeding on nymphs and adults, the predator would often stop feeding, turn to cleanin g activities, then return to continue feeding on the same prey item, thus prolonging the hand ling time. The greater consumption of eggs compared to the other stages could also be ex plained by differences in the body weight of each stage. Because the body weight of the prey di ffers among the life stages, the maximum number of each prey stage which the predator can eat vari es as well. More important is the nutritional benefit that each stage provides for A. largoensis which could be indirectly estimated by comparing the fecundity of females fed separa tely on the different stages. That kind of experiment was not carried out in this study. However, the oviposition rate of females fed on an egg diet (2.36 0.11 eggs/ day) was higher than those observed in a previous study when a diet of mixed stages was provided (1.63 0.27, Carrillo et al. 2010, chapter 3), suggesting that a diet of eggs alone may result in a grea ter or at least similar fecundity than with a diet of mixed prey. Similar results were found with Neoseiulus fallacis (Garman); a diet of Tetranychus urticae Koch (Acari: Tetranychidae) eggs alone resulted in significantly higher fe cundity than a diet of

PAGE 75

75 mixed prey stages (Blackwood et al. 2001). However our results should be analyzed with caution as we did not evaluate the effect of long-term feeding on the diffe rent stages which could provide a more accurate estimate of the nutritional bene fits of each life stage (Castagnioli and Simoni 1999). Most studies on prey-stage preferences of phytoseiid mites have been conducted using tetranychid mites as hosts. In studies using T. urticae eggs and larvae as prey, Blackwood et al. (2001) suggested that type III generalist predat ors had no prey-stage preference or preferred larvae over eggs (Blackw ood et al. 2001). In our study A. largoensis, which was classified as a generalist type III predator by McMurtry and Croft (1997), showed a marked preference for R. indica eggs. These results suggest that prey-stage preference of A. largoensis varies depending on the prey species and is probably marked by inte ractions between the searching behavior of the predator, hunger level, prey protection mechanis m and nutritional value of each prey individual (Sandness and McMurtry 1970; Blackwood et al. 2001). For instance, the ability of A. largoensis to find and capture prey wa s affected by the webbing of Oligonychus punicae (Hirst.) (Sandness and McMurtry 1972); in contrast, R. indica females congregate and gradually form large groups of eggs that are exposed without any protection. Thus, R. indica eggs could be considered easily accessible prey for A. largoensis The duration of the egg stage is approximately 9 days, which is the longest immature stage of R. indica (Nageshachandra and Channabasavanna 1984), suggesting a greater availability of eggs compared to other immature stages. From a management point of view, a preference for R. indica eggs could be viewed as a desirable attribute of A. largoensis, because it kills the prey before it causes damage to the host plants. However, it is likely that other mortality factor s will be required to suppress R. indica Other natural enemies

PAGE 76

76 attacking mobile and reproductive stages of R. indica could be very useful for managing this pest. Predation and oviposition of A. largoensis increased as a function of an increase in the population density of R. indica Females consumed almost 100% of prey at relatively low prey densities (<40 eggs/day) suggesti ng a high searching ability of A. largoensis However, the proportion of prey killed decreased at high populations densities (>40 eggs/day) probably because of satiation or interference on their pred ation capacity related to prey density (Mori and Chant 1966; Sandness and McMurtry 1970). Amblyseius largoensis showed a functional type II response feeding on R. indica eggs which suggests that this pred ator could be more efficient at low to moderate prey densities, a common patte rn observed among many predatory mites of the group Gamasina (Koehler 1999) and other predat ors (Krebs 1978). In accordance with our results, Sandness and McMurtry (1970) studied the functional response of A. largoensis fed on O. punicae and found a type II response curve. Interestingly, the functional response curves of A. largoensis feeding on O. punicae and R. indica did not show a domed shape. The domed curve has been attributed to a confus ion or disturbance component at higher prey densities (Holling 1961; Mori and Chant 1966). For some predator s, high prey numbers may cause conflicting stimuli, blocking the feeding response, or causi ng disturbance resulti ng in repeated erratic movements. Amblyseius largoensis did not show signs of confusion or a reduction on the number of prey killed at high prey population densit ies which, similar to what was observed with O. punicae as prey (Sandness and McMurtry 1970), may explain why a domed curve does not occur. The type II functional response is the most common functional response of phytoseiid species (Sabelis 1985). A T ype II response was also observed in the phytoseiids: Euseius

PAGE 77

77 concordis (Chant) and Phytoseius floridanus Muma (Sandness and McMurtry 1970), Phytoseiulus persimilis Athias-Henriot Galendromus occidentalis (Nesbitt) and Neoseiulus chilenensis (Laing and Osborn 1974); Amblyseius longispinosus (Zhang et al. 1999); P. persimilis, G. occidentalis (Nesbitt), and Neoseiulus californicus (McGregor) (Xiao and Fadamiro 2010); Chileseius camposi Gonzlez y Schuster (Sepulveda and Carrillo 2008); Euseius (Amblyseius) finlandicus and Amblyseius andersoni (Koveos and Broufas 2000); Euseius hibisci (Chant) (Badii et al. 2004). Howe ver, all these examples used tetranychids as prey; in contrast, few studies have addressed the func tional response of phytos eiid mites preying on tenuipalpids. Reis et al. (2003) studied the functional response of Euseius alatus DeLeon and Iphiseiodes zuluagai Denmark and Muma on Brevipalpus phoenicis (Geijskes) and found a type II and type I functional responses, respectively. The predatory mite Amblyseius herbicolus (Chant) showed a type II func tional response when feeding on B. phoenicis (Reis et al. 2007). Euseius mesembrinus (Dean) showed the same type of response when feeding on Brevipalpus californicus (Banks) (Badii et al. 2004). Even though type II responses seem to be common among phytoseiid species it is impor tant to note that functional responses may change depending on the prey stage (Santos 1975), pl ant species (Skirvin and Fenl on 2001) and spatial distribution pattern of the prey (Ryo o 1996), among other factors. The best parameter estimate was obtained through the Fan and Petitt model (Table 4-1). Our results coincide with those observed by Sepul veda and Carrillo (2008) and indicate that the Fan and Petitt (1994) equation provides better parameter estimates. Based on our results, Hollings and Rogerss models unde restimated (i.e. generated lowe r values) the handling time of A. largoensis On the other hand, Rogerss model overe stimated the attack rate coefficient whereas Hollings model underestimated the same parameter.

PAGE 78

78 Results of the functional and numerical res ponses suggest that an optimum diet for A. largoensis where prey consumption and oviposition by A. largoensis were maximized, is approximately 45 R. indica eggs/day. The number of eggs laid/day by A. largoensis increased as a function of prey killed, with a positive and hi ghly significant correlation. Similar results were observed for N. fallacis (Smith and Newsom 1970) and for P. persimilis, G. occidentalis and N. chilenensis fed on tetranychids (Laing and Os born 1974). Our findings suggest that A. largoensis has a high numerical response when feeding on R. indica which could explain previous reports showing that populations of A. largoensis increased in numbers on coconut leaves after the arrival of R. indica in south Florida (Pea et al. 2009). A similar situation was observed in Puerto Rico and Trinidad and Tobago (Pe a et al. 2009; Roda, pers. comm.). A pattern in the coloration of the alimentary tract of A. largoensis was observed. The alimentary tract of A. largoensis turns bright red soon af ter it starts feeding on R. indica In our experiment, only females that ate more than 30 eggs per day retained the red coloration throughout the four days of evalua tion. In contrast, females fed w ith 20 eggs/ day or less turned red after feeding but lost the coloration by the following eval uation day. This suggests that predators showing red coloration, which are abunda nt in field samples (Carrillo, pers. observ.), must be repetitively feeding on a relati vely large number (>30 eggs/day) of R. indica This study was designed to gain insight into the potential capability of A. largoensis to control R. indica A previous study demonstrated that A. largoensis can complete development and reproduce when fed solely on R. indica, showing high survival rates, short developmental times and high reproductive rates compared with single other food sources (Carrillo et al. 2010, chapter 3). Another consideration is that A. largoensis has a shorter develo pmental period than R. indica The predator can complete its immature development in less than one week (5.92 0.67

PAGE 79

79 day, Carrillo et al. 2010), a much sh orter developmental period than R. indica (24.5 day) (Nageshachandra and Channabasavanna 1984). Resu lts of the present study provide a better understanding of the new associatio n between the Florida predator A. largoensis and the invasive R. indica The available evidence suggests that A. largoensis could play a role in reducing R. indica populations; however, results of the functional response study suggest that this predator could be more efficient at regu lating low prey population densities Nevertheless, it is necessary to consider the results with caution because all experiments have been conducted under simplified conditions using excise d leaves. In a whole plant syst em the predators can disperse and interact with other prey and predator communities, which could significantly affect the efficiency of this predator. Studies on the prey preference of A. largoensis and possible intraguild predation may contribute to understanding the potential of this pred ator as a biological control agent of R. indica In addition, field studies with known proportions of A. largoensis and R. indica are necessary to determine th e efficiency of this predator under more realistic conditions.

PAGE 80

80 Table 4-1. Parameters of th e functional response of A. largoensis feeding on R. indica eggs estimated by three functional response models. Attack rate coefficient (in units of the proportion of prey captured by each pr edator per unit of searching time) and handling time (in units of the proportion of 24hr exposure period). Model r2 P value Rogers 1.840.00390.95<0.001 Holling 1.020.006170.97<0.001 Fan & Petitt 1.550.0170.98<0.001 Figure 4-1. Daily consumption and oviposition by A. largoensis when offered R. indica stages in a no-choice condition. Error bars repres ent standard error of the mean.

PAGE 81

81 Figure 4-2. Preference of A. largoensis females for R. indica stages. Paired comparisons among R. indica stages. The preference index assigns preference values from 0 to 1, where 0.5 represents no preference. The position of in the bottom (0
PAGE 82

82 Figure 4-3. Functional response of A. largoensis to increasing densities of R. indica eggs estimated through three curve fitting m odels. Dots represent observed data.

PAGE 83

83 Figure 4-4. Daily oviposition of A. largoensis females as a function of the number of R. indica eggs consumed/day. Relation calculated through regression analysis using a hyperbolic model. Dots represent observed data. Error bars repr esent the standard error of the mean.

PAGE 84

84 CHAPTER 5 VARIABILITY IN RESPONSE OF FOUR POPULATIONS OF AMBLYSEIUS LARGOENSIS (ACARI: PHYTOSEIIDAE) TO RAOIELLA INDICA (ACARI: TENUIPALPIDAE) AND TETRANYCHUS GLOVERI (ACARI: TETRANYCHIDAE) EGGS AND LARVA4 Summary Raoiella indica (Acari: Tenuipalp idae) is a phytophagous mite th at recently invaded the Neotropical region. A predatory mite Amblyseius largoensis (Acari: Phytoseiidae) has been found associated with R. indica in Florida. This study evaluated A. largoensis by determining its likelihood of consuming eggs and larvae of R. indica and Tetranychus gloveri (Acari: Tetranychidae) under no-choice a nd choice conditions. To detect variations in the response of A. largoensis to R. indica four populations of predators were examined: (1) predators reared exclusively on R. indica in the laboratory for two y ears, (2) predators reared on T. gloveri in the laboratory for 2 months but reared on R. indica for two years previously, (3) predators collected from a field infested with R. indica and (4) predators collected from a field that had never been infested with R. indica Results of this study suggest that A. largoensis is likely to accept and consume high numbers of R. indica eggs regardless of their prev ious feeding experience. In contrast, all populations c onsumed relatively fewer R. indica larvae than the other prey tested. Predators previously exposed to R. indica were more likely to consume R. indica larvae. By contrast, predators not previously exposed to R. indica showed the lowest likelihood of choosing to feed on this prey item. Pl asticity in the response of A. largoensis to R. indica larvae could be associated with genetic selec tion, learning, or a combination of both. The possible implications of the observed differences in terms of biological control of R. indica are discussed. Key words: mites, invasive species, Cocos nucifera Raoiella indica Amblyseius largoensis, preypreference, nave predators, prey switch, learning, genetic selection. 4 Reprinted with permission from Biol Control

PAGE 85

85 Introduction Raoiella indica Hirst (Acari: Tenuipalpidae) is a phyt ophagous pest nativ e to tropical and subtropical areas of Asia that recently invaded the Neotropical region (Etienne and Flechtmann 2006; Vsquez et al. 2008; Marsaro Jr. et al. 2009). This mite is a multivoltine and gregarious species that can reach high populat ion densities and cause signifi cant damage to various plant species, especially palms (Arecaceae). Establishment of this i nvasive species in the Neotropical region has given rise to con cerns about its potential eff ect on several economically and ecologically important plants (C arrillo et al. 2011a, Appendix). Efforts were undertaken to id entify and evaluate potential biological control agents of R. indica in the Neotropical region (Pea et al. 2009; Carrillo et al. 2010, chapter 3). The predatory mite Amblyseius largoensis Muma (Acari: Phytoseiidae) ha s been found associated with R. indica in several areas of recent invasion including Florida, Puerto Rico, Trinidad and Tobago, Colombia, Cuba, and Mexico. Moreover, this predator has been found associated with R. indica in India (Taylor et al. 2011), Philippines (Galle go et al. 2003), Mauri tius and Dominica (Hoy, pers. comm.), and Benin and Tanzania (Zannou et al. 2010). It would be important to know whether all the predators identified as A. largoensis are equally efficient natural enemies of R. indica The geographically diverse dist ribution of this phytoseiid sugg ests that strains, biotypes, or cryptic species could exist (Noronha and Moraes 2004; Bowman 2010), and that differences in predatory behaviors among the various populations of A. largoensis could affect their propensity to feed on R. indica, as has been found for other phytoseiid species (Tixier et al. 2010). Information on actual predation by A. largoensis upon R. indica is limited to a single population of predators that was co llected from a coconut plantation located in one of the sites where R. indica was first detected in Florida in 2007. This population was used to initiate a

PAGE 86

86 laboratory colony which was used to study survival and reproduction (Carrillo et al. 2010, chapter 3), prey-stage preference, and functional and numerical responses of A. largoensis on R. indica (Carrillo and Pea 2011, chapter 4) These studies suggested that R. indica could be considered an appropriate prey for this population of A. largoensis in terms of its reproductive success and consumption rates. However, it is unknown whether other Florida populations of A. largoensis will respond differently to R. indica. This study was designed to determine th e likelihood of Florida populations of A. largoensis using the invasive R. indica as prey. Knowledge of the mechanisms that underlie prey acceptance and choice in predatory mites has rece ntly expanded. Studies on the predatory mites Hypoaspis aculeifer Canestrini (Acari: Laelapidae) and Phytoseiulus persimilis Athias-Henriot (Acari: Phytoseiidae) demonstrat ed that foraging traits and pr ey preferences are genetically determined and that genetic polymorphism ma y occur within local populations (Lesna and Sabelis 1999; Jai et al. 2002). In addition, ther e is increasing evidence suggesting that learning plays an important role in prey recognition and acceptance in phytoseiid mites (Rahmani et al. 2009; Schausberger et al. 2010) The likelihood of choosing R. indica as a prey could vary among local populations of A. largoensis as a result of genetically determined prey preferences, predator learning abilities, or a combination of both. This study evaluated four populations of A. largoensis from Florida by determining their likelihood of consuming R. indica when offered eggs and larvae of R. indica and Tetranychus gloveri Banks (Acari: Tetranychidae) under no-choice and choice conditions. To detect possible genetic or behavioral plas ticity in the response of A. largoensis to R. indica the four populations of predators had different feedi ng histories (experience feeding on R. indica ). Prey were chosen

PAGE 87

87 based on a previous study in which R. indica and T. gloveri were the most suitable prey for the above mentioned laboratory colony of A. largoensis (Carrillo et al. 2010, chapter 3). Materials and Methods General Experimental Procedures Experiments were conducted at 26.5 1 C, 70 5 % RH under a 12:12 L:D photoperiod. The experimental arenas consis ted of rectangles (4 2.5 cm) cut from mature unblemished coconut (Cocos nucifera L.) leaves. Leaf rectangles were placed with the abaxial surface facing up on cotton squares (6 6 2 cm) saturated with water in a plastic tray (hexagonal polystyrene weighing dishes 12.7/ 8.9 cm, Fisherbrand cat. No. 02-202-103). Paper strips (Kimwipe, Kimberly Clark Corporation, Roswell, GA) were placed along the edges of the leaf squares to minimize mite escape. A small plastic square (0.5 0.5 cm) was placed on top of each arena as a shelter for A. largoensis Prey Mites The two prey species, R. indica and T. gloveri were obtained from stock colonies reared on potted coconut palms kept in separate gr eenhouses following the procedure previously described (Carrillo et al. 2010). Eggs of bot h prey were obtained by placing 100 ovipositing females in each arena and allowing them to depo sit eggs for 5 d. After this period the females were removed and the desired number of eggs adjusted by removing excess eggs with a fine brush. In the case of T. gloveri, extensive webbing resulted from the crowded conditions of females on the ovipositing arenas. Such extensive webbing is not observed under natural field conditions, so a layer of webbing was carefully removed to expose T. gloveri eggs and allow more common conditions for A. largoensis to access its prey. Larvae of both prey species were transferred from the stock colonies into the aren as according to the desired prey ratios. Arenas

PAGE 88

88 with two prey species had an additional moistened paper strip on the mid vein of the leaf to allow separate placement of prey, but the strip was removed befo re releasing the predators. Predatory Mites Four populations of A. largoensis from Florida with disparate feeding histories (previous diet) were used. Two populations we re laboratory colonies that had been fed contrasting diets (following the procedures previously described, Carrillo et al. 2010, chapter 3). The first colony had been fed exclusively with R. indica for more than two years and will be referred to as the R colony This colony was initiated in November 2008 with 150 predators obtained from a coconut palm plantation (Malayan dwarf variety, pestic ide-free, 26.93 N, 80.82 W) that had been infested with R. indica since 2007. Individuals collected from the same site and confirmed as A. largoensis were added to the colony every six m onths. The second laboratory colony was initiated with 50 females obtained from the R -colony, which were subjected to a diet shift from R. indica to T. gloveri two months before testing, which w ould have allowed the predators to develop approximately 6 generations on the new diet. This colony will be referred to as the T colony The other two populations were collected from coconut palm plantations and maintained in the laboratory for one week befo re testing in arenas constructed with the same leaves and prey that were present at the time they were coll ected. The first field population was obtained from the same coconut plantation where two years before A. largoensis was obtained to initiate the R colony and will be referred to as the experienced field population because of its long-term previous exposure to R. indica although we cannot exclude the po ssibility that this population fed on pollens and T. gloveri as well. The second field population was collected from a pesticide-free coconut plantation (Malayan dwarf variety) that had never been infested with R. indica (25.49 N, 80.03 W, Tropical Research and Education Center, University of Florida, Homestead) and will be referred to as the nave field population for its lack of previous exposure

PAGE 89

89 to R. indica Most likely, it had fed on pollen, T. gloveri Oligonychus aff modestus and Oligonichus sp. (Acari: Tetranychidae); Nipaecoccus nipae (Maskell) (Hemiptera: Pseudococcidae); Aonidiella orientalis (Newstead) (Hemiptera : Diaspididae); and Aleurocanthus woglumi Ashby (Hemiptera: Aleyrodidae) (Pea et al. 2009; Carrillo et al. 2010, chapter 3). No-choice Tests No-choice tests were included to ensure that each population of predators could feed on R. indica and T. gloveri and to determine whether feeding e xperience and laboratory rearing could affect prey consumption. Predators from the four populations were offered se parately 50 eggs or 50 larvae of either R. indica or T. gloveri. Ten replicates per treatment (prey item) were used. In addition, absolute controls consis ting of arenas with prey but w ithout predators were included to ensure that mortality was caused by predation. Choice Tests Prey preference of the four populations was addressed by choice experiments offering 50 individuals of each prey item in the following combinations: R. indica eggs and T. gloveri eggs, R. indica eggs and T. gloveri larvae, R. indica larvae and T. gloveri eggs and R. indica larvae and T. gloveri larvae. Prey type(s) were randomly assigned to either side of the arena to discard any effect of prey location on the choice by A. largoensis Fifteen replicates per treatment (prey combination) were used, including absolute controls to ensure that prey mortality was caused by predation. Predators used in all experiments were A. largoensis mated females in their oviposition peak (3-5 days after adult emergence) that were starved for 8 hr before being transferred to the arenas. The numbers of each R. indica or T. gloveri stage consumed during a period of 24 h were recorded by counting the number of shriveled corpses (larvae) and by subtracting the number of

PAGE 90

90 eggs remaining from the number of eggs provided. Only replicates in which the female remained within the test arena and produced at least one egg within 48 hr were c onsidered for analyses. Data Analysis Data were separately analyzed for each e xperiment using SAS 9.2 (SAS Institute Inc.). Predation by the four populations of A. largoensis in the no-choice experiments was analyzed by Kruskal-Wallis tests due to variance heterogene ity and non-normality of data. Predation data under choice conditions were normally distributed (KolmogorovSmirnov P > 0.05) and had homogeneous variances (Levene test P >0.05) and were analyzed by two-tailed t-tests for each A. largoensis population and prey comb ination. The likelihood of choosing one prey item over another (preference) was quantified using the index proposed by Manly et al. (1972): where N and N are the numbers of each prey-stage provided and Nc and Nc are the numbers of each prey consumed. The index assigns preferen ce values from 0 to 1, where 0.5 represents no preference. Mean -values were considered significant wh en 95% confidence intervals based on the t -distribution did not overlap with =0.5. Preference () values were normally distributed (KolmogorovSmirnov P > 0.05) and had homogeneous variances (Levene test P >0.05) in the experiments involving choice between R. indica larvae and T. gloveri eggs or larvae. Effects of experience and laboratory r earing on prey preference () by A. largoensis were analyzed through ANOVAs and means separated by Tukeys test. Data of the experiment s involving choice tests between R. indica eggs and T. gloveri eggs or larvae were not norma lly distributed, so KruskalWallis tests were used in this case.

PAGE 91

91 Results Prey survival in the absolute controls was always greater than 95%, which ensures that mortality in the choice and no-choice experiments was caused by predation by A. largoensis females. No-choice Tests The four populations of A. largoensis consumed equivalent numbers of R. indica eggs, with consumption ranging from 44 to 47 eggs per female/day regardless of their previous feeding experience, [ 2 (3, N=40)= 3.90, P =0.27] (Table 5-1). By cont rast, the four populations consumed relatively fewer R. indica larvae (ranging from 9 to 23 larvae per female/day) compared with the other offered prey items and differed in their predation rates ( P < 0.001, Table 5-1). The nave field populati on, not previously exposed to R. indica, consumed significantly fewer R. indica larvae than the othe r three populations [ 2 (3, N=40)= 19.0, P <0.0001]. With respect to use of T. gloveri as prey, the four populations preyed on 39 to 44 T. gloveri eggs per female/day, showing no statistical diffe rences among the predator populations [ 2 (3, N=40)= 2.96, P =0.4]. Consumption of T. gloveri larvae ranged from 36 to 46 larvae per female/day. The R -colony, fed on R. indica for two years without exposure to T. gloveri prey, consumed significantly fewer T. gloveri larvae than the other three populations ( 2 (3, N=40)= 7.5, P =0.05). Choice Tests All four populations of A. largoensis consumed significantly more R. indica eggs than T. gloveri eggs or larvae (Figure 5-1A). C onsequently, preference indices ( ) were significant (not overlapping with =0.5) and favoring R. indica eggs without significan t differences among the four populations tested [ 2 (3, N=52)= 0.68, P =0.871) and 2 (3, N=58)= 6.54 P =0.08 for the R.

PAGE 92

92 indica eggs vs. T. gloveri eggs and R. indica eggs vs. T. gloveri larvae, respectively] (Figure 52). Despite the high rate of consumption of R. indica eggs, the likelihood that the four A. largoensis populations would consume R. indica larvae was generally lower and differed among them (Figure 5-1C & D). The two field populations consumed significantly more T. gloveri eggs than R. indica larvae, whereas both laboratory colonies consumed similar amounts of these prey items (Figure 5-1C). Preferen ce analysis showed that the R -colony, fed on R. indica for two years, was more likely to choose to prey on R. indica larvae than the other three populations tested, which showed significant preferences for T. gloveri eggs ( F =5.29; df= 3, 52; P <0.003, Figure 5-2). A significant eff ect of laboratory rearing on the likelihood of choosing R. indica larvae over T. gloveri eggs was observed ( F =9.08; df= 1, 52; P =0.004). Effects of experience on prey selection approached significant levels ( F =3.99; df= 1, 52; P =0.05). Finally, in the R. indica and T. gloveri larval-choice test, three populations of A. largoensis showed no significant differences in their cons umption rates or preference among these prey items (Figures 5-1D and 5-2). However, the na ve field population, not previously exposed to R. indica showed a significantly lowe r rate of consumption of R. indica larvae ( F =5.29; df= 3, 40; P =0.004) (Figures 5-1D and 5-2). Discussion Our experiments showed that populations of A. largoensis consistently fed on R. indica eggs but larvae of R. indica were less often consumed. Raoiella indica eggs were consumed in high numbers under no-choice conditions, and chosen over T. gloveri eggs and larvae by the four populations of A. largoensis in the choice tests, regardless of their previous fe eding experience. Nave predators, not previously exposed to R. indica responded to R. indica eggs similarly as the laboratory colony fed exclusively on R. indica for two years, suggesting th at the four populations

PAGE 93

93 of A. largoensis rapidly associated R. indica eggs with a reward and started to consume and prefer this prey item. Thus, the cues used by A. largoensis to recognize and associate R. indica eggs with a reward must be well defined and eas ily assimilated by the predator. Previous studies showed that oviposition rates of A. largoensis females (i.e., the R -colony) fed on a diet of R. indica eggs (2.36 0.1153 eggs/ day, Carrillo and Pe a 2011) were higher than oviposition rates when the predator was provided a diet of mixed R. indica stages (1.63 0.27 eggs/ day, Carrillo et al. 2010), suggesting that a diet of R. indica eggs is more rewarding in terms of the predators reproductive success than a mixed diet. A similar condition was observed with Metaseiulus occidentalis (Nesbitt) (Acari: Phytoseiidae), which s howed better reproductive parameters when fed on Tetranychus pacificus Mcgregor (Acari: Tetranychidae) e ggs compared to a diet of mixed immature stages (Bruce-Oliver and Hoy 1990). Preference of A. largoensis for R. indica eggs might be explained by its ability to recognize this prey item and associate it with high quality food. By contrast, the four A. largoensis populations differed in th eir likelihood of preying upon R. indica larvae. The no-choice experiments showed that A. largoensis generally consumed fewer individuals of R. indica larvae compared with the other offered prey items. The laboratory R -colony showed the highest likelihood of preying on R. indica larvae in the choice experiments and the lowest likelihood of consuming T. gloveri larvae in the no-choi ce experiments, which could have resulted from genetic se lection due to it having access only to R. indica for over two years (approximately 50 generations). Moreover, prey consumption analysis showed that laboratory colonies had a hi gher likelihood of consuming R. indica larvae over T. gloveri eggs than the two field populations. Thes e results suggest that laboratory colonies could have been genetically selected to prey on R. indica and that this genetic cha nge was not eliminated by two

PAGE 94

94 months of rearing on T. gloveri in the T -colony. For instance, the th ree populations that had previous exposure to R. indica ( R -colony, T -colony and the experi enced field population) consumed significantly more R. indica larvae than the nave fiel d population with no previous exposure to R. indica The nave field population showed the lowest likelihood of consuming R. indica larvae both in choice and no-c hoice experiments, suggesting that its lack of experience in preying on R. indica might have affected its ab ility to consume this prey. Our experiments did not test R. indica nymphs and adults as prey. However, previous studies indicated that female A. largoensis (the R -colony) had difficulty handling nymphs and adults and consumed significantly more R. indica eggs and larvae than nymphs and adults (Carrillo and Pea 2011, chapter 4) While preying on nymphs and adults, predators would often stop feeding, turn to cleaning activities, then retu rn to continue feeding on the same prey item, thus prolonging the handling time. Consequently one could assume that the likelihood of consuming R. indica nymphs and adults would have been lower than the likelihood of consuming R. indica eggs or larvae observed in this study. Overall, the results of our studies suggest th at multiple mechanisms can be involved in the response of A. largoensis to R. indica It is possible that these mechanisms do not act independently but are conditionally expressed depending on the conditions that the predator faces (Sabelis and Lesna 2010). For instan ce, it is likely that populations of A. largoensis are subject to relatively long periods of exposure to R. indica, which has become by far the dominant arthropod on coconuts in areas of recent inva sion. Under these circumstances, genetically determined prey preferences could be selected and the likelihood of A. largoensis preying on R. indica active stages may increase. Simultaneously, experience d predators could learn to optimize their foraging activities and also increase the likelihood of preying on R. indica larvae However,

PAGE 95

95 when food quality declines and predators are forced to disperse, dispersal might bring A. largoensis to a location where another food type is dom inant. Generalist predators must retain the ability to switch prey (Sabelis an d Lesna 2010). Our results suggest that A. largoensis can switch to novel prey after a peri od of starvation despite possible conditioning of prey preferences due to genetic selection or experience. In conclusion, A. largoensis showed plasticity in its use of R. indica larvae, which could be related to genetic selection, learning or a combination of both. Moreover, the environmental conditions that pred ators face can modulate the expression of these. Another factor that could affect the prey pref erence of phytoseiid mites is the presence of feeding attractants and deterrents (Hoy and Sm ilanick 1981; Vet and Di cke 1992). It has been suggested that R. indica could produce repellent compounds b ecause of the presence of droplets located at the tips of the dorsal body setae and at the tip of the eggs pedicel. The marked preference for R. indica eggs observed in the present study a nd the high profitability of this prey for A. largoensis (Carrillo et al. 2010, chapter 3) do not support the hypothesis of toxic or repellent compounds being pres ent in the droplets of R. indica eggs. However, it is possible that the droplets found on other R. indica larvae (and other life stag es) could have a different composition and affect the ability of A. largoensis to feed on them. The possible existence of kairomones or allomones produced by R. indica together with herbivor e-induced plant volatiles that may alter the searching behavior of A. largoensis should be investigated to better understand their role on the prey choice of this predator. In addition, mites can have other defense mechanisms that could directly influen ce their attractiveness to predators. The hardness of the cuticle could serve as a defense mechanism against pred ators (Alberti and Crooker 1985). Observations made during the experiments showed that A. largoensis usually required a single attack to successfully penetrate the chorion and imbibe all of the egg contents. In contrast, A.

PAGE 96

96 largoensis females probed several times in order to penetrate the larval cuticle, had a longer handling time, and did not consume the larv al prey entirely. Thus, preference of A. largoensis for R. indica eggs and the lack of preference for larvae could be related to the energy spent and the nutritional reward obtained wh en feeding on each of them. Findings of this study suggest that disparate populations of A. largoensis may be responding to the invasion by R. indica either by learning or evolving genetically to be better predators of this invasive pest These findings may be important for future biological control programs targeting R. indica Since R. indica gained importance as an invasive pest in the Neotropics, surveys for natural enemies in several places of the world (Roda et al. 2008; Pea et al. 2009; Ramos et al. 2010; Zannou et al. 2010; Carrillo et al. 2011; Taylor et al. 2011; Hoy, pers. comm.) have identified A. largoensis as the most abundant pr edator, and often the only phytoseiid species associated with R. indica Further studies are needed to determine the efficacy of these predator popula tions in controlling R. indica, and whether there are additional differences in populations of A. largoensis with regard to their use of R. indica

PAGE 97

97 Table 5-1. Predation of R. indica and T. gloveri eggs and larvae by four local populations of A. largoensis with disparate feeding history (pre vious diet) under no-choice conditions. Mean number prey items consumed in 24 hrs Std. Dev. Amblyseius largoensis Population Number of individuals consumed/day R. indica eggs R. indica larvae T. gloveri eggs T. gloveri larvae R -colony 47.2 2.7 a 23.4 6.7 a 39.4 8.2 a 35.5 11.1 b Experienced field 44.4 4.2 a 24.3 7.1 a 40.5 3.9 a 41.7 6.4 a T -colony 46.6 3.5 a 21.7 5.4 a 42.6 5.8 a 46.3 2.3 a Nave field 45.5 3.0 a 9.2 4.8 b 43.7 5.0 a 41.3 9.6 a 2 (3,N=40) 3.9 18.9 2.96 7.5 P value 0.3 <0.001 0.4 0.05

PAGE 98

98 Figure 5-1. Percentage of individuals of R. indica (black bars) and T. gloveri (grey bars) consumed by four populations of A. largoensis with disparate feeding histories (previous diet) under choice conditions. (A) R. indica eggs vs. T. gloveri eggs (B) R. indica eggs vs. T. gloveri larvae (C) R. indica larvae vs. T. gloveri eggs (D) R. indica larvae vs. T. gloveri larvae.* Represents signific ant differences in two-tailed t -tests ( P<0.05) for each A. largoensis population and prey combination.

PAGE 99

99 Figure 5-2 Prey preference of f our local populations of A. largoensis with disparate feeding history (previous diet) when preying on R. indica and T. gloveri eggs and larvae under choice conditions. The preference index assigns preference values from 0 to 1, where 0.5 represents no preference. Mean -values we considered significant when 95% confidence intervals (error bars) based on the t-distribution did not overlap with =0.5. Differences in -values among the four populations of A. largoensis represented by letters were analyzed through ANOVAs or Kruskall-Wallis tests depending on normality of data.

PAGE 100

100 CHAPTER 6 EFFECT OF AMBLYSEIUS LARGOENSIS (ACARI: PHYTOSEIIDAE) ON RAOIELLA INDICA (ACARI: TENUIPALPIDAE) USING PR EDATOR EXCLUSION AND PREDATOR RELEASE TECHNIQUES Summary Exclusion and release tactics were used to obt ain coconut palms with disparate levels of A. largoensis in order to quantify their effects on R. indica densities. Four treatments consisting of a range of A. largoensis release rates (0= control, 1:10, 1:20 and 1:30 A. largoensis : R. indica ) were tested. The release of A. largoensis resulted in a significant reduction of R. indica densities and less leaf area damaged in the coconut palms compared to the controls. The largest pest density reduction (~92%) was observed at th e highest predator release rate (1:10 A. largoensis : R. indica ). The other two release rates (1:20 and 1:30 A. largoensis : R. indica ) caused significant and equivalent reductions in pest densities (5 5and 43%, respectively). Results of this study support the hypothesis that A. largoensis is an important mortality factor of R. indica and should be considered as a key biological co ntrol agent in IPM programs targeting R. indica. Key words: mites, Cocos nucifera invasive species, Raoiella indica Amblyseius largoensis, exclusion, predator release, pest densities, damage. Introduction The arrival and establishment of Raoiella indica has had a serious economic effect on coconut production and the nursery palm industry in the Caribbean, Florida and other sites in the Neotropics (Roda et al. 2008; Carrillo et al. 2011a, Appendix). Efforts have been made to identify natural enemies with potential in biological control and IP M programs targeting R. indica The predatory mite Amblyseius largoensis (Muma) (Acari: Phytoseiidae) responded numerically to the arrival of R. indica in Florida (Pea et al. 2 009), which motivated detailed studies on the efficiency of this predator. An initial study determined that this predator was able

PAGE 101

101 to feed, develop and reproduce on a diet consisting solely of R. indica showing better reproductive parameters than when feeding on othe r prey and pollen (Carr illo et al. 2010, chapter 3). Follow-up studies determined that A. largoensis responded both functio nally and numerically to increasing densities of R.indica, and showed a marked pref erence for eggs over larvae, nymphs and adults of the phytophagous mite (Ca rrillo and Pea 2011, chapter 4). Further studies examined the response of four populations of A. largoensis with disparate previous exposure to R. indica and Tetranychus gloveri (Acari: Tetranychidae). The four populations of A. largoensis, including predators never exposed to R. indica were likely to accept and consume high numbers of R. indica eggs. However, predators previously exposed to R. indica were more likely to feed on R. indica larvae than were nave predators. The combined previous studies (Chapters 2-5) indicate that A. largoensis is actively responding to the invasion by R. indica and provide a framework to hypothesize that this predator has potential to be used in biological control and IPM programs targeting R. indica The objective of this study was to provide a quantitative evaluation of the ability of A. largoensis to reduce R. indica densities at four predator-prey ratios. Materials and Methods The experimental approach combined ex clusion and release techniques to obtain differential levels of A. largoensis on palms infested with R. indica. The experiment was conducted inside a c limate-controlled 43.2 m2 glasshouse house (26.5 4C, RH 70 20%) located at the Tropical Resear ch and Education Center in Homestead, FL (25.49 N, 80.03 W). Thirty-two 1-year-old potte d, pest-free and unsprayed Mala yan dwarf coconut palms (about 1.5 m in height) were used in this experiment. A middle frond in each palm was selected as the experimental unit and remaining fronds were manually removed. A yellow tagging tape (~5 cms wide) coated with Tanglefoot was tied around the base of th e frond to exclude any crawling arthropods. Palms were arranged on benches and clear plastic sheets (1.5 3 m) were hung from

PAGE 102

102 the roof of the glasshouse to isolate each indi vidual palm. Fronds were then inspected with a hand lens (20 ) and any arthropod [ Aonidiella orientalis Newstead (Hemiptera: Diaspididae) Tetranychus gloveri Banks (Acari: Tetranychidae), A. largoensis ] was removed by hand every three days during 2 consecutive weeks before being inoculated with R. indica. Raoiella indica specimens used for infestation were obtained from a red palm mite colony stock (Carrillo et al. 2010). Two R. indica -infested pinnae were then attached to the abaxial surface of each palm frond with a hair clip. Infested palms were le ft undisturbed for 30 days. After this period, R. indica establishment was evaluated by counting the number of motile stages and eggs per pinna. Treatments were allocated to each palm de pending on the degree of infestation. Four A. largoensis release rates (0= cont rol, 1:10, 1:20 and 1:30 A. largoensis : R. indica ) were tested with 8 replicates per treatment. Predators we re obtained from a coconut palm plantation (Malayan dwarf variety, pesticide-fr ee, 26.93 N, 80.82 W) infested with R. indica since 2007. Predators were confirmed as A. largoensis and maintained on re aring arenas under laboratory conditions for one week before tes ting. Predators were then transferred from the rearing arenas to the experiment al coconut fronds using a camel hair brush. After treatment, all fronds were carefully inspected every two w eeks with a 20 hand lens and any undesirable predators or phytophagous arthropod s removed. Three months later, fronds were excised, taken to the laboratory, and the total number of R. indica and A. largoensis eggs and motiles per pinna were inspected under a dissecting microscope (5 0). The surface area of each of the pinnae was estimated by measuring its width and height a nd calculating the area usin g base height in order to account for the triangular shape of the pinnae. Leaf dama ge was quantified by measuring the width and height of necrotic areas and calculated in the same manner.

PAGE 103

103 Statistical Analysis: Raoiella indica and A. largoensis density values were normally distributed (Kolmogorov-Smirnov P > 0.05) and had homogeneous variances (Levene test P >0.05). The effect of releas ing rates on density of R. indica eggs and motiles per pinna was analyzed through covariance analysis and Tuke y separation tests (Proc ANCOVA SAS Inc.). The number of predators released per plant was used as the co variate and the analysis was weighted with the number of pinnae per palm. Da ta on the leaf area dama ged (showing necrosis) were not normally distributed, so Kruskal-Wallis tests were used in this case. Results One month after inoculation, R. indica mean infestation levels ranged from 41.3 14.6 to 46.6 16.5 (Mean SEM) R. indica motile stages per pinna (Table 6-1). Random allocation of treatments ensured similar infestation levels betw een treatments before predators were released (df= 3, 8; F = 0.15; P = 0.92). The number of predators rel eased ranged from 0 (control treatment) to 4.1 1.1 A. largoensis females per pinnae according to the release rates and the infestation level (Table 6-1). Three months after predators were released, R. indica mean infestation levels ranged from (124.7 94.1 to 999.6 94.6) R. indica motile stages per pinna (Table 6-1, Figure 6-1). Although release of A. largoensis females did not resu lt in decline of R. indica initial infestation levels, the release of A. largoensis resulted in a significant reduction in R. indica densities relative to the densities fo und on predator-free palms (df= 4, 28; F = 7.76; P < 0.001). Raoiella indica reached 70.7 7.2 individuals/ cm2 (mean SEM) in predator-free palms, and 5.3 7.1, 31.7 5.7 and 40.4 5.9 (individuals/ cm2) when A. largoensis was released at three rates of 1:10, 1:20 and 1:30 A. largoensis : R. indica resulting in a 92, 54 and 42 percent reduction of R. indica densities, respectiv ely. The highest R. indica density reduction was observed at the highest predator release rate (Figure 6-2).

PAGE 104

104 More motile stages of R. indica were present than eggs on the two treatments with lower A. largoensis release rates and on the predator-free palms. In contrast, R. indica populations consisted of more R. indica eggs than motile stages in the treatment with the highest A. largoensis release rate (Figure 6-2C). Motile stage densities in predator-free palms (51.5 5.9 R. indica motiles / cm2, mean SEM) were significantly hi gher than those found in the release treatments (2.0 5.8, 20.1 4.7 and 26.5 4.9 R. indica motiles/cm2, mean SEM)), resulting in 96, 61 and 48 percent reduction of R. indica motile stages in the 1:10, 1:20 and 1:30 A. largoensis : R. indica release rates, respectively (df= 4, 28; F = 6.87; P < 0.001) (Figure 6-2B). Raoiella indica egg densities on predator-free palms (19.3 2.2 individuals/ cm2, mean SEM) were significantly higher than those found in pa lms where predators were released (3.3 2.2, 11.6 1.7 and 13.9 1.8); predator re leases resulted in an 82, 39 and 27 percent reduction in R. indica egg densities in the three A. largoensis/ R. indica release rates, resp ectively (df= 4, 28; F = 5.13; P< 0.01) (Figure 6-2A). Amblyseius largoensis established on all palms where it was released and was not detected in the control treatments. However, the numbers of A. largoensis recovered in the final destructive sampling (0.25 0.8, 0.36 1.4, 0.4 1.3 a nd 0 0 predators per/pinnae, in the 1:10, 1:20 and 1:30 A. largoensis : R. indica release rates, respectively) were lower than the numbers released (Table 6-1), and did not vary along the release treatments (df= 4, 28; F = 10.04; P < 0.001). Besides A. largoensis, no other predators were detected. The leaf area per pinna and the number of pinnae per palm were equivalent among the palms assigned to each A. largoensis release rate (F = 0.69 P =0.56 and F = 0.11.4 P =0.95, respectively). In contrast, the percentage of l eaf area showing necrosis was significantly higher

PAGE 105

105 in palms where A. largoensis was excluded compared to palms where the predator was released [ 2 (3, N=32)= 99.83, P <0.0001]. Discussion Convincing evidence of the efficacy of a na tural enemy requires not only quantitative evaluations of pest densities with and without the presence of the natural enemy, but also needs to differentiate the mortality caused by the natura l enemy from other biot ic and abiotic factors (Luck et al 1988; Van Driesche a nd Bellows 1996). In our experiment the use of exclusion tactics (sticky barriers, hand remova l and cages) was used to quantify R. indica densities in palms with different ratios of A. largoensis including predator-free pa lms. No other predators were detected and palms were kept under cont rolled environmental conditions, which suggest that the observed differences in R. indica densities could only be attributed to A. largoensis Previous studies indicated that A. largoensis had a marked preference for R. indica eggs and difficulties feeding on motile stages (Carrill o and Pea 2011, chapter 4). In this experiment, pest density reductions we re observed both in the e gg and motile stages of R. indica However, the proportion of eggs and motile stages in the treatment with the highest release rate, where the highest pest reduction was observe d, differed from all other treatments. This could have been caused by a higher predation of eggs in this treatment that resulted in fewer individuals developing into motile stages. By contrast, the treatments with the two lowest release rates and the predator-free treatment resulted in an accumu lation of motile stages and substantially higher prey densities. It is possible that at low predator densities more R. indica individuals completed the egg stage and became larvae, a stage when they are less likely preyed upon by A. largoensis. These results suggest that A. largoensis can be effective in controlling R. indica only when it consumes a large proportion of R. indica eggs and reduces the accumulation of motile stages. Moreover, the ability of A. largoensis to suppress R. indica might depend not only on the

PAGE 106

106 predator prey ratio but also on the population density of R. indica It is possible that as for other generalist predators (Jam es 1990; McMurtry 1992), A. largoensis could have its major effect at low population densities of R. indica Predation on R. indica by A. largoensis resulted in a re duction of the coconut leaf area showing damage. The proportion of damaged leaf tissue differed between treated and untreated palms, but not among the three release rates test ed. Predator-free palms showed necrosis four months after being inoculated with R. indica Therefore, the experiment was concluded at that time in order to prevent advanced necrosis and subsequent reduc tion of host plant quality, which would eventually affect R. indica survivorship. In conclusion, our results suggest that A. largoensis can reduce R. indica densities and the damage in flicted to coconut fronds under greenhouse conditions. Although A. largoensis caused a significant reduction in R. indica densities relative to the predator-free palms, predators were unable to eliminate R. indica during the period evaluated. It is possible that a higher number of predators is required to suppress R. indica or that a longer time could have resulte d in elimination of R. indica The highest pest reduction (~92% of total density) in our experiment was observe d at the highest release rate (1:10 A. largoensis: R. indica ), which suggests that inundative releases at higher re lease rates than the ones used in this experiment could result in larger pest density reductions a nd better control of R. indica The use of exclusion tactic s was useful to quantify R. indica densities in palms with different levels of A. largoensis An alternative to exclusion by physical means used in our experiment is the use of acaricides to reduce m ite numbers (Braun et al. 1987; McMurtry et al. 1992; Cuthbertson et al. 2003). Some chemicals are highly toxic to predators but less toxic to phytophagous mites (Roush and Hoy 1978) so they can be used for predator exclusion purposes.

PAGE 107

107 However, the use of pesticides has often led to problems of interpreting results because of possible pesticide-induced physio logical effects on th e plants (Jones et al. 1983), pesticideinduced sex ratio bias and s timulation of the reproductive po tential of the prey population (hormoligosis) (Bartlett 1968; Dittrich et al 1974; Hoy et al. 1979; Maggi and Leigh 1983; Cuthbertson et al. 2003). Despite these problems, acaricidal disruption is a quick and easy way to evaluate the effect of natural en emies at a large scale (Luck et al. 1988). Thus, the feasibility of using this technique to test the effects of different of A. largoensis release rates on R. indica populations under natural conditions should be investigated. Our experiments showed that A. largoensis is an important mortality factor of R. indica and its effects can reduce damage on coconut leaves caused by this pest. However, additional mortality factors may be needed to control R. indica populations, especially mortality factors affecting motile stages of R. indica

PAGE 108

108 Table 6-1. Initial mean R. indica infestations and A. largoensis release rates on coconut palms and mean R. indica infestations and number of A. largoensis recovered three months after predator release (26.5 4C, RH 70 20%) in a climate-cotrolled glasshouse at TREC, Homestead, FL. Treatment release rates A. largoensis : R. indica Infestation level before predator release Infestation level and predators recovered R. indica motiles/pinna (mean SEM) A. largoensis released/pinna (mean SEM) R. indica motiles/pinna (mean SEM) A. largoensis recovered/pinna (mean SEM) 1 : 10 41.3 10.9 4.1 1.1 124.7 94.1 0.3 0.1 1 : 20 46.6 11.6 2.3 0.6 376.7 .4 0.3 0.1 1 : 30 41.5 7.50 1.3 0.2 546.7 .1 0.4 0.1 0 : X(Control) 43.3 10.9 0.0 0.0 999.5 .5 0.0

PAGE 109

109 Figure 6-1. Effect of fo ur release rates of A. largoensis (0= control, 1:10, 1:20, 1:30 A. largoensis : R. indica ) on R. indica motile stage densities (mean SEM represented by the error bars).

PAGE 110

110 Figure 6-2. Effect of fo ur release rates of A. largoensis (0= control, 1:10, 1:20, 1:30 A. largoensis/R. indica ) on R. indica densities. A. R. indica egg density B. R. indica motile stages density, C. R. indica total densityshowing th e proportion of eggs and motile stages. Error bars represent standard errors.

PAGE 111

111 CHAPTER 7 CONCLUSIONS The objective of this dissertation was to determine the role of key mortality factors affecting R. indica within the natural enemy complex that inhabits coconut palms in Florida. Previous reports (Pea et al. 2009) and the inform ation presented here i ndicate that 9 predator species have been reported feeding on R. indica in Florida. However, with the exception of A. largoensis, all other predators are fo und only occasionally, in low numbers and depend on other prey species (Chapter 2). The studi es presented here suggest that A. largoensis is the most important biotic factor causing mortality of R. indica in Florida. The first study dealt with the development and reproduction of A. largoensis feeding on R. indica, pollen, and on different potential prey. In c ontrast to the results of tests with other predators, the study showed that R. indica is a good prey for A. largoensis. Predators fed on a R. indica diet had faster development and greater intrin sic rate increase than those fed on other prey or pollen, suggesting that the obs erved relationship between these two species could be important in terms of biological control of R.indica The second study of prey stage-preferences and functional and nume rical responses of A. largoensis to R. indica density, was useful to gain insight on the predator-p rey interactions between these two species. The study showed that A. largoensis has a marked preference for R. indica eggs over other developmental stages, which ha s implications for the effectiveness of the predator. By consuming primarily eggs, the predator kills its prey before it can cause injury to the host and before it reproduces (Huffaker et al 1970). In contrast, it was observed that the predator had difficulties preying on the active stages of R. indica, indicating that other mortality factors are necessary to hold in check the reprod uctive stages. The second part of the study also showed that A. largoensis is capable of responding both func tionally and numerically to changes

PAGE 112

112 in the population density of R. indica These results signaled that A. largoensis could prey on R. indica in a density dependent manner by increasing prey consumption and their own reproduction in response to incr easing prey density, an important attribute for potentially effective natural enemies (Rosen and Huffaker 1983). The third study explored differences in the use of R. indica by four populations of A. largoensis that differed in their previous exposure to R. indica This study revealed that all predators, including those with no previous exposure to R. indica had a high likelihood of consuming and preferring R. indica eggs over eggs and larvae of T. gloveri However, it also showed that populations of A. largoensis varied in their consumption of R. indica motiles. Predators with previous exposure to the inva sive species were more likely to consume R. indica larvae than nave predators. Over all, the findings of this study suggest that disparate populations of A. largoensis may be responding to the invasion by R. indica either by learning or evolving genetically to be better pred ators of this invasive pest. Finally, the fourth study using predator excl usion and release tec hniques proved that A. largoensis can cause significant reduction in pest de nsities and can be considered an important mortality factor of R. indica in Florida, and also contributes to a reduction of the damage to coconut caused by this pest. However, additional biotic mortality factors may be needed to suppress the high populations of this pest observed after its recent invasion of Florida. One method to achieve greater pest suppressi on could be through augm enting the effect of biotic factors that are already acting over R. indica in Florida. For instance, besides A. largoensis, the other natural enemy showing some potential as predator of R. indica is the lacewing species Ceraochrysa claveri Navs (Neuroptera:Chrysopida e). Although this predator depends on other prey to complete development, it has been repeatedly found associated with R.

PAGE 113

113 indica It has been observed that its first larval instars are vor acious predators of this pest. Methods of increasing the effects of A. largoensis and C. claveri on R. indica merit investigation. Techniques such as augmentative release, use of predator attractants and provision of alternative food sources for increasing predator populations in R. indica infested areas s hould be explored. In addition, research towards identifying chemical control practices that can preserve these natural enemies for their use in local IPM plans targeting R. indica should be conducted. An alternative to biological control with the natural enemy fauna pres ent in Florida is the introduction of natural enemies in a classical biological control approach. Considering that the natural enemy complex of R. indica in Florida is composed of onl y generalist predators it will be desirable to find natural enemie s more narrowly specific to R. indica However, until now there are no reports or evidence of the existence of more effective predators in any place recently surveyed for natural enemies of R. indica. Other geographical areas should also be searched for natural enemies of this pe st. The high diversity of Raoiella species in the Middle East region (Chapter 1) suggests the species could have originated in this region and, theoretic ally, its most effective natural enemies could be found there. Ne vertheless, results from one of the studies of this dissertation suggest that a long time of asso ciation between A. largoensis and R. indica could result in more aggressive strains of this pred ator. The possible existence of more aggressive strains, biotypes or cryptic species of A. largoensis from other parts of the world that are reproductively isolated from the A. largoensis already present in Florid a should be investigated. Moreover, the recent report of four speci es of acaropathogenic fungi attacking R. indica in Puerto Rico (Rodrigues and Col on, unpublished) is promising and deserves special attention. The tropical conditions and high R. indica populations found in areas of recent invasion could favor

PAGE 114

114 epizootics caused by acaropathogenic fungi which could cause a decline in R. indica densities and possibly enhance the regulatory capacity of A. largoensis and other predators. In closing, an important consideration for this research is that a ll studies were conducted excluding abiotic factors that could cause mortality on R. indica In natural conditions populations of R. indica will be affected not only by natural enemies, but also by factors such as temperature and precipitation (Moutia 1958; Sakar and Somchoudhury 1989). For instance, R. indica infested areas in three site s in Florida (Palm Beach, Brow ard and Miami-Dade counties) were sampled from January 2008 to March 2010. At all three sites the populations were highest in the first four months after the initial in festation and a steady ne gative trend on population density was observed ther eafter. The decline in R. indica densities could be related to the subtropical conditions of the state as well as the build-up of predators, primarily A. largoensis (Duncan et al. 2010; Pea, unpublished). As seen in other invasive species, the rates of population growth can vary markedly during the early stages of invasion (Crooks and Soule 1999). It is possible that the high rates of population growth that R. indica has displayed in its early stages of invasion will change over time when other mortality factors present in the environment affect this invasive species. It will be important to quantify the effects of abiotic factors on R. indica. Moreover, the complementarily or an tagonisms between abiotic and biotic factors affecting R. indica should be investigated to better understand the populat ion dynamics of this invasive pest in Florida.

PAGE 115

115 APPENDIX HOST PLANT RANGE OF RAOIELLA INDICA HIRST (ACARI: TENUIPALPIDAE) IN AREAS OF INVASION OF THE NEW WORLD5 Summary Raoiella indica has spread rapidly through the Neotro pical region where the mite damages economically and ecologically important plants. Th ree studies were conduc ted to determine the host plant range of R. indica using the presence of colonies containing all life stages as an indicator of reproductive suitabil ity. Periodic surveys at the Fair child Tropical Botanic Garden (Miami Dade County, FL, USA) and the Royal Bo tanical Gardens (Port of Spain, Trinidad and Tobago) identified 27 new reproductive host pl ants. The reproductive suitability of 2 dicotyledonous species and 3 na tive Florida palm species was examined. An updated list of reproductive host plants of R. indica is presented. All reporte d reproductive hosts (91 plant species) of R. indica are monocots from the orders Arecales (Arecaceae), Zingiberales (Heliconiaceae, Musaceae, Strelitziaceae, Zingi beraceae) and Pandanales (Pandanaceae). Most are palms of the family Arecaceae that originated in areas of the Eastern Hemisphere; about one fourth of the reported hosts are native to th e New World and could be considered new host associations of R. indica Six years after the initial detection in the Caribbean, R. indica has expanded its host plant range. Here we report 27 new reproductive host of R. indica that represent 30 % of increase on prev ious host plant records. As this mite continues spreading in the Neotropical region a great dive rsity of plants could potentia lly be affected. Key words: Raoiella indica reproductive hosts, monocotyledons, Arecaceae, palms. 5 Reprinted with permission from Exp Appl Acarol

PAGE 116

116 Introduction Raoiella indica Hirst (Acari: Tenuipalpidae), also called the red palm mite, is a phytophagous mite that recently invaded the Western Hemisphere. This mite is a polyphagous species that can reach very hi gh populations and cause signific ant damage to various plant species. It is also the first mite species observe d feeding through the stomata of its host plants (Ochoa et al. 2011). Through this specialized feeding habit R. indica probably interferes with the photosynthesis and respiration processes of thei r host. However, the damage caused by this species to most of its host plants has not yet been characte rized. In coconut, R. indica feeding causes an initial bronzing of the leaves which later turns into necrotic tissues. During 2004, R. indica was detected in Martinique and St. Lucia, and rapidly expanded its geographical range throughout th e Caribbean (Kane et al. 2005; Etienne and Flechtmann 2006; Rodrigues et al. 2007). In December 2007, R. indica was detected in the West Palm Beach area of south Florida (FDACS 2007) and spread to six counties of th e state thereafter. Raoiella indica has also reached Venezuela (Vsqu ez et al. 2008), Brazil (Navia et al. 2011), Colombia (Carrillo et al. 2011b) and Mexico (NAPPO 2009). Concerns have arisen about the consequences of the establishment of this exotic species in th e Neotropical region where economically and ecologically important plants c ould potentially be affected. Most studies and reports of this mite ha ve focused on one of its more suitable and economically important host, Cocos nucifera (Arecales: Arecaceae) (Nageshachandra and Channabasavanna 1984 and 1983b; Sarkar and Somc houdry 1989; Pea et al 2009). Few studies have addressed the po tential effects of R. indica on other host plants in the New World. In 2006, Welbourn made available a list of reported host plants of R. indica in the Caribbean region compromising 59 plant species. All the reported plants were monoc otyledons that belong to the orders Arecales [family Arecaceae (42 species )], Zingiberales [families Musaceae (6),

PAGE 117

117 Heliconiaceae (5), Streli tziaceae (2) and Zingiber aceae (3)] or Pandanales [family Pandanaceae (1)] (Table A-4). In 2009, Cocco and Hoy publishe d an expanded list of 72 reported host plants. Interestingly, these reports included 7 dicotyledon plant species from the families Aceraceae (1), Celastraceae (1), Myrtaceae (2), Lamiaceae (1 ) Oleaceae (1) and Fabacaeae (1) raising the question whether this mite is a highly polyphagous species or a stenophagous species that feeds and reproduces on relatively few plant families within the orders Arecales, Zingiberales and Pandanales. As this exotic species keeps expa nding its geographical rang e in the New World, a comprehensive study on the host plant range of R. indica is needed in order to devise feasible management options for the pest. We present results of three st udies that contribute to the knowledge of the host plant range of R. indica in the New World. We evaluated the poten tial host range of red palm mite by surveying two botanical gardens that had extensiv e tropical plant and palm collections located in areas of recent R indica invasion (Miami Dade, FL, USA a nd Port of Spain, Trinidad and Tobago). The reproductive suitability of 2 di cotyledonous species and 3 native Florida palm species was examined through infesti ng the plants with known numbers of R. indica in a plant nursery. Further field surveys evaluated the viabil ity of the Florida native palms as reproductive hosts. Based on the studies, we present an updated host plant list a nd discuss the known distribution and phylogenetic placement of the reported reproductive hosts of R. indica Material and Methods Fairchild Tropical Botanic Garden Survey, Florida, USA This botanical garden, located in the Coral Gables area of Miami Dade County, has a large collection of taxonomically arranged and well-do cumented tropical plants with emphasis on palms. All plants in the Palmet um area of the garden reachable with an extension pole (5 m) were inspected three times (25 February 2009, 24 November 2009 and 25 February 2010). Plants

PAGE 118

118 were inspected by looking at the underside of five fronds per plant using a hand lens (20 ). Plants with at least one R. indica individual were recorded and sa mples of 5 pinnae/ plant were taken to the laboratory to count the total number of individuals (eggs, im matures and adults) per pinna under a dissecting microscope (50 ). Leaf area was measured us ing a leaf area meter (LI3000C Portable area meter; LI-COR Biosciences) and mite densities estimated by dividing the number of mites by the area per pi nna. Plants with established R. indica colonies, having all developmental stages (egg, larva, protonym ph, deutonymph and adult), were considered reproductive hosts. Plants with only R. indica adults were considered non-reproductive hosts. In addition, all natural enemies (predators or pathogens) found associated with R. indica were also recorded and counted. The study was conducted appr oximately one year afte r the first report of R. indica in the area. Royal Botanical Gardens Survey, Port of Spain, Trinidad and Tobago Twenty five species of palms were selected from this 61.8 acre botanical garden located at 10.675284,-61.513696. Three trees were selected for each species when possible. The trees received no regular fertilizer or pesticide application, but mowing was performed periodically around them. The size of the palm depended on the species and ranged from 2-8 meters tall. One frond from the upper 1/3rd of the canopy (leaf 3 from the central spike), the middle 1/3rd canopy and the lower 1/3rd canopy were removed during the dry season (March 26-April 14, 2008) and again in the rainy season (May 29-June 11, 2008) The fronds were individually placed into plastic bags and transported to the laboratory in coolers. All st ages of red palm mites (adults, nymphs, larvae and eggs) and phytoseiid motiles (adults, nymphs and larvae) were counted using a mite brushing machine (Leedom Engineering, Twai n Harte, CA). As in the Fairchild Tropical Botanic Garden study, plants with R. indica colonies of a ll stages (egg, la rva, protonymph,

PAGE 119

119 deutonymph and adult) were considered reproductive host s while plants with only R. indica adults were considered non-reproductive hosts The surface area of each of frond was estimated taking the width and height for the secti on and calculating the area using base height in order to account for the triangular shape of th e sample. In addition, phytoseiid mites were collected from each palm species from which they were encountered for identification. The study was conducted approximately 2 years after the fi rst report of red palm mite in the area. Evaluation of 3 Florida Native Palms as Reproductive Host of R. indica The potential of three species of native palms, cabbage palm ( Sabal palmetto ), saw palmetto (Serenoa repens ), and Florida thatch (Thrinax radiata ), as reproductive hosts of R. indica was evaluated under nursery (University of Florida, West Palm Beach Extension Office, Palm Beach County, FL) and field conditions. Coconut was included in the study as positive control plant. In the nursery experiment, potted (3 gal. vol .) palms of each species were infested by placing 20 field collected R. indica adult females per leaf on 5 random ly selected leaves (total of 100 adults per palm). The five infested leaves were marked in each palm and R. indica infestation or lack of, examined every m onth thereafter during 5 months (August 2008 to December 2008) until the coconut palms started to show severe bronzing and leaf necrosis. All R. indica eggs, nymphs and adults were counted in 15 randomly selected locations of the frond using a hand lens (10 ) with a viewing field of 6.45 cm2. The experiment had 4 treatments (palm species) and 4 replications. All palms were free of pesticides, fertilized (Osmocote 15-9-12/ NP-K at a rate of 2 Tbsp per 3-gal pot) and wate red daily using automated overhead irrigation. The 3 native palms had been grown for approximately 2-years and the coconut palms for 1-year prior to infestation.

PAGE 120

120 In addition to the nursery experiment, eight locations in South Florida known to have R. indica infestations (4 in Miami-Dade County, 2 in Broward County and 2 in Palm Beach County) were surveyed every 2 months from December 2008 to August 2009. Each site was approximately 2 km2 and contained at least one cabbage palmetto, 1 saw palmetto and 1 Florida thatch palm. All study sites were located with in 300 to 800 meters of coconut palms with moderate to high infestation of R. indica which were considered a source of R. indica infestation for the native palms. At each site, one plant of each palm species was used to count R. indica each sample date. Three fronds were chosen randomly and R. indica counts done as described for the nursery evaluation. Coconut palms were ev aluated by taking 15 pinnae randomly selected from a single frond located in the lower canopy. Evaluation of Two Dicotyledon Plant Species, Phaseolus vulgaris and Ocimum basilicum, as Hosts of R. indica Potted O. basilicum (approximately 4 months old, obtained from a local nursery), C. nucifera plants (approx 1 year old) and P. vulgaris (grown from seed for 4 weeks) were infested with R. indica in a similar way to the nursery experi ment previously described. Fifty fieldcollected adult R. indica females were placed on the bean and basil leaves and on 1 selected coconut frond. The plants were then examined weekly for R. indica or lack of for 5 weeks. All R. indica life stages were counted using 10 hand lens. After 6 weeks, all plants were taken to the laboratory and destructively sa mpled. The total number of R. indica (eggs, immatures and adults) per leaf or pinna was counted us ing a dissecting microscope (50 ). The leaf area for each plant was estimated by taking the width and height for the section and calculating the area using base height in order to account for the triangular shape of the sample. This experiment had ten replicates.

PAGE 121

121 Results Fairchild Tropical Botanic Garden Survey, Florida, USA In the first survey (Feb 25 2009) R. indica was found on three palms at variable population densities (Table A-1). Cocos nucifera and Phoenix canariensis had established colonies of R. indica whereas Roystonea lenis had only isolated individuals (adults) that were not forming colonies. In the second survey (Nov 24 2009), R. indica was found on 24 palm species at variable population densiti es; eighteen had established colonies and six had isolated individuals (adults) not forming col onies (Table A-1). As in the first survey, C. nucifera and P. canariensis had R. indica colonies. In contrast, no mites were found on R. lenis during the second survey. In the third survey (Feb 25 2010), R. indica was found on 26 palm species at variable population densities; 24 had established co lonies and two had isolated in dividuals not forming colonies (Table A-1). Among the 24 plants with multigenerat ional colonies 12 also had colonies in the previous survey. In contrast, six palm species that had established co lonies during the second survey completely lost R. indica or had only isolated individuals on them during the third survey (Table A-1). Phytoseiid mites, identified as Amblyseius largoensis (Acari: Phytoseiidae), were found associated with R. indica on 12 and 13 of the plants during the second and third survey respectively (Table A-1). Overall, R.indica was found only on 36 palms among the high diversity of palm species in Fairchild Tropical Botanic Garden (a complete list of the plants is available at http://www.fairchildgarden.org/livingcollections/li stoflivingplantsinfairchildtropicalbotanicgarde n/). Thirty one of these palms were considered reproductive hosts because they had established colonies with all developmental stages. Among th e reproductive hosts, nine palm species were previously reported as hosts of R. indica and 22 palm species are repor ted here for the first time (Table A-4).

PAGE 122

122 Royal Botanical Garden Survey, Port of Spain, Trinidad The number of R. indica varied greatly between the sa mple periods and on the 25 palm species (Table A-2). Raoiella indica numbers were highest in the dry season and decreased 80100% with the onset of the rainy season on palm s sampled at both time periods. Mite populations were highest on coconut ( C. nucifera ), pygmy date palm ( Phoenix roebelenii ) and lady palm ( Rhapis excelsa ) during the dry season (Table A-2). Raoiella indica were relatively rare on 68% of the species sampled where 7 species of palm had no mites, 5 species had very low populations (<0.05 R. indica /cm2) and 5 other species had low numbers (< 1 R. indica /cm2). Raoiella indica were found throughout the palm canopy and were not found in greater numbers in any portion of the canopy. All phytoseiid mites collected were identified as Amblyseius largoensis (Welbourn, pers. comm.). Seventy eight percent of the palms with R. indica also had phytoseiid mites present. Only three species of palms had R. indica and no detectable populat ions of phytoseiid mites (Table A-2). Only one palm species, Elaeis guineensis, had phytoseiid mites and no detectable population of red palm mites. The highest num ber of phytoseiids was found on the palms with the highest number of R. indica Similar to R. indica the numbers of phytose iids were highest in the dry season. In summary, R. indica was found on 18 palm species out of the 25 palms that were sampled (Table A-2). Fourteen of these palms were considered reproductive hosts for having established colonies with all developmental stages. Among the reproduct ive hosts eight palm species were previously reported as hosts of R. indica and six palms are reported here for the first time (Table A-4).

PAGE 123

123 Evaluation of 3 Florida Native Palms as Reproductive Host of R. indica In the nursery experiment, R. indica colonies developed on the coconut palms but did not develop on the native palms (Figure A-1). Raoiella indica adults were recorded on Florida thatch during the first September sampling, however no colonies were found on the succeeding sample dates. Raoiella indica populations on coconut gradually increased with the highest density recorded in November (15.5 2.1 R. indica / cm2), then decreased in December (Figure A-1). In the field evaluation, R. indica colonies were found on coc onuts at all sites and on all sampling dates with peak population palms occurring on June 2009. Raoiella indica were not found on the native palms during the first 2 samp ling dates (Figure A-2). In the February 2009 survey, R. indica adults were recorded on Florida thatch at Site 2 in Miami Dade and at Site 6 in Broward County (Figure A-2). In April 2009, R. indica colonies were found on Florida thatch at all sites except site 4 but the densities (0.24 0.1 R. indica / cm2) were low compared to the ones found on coconuts (6.9 1.8 R. indica / cm2). Raoiella indica were never found on cabbage and saw palmettos. The nursery and field studie s suggest that cabbage palms and saw palmetto are not reproductive hosts of R. indica Detection of established R. indica colonies on Florida thatch from the field evaluation confirmed this palm as a reproductive host. Evaluation of Two Dicotyledon Plant Species, P. vulgaris and O. basilicum, as Hosts of R. indica Adult R. indica adults were found on all plant specie s one and two weeks after infestation. However, only the coconut palms ha d established colonies with adults and all immature stages. By the third week, R. indica was found only on coconut. On the final sample date, R. indica adults and established colonies only occurred on coconut (Table A-3). These controlled

PAGE 124

124 infestations studies suggest that P. vulgaris and O. basilicum are not reproductive hosts for R. indica. Updated List of Reported Host Plants of R. indica The surveys of the botanical gardens in Florida and Trinidad identified 27 new reproductive host plants of R. indica (Table A-4). The new reports compromise 13 genera of the family Arecaceae, seven of which had not been previously reported as R. indica hosts [ Arenga (6 spp.), Heterospathe (4 spp.), Allagoptera (1 sp.), Brahea (1 sp.), Gaussia (1 sp.), Guihaia (1 sp.), Latania (1 sp.) and Neoveitchia (1 sp.)] and six from which other species were previously reported [ Livistona (7 spp.), Acanthophoenix (1 sp.), Caryota (1 sp.), Licuala (1 sp.), Phoenix (1 sp.) and Pritchardia (1 sp.)]. All dicotyledonous plants were excluded from the reproductive host plant list presented here. Four dicotyledonous plants ( Eugenia sp., Eucalyptus sp., Olea sp. and Cassine transvaalensis) were included in a previous list citing a reference by Kane and Ochoa (2006). However, the authors referred to these plants as hosts of Raoiella spp., not necessarily of R. indica The references on Phaseolus sp. and O. basilicum as hosts of R. indica do not substantiate its reproductive host status (Chaudhri et al. 1974; Gupta 1984). In contrast, our controlled infestation ex periments showed that P. vulgaris and O. basilicum were not suitable hosts for R. indica The other mention of a dicotyledonous plant ( Acer sp.) as host of R. indica was probably induced by a spelling error (O choa, pers. observ.). The original document (Mitrofanov and Strunkova 1979) cited a study on Areca sp. but it was misspelled once as Acer sp. leading to erroneous citations and translations thereafter. In conclusion, it is unclear whether R. indica can reproduce on any dicotyledonous plant. In contrast, all the available evidence suggests that R. indica feeds and reproduces only on monocot hosts.

PAGE 125

125 The host plant list of R. indica now comprises 91 plants speci es. All reported reproductive hosts of R. indica are monocots in the orders Arecales (A recaceae), Zingiberales (Heliconiaceae, Musaceae, Strelitziaceae, Zingibe raceae) and Pandanales (Pandanaceae) (Table A-4). Most ( 80%) are palms of the family Arecaceae. Within the Arecaceae (as classified by Baker et al. 2009) the subfamily Arecoideae is represented by 33 species in four tribes (Areceae 19 spp., Cocoseae 10 spp., Chamaedoreeae 2 spp. and Royst oneeae 2 spp.), Coryphoideae are represented by 38 species in six tribes (Trachycarp eae 18 spp., Caryoteae 8 spp., Phoeniceae 5 spp., Thrinaceae (= Cryosophileae) 4 spp., Borasseae 2 s pp. and Corypheae 1 sp.) and Ceroxyloideae are represented by two species of the tribe Cyclospatheae (Table A-4). Discussion Knowledge of the repro ductive host range of R. indica is important to identify plants that could transport and increase the in festations of this exotic specie s in areas of invasion of the New World. The best way to determine whethe r a plant is a repr oductive host of R. indica is to infest the potential host with known numb ers and ages of mites and co mpare the population changes to infested known reproductive hosts. In the field, parameters are needed to verify that plants growing in natural conditions can be considered reproductive hosts of R. indica Observations in areas with high R. indica infestations suggest that mites can be blown by the wind and land on various plants that are not necessarily reproductive hosts of R. indica. In this study, the status as a reproductive host was limited to plants where multigenerational R. indica colonies were found. Our studies showed a high variability of R. indica population densities on the disparate reproductive host plants. While most plants had relatively low R. indica densities, other reproductive hosts had mite densities similar to those found to cause damage on coconut. Plants with relatively high populations of R. indica included 11 species found in the Florida survey ( A. tremula A. engleri A. australasica, C. urens, G. princep s, G. grossefibrosa, H. elata, H.

PAGE 126

126 intermedia, L. mariae, L. mue lleri, L. rigida, N. storckii, and P. canariensis ), five in the Trinidad survey ( A. merrillii, P. aucalis P. pacifica P. macarthurii W. robusta ) and two found in both places (P. roebelenii and R. excelsa ). These species are capab le of sustaining large R. indica populations and could serve as a source for infesta tions at other locations. Interestingly, three of these species (A. tremula, A. engleri and H. intermedia ) had high populations on one sampling date but few or no mites on the succeeding sampling dates, despite the palms being under the same environmental conditions as other sp ecies that retained large numbers of R. indica A similar situation was observed in other plants ( Heliconia sp ., L. australis and L. carinensis in Florida and A. pinnata inTrinidad) suggesting that not all reproductive hosts will necessarily maintain high populations of R. indica In general, results of these studies suggest that there are differences in the fitness of R. indica on the different host plants. The fitness of a phytophagous arthropod on its host plants can be influenced by direct interactions between the arthropod and plant, including differences in adaptation to plant constitu tive and induced defenses (e.g., chemical detoxification ability, Agrawal 2000), a nd the physical characteristics of the plant surface. For instance, the feeding habits of R. indica suggest that the host plants must have special stomata that could allow mites to insert th eir stylet and feed inside them (Ochoa et al. 2011). Also, the performance of a phytophagous ar thropod on a host plant can result from interactions with competitors or enemies at other trophic levels, interactions with mutualists, or abiotic differences (e.g., temperature, sunlight) in the microhabitats in which the plant species grow (Thompson 1988). For example, a large drop in red palm mite populat ions during the rainy season was observed in Trinidad suggesting that R. indica is may be easily dislodged off of certain species of palm. The ability of R. indica to remain attached to the palm may be a possible reason for the differences in populat ions on the different palm species.

PAGE 127

127 Most R. indica host plants ( 70%, Table A-4) originated in the Eastern Hemisphere, primarily in tropical areas of Asia, Australia and/or Africa (Uhl and Dransfield 1987), where R. indica is widely distributed. Nevertheless, about one fourth of the reported hosts are native to the New World and could be consider ed new host associations of R. indica This suggests that R. indica has a high ability to adapt to new environments and feed on new plant species. However, not all New World palms are vulnerable to R. indica infestations. Our study showed that Florida native palms such as that cabbage palm and saw palmetto were not suitable hosts for R. indica. In another study Cocco and Hoy (2009) also found that R. indica females did not survive on cabbage and saw palmettos leaf disks. In the same study, R. indica completed a generation on the Florida needle palm ( Rhapidophyllum hystrix ) but with a longer development time, higher mortality, and lower fecundity than when reared on coconut leaf disks. In the botanical garden surveys R. indica populations were low on the majority of the new host asso ciations recorded. Most plant species with high R. indica densities were native to the Eastern Hemisphere; G. princeps and W. robusta were exceptions where mite populati ons were comparable to or higher than R. indica numbers found on coconut. These observations suggest that the evolution of host range patterns of R. indica in the New World is a complex process that needs to be investigated in detail to determine the potential threats that th is exotic species could represent for native palm species and important crops in the Neotropical region. The evolution of host associations has been studied in other phytophagous mite species Gould (1979) demonstrated that host range evolution in Tetranychus urticae (Acari: Tetranychidae) can be a rapid process influenced by the ecological proximity of the plant species a nd cross-adaptation to sets of plants. Most R. indica hosts are palms of the family Arecaceae with two subfamilies, Arecoideae and Coryphoideae, accounting for 75% of the reporte d host plants. Within the Arecoideae, plants

PAGE 128

128 in four tribes (Areceae, Ch amaedoreeae, Cocoseae and Roystoneeae) have been reported as R. indica reproductive hosts (Table A-4). All reported hosts of the tribe Areceae are native to the Eastern Hemisphere. In contrast, R. indica host species in the tribes Chamaedoreeae, and Roystoneeae are native to the Neotropics (Uhl and Dransfield 1987). Within the Cocoseae, six reported hosts are native to the Neotropics and four to the Old World. This tribe contains several economically important plants including coconut, a major host of R. indica with pantropical distribution (Gunn 2004). Another important plant E. guineensis (African oil palm) was reported as reproductive host of R. indica (Welbourn 2009), but in both the Fairchild and Trinidad botanical garden surveys R. indica populations were not dete cted on that host. The Cocoseae contain other species of value in loca l Neotropical economies that could be vulnerable to R. indica For example, plants in the genera Bactris and Attalea are widely used as food and fiber source. Further studies are required to determine the susceptibility of these plants to damage caused by the invasive mite. Another importa nt subfamily of palms, in terms of the host range of R. indica is the Coryphoideaea. Plants in fi ve tribes within the Coryphoideaea (Borasseae, Caryoteae, Corypheae, Trachycarpe ae, Phoeniceae and Thrinaceae) have been reported as R. indica reproductive hosts (Table A-4). A ll reported hosts of the Borasseae, Caryoteae, Corypheae and Phoeniceae are native to the Eastern Hemisphere (Uhl and Dransfield 1987). Reported hosts of the tribe Trachycarpeae included primarily species from the Old World but also four species from the Neotropics ( W. filifera, W. robusta, A. wrightii and B. armata ) (Table A-4). In contrast, all R. indica reproductive hosts in the tr ibe Thrinaceae (=Cryosophileae) are native to the New World (Uhl and Dransfield 1987). Besides palms, high populations of R. indica have been observed on other economically important hosts, including bananas (Musaceae), He liconias (Heliconiaceae) and bird of paradise

PAGE 129

129 (Strelitziaceae). However, the pest status of R. indica on these plants is unknown. Further studies are needed to determine the potential effect R. indica may have on these economically and ecologically important plants, which are widely distributed in the Neotropics. In closing, almost six years after the initial detection of in the Caribbean, R. indica has expanded its host plant range to various plants Here we report 27 reproductive hosts of R. indica that represent a 30 % increase on pr evious host plant records. As this mite continues spreading throughout the Neotropical region a gr eat diversity of plants could potentially be affected. More detailed studies are necessary to determine the potential effects of R. indica in its areas of invasion.

PAGE 130

130 Table 1-A. Mean densities SEM of R. indica and phytoseiid mites found on various palms in the Fairchild Tropical Botanic Garden, Miami, FL. Date 24-Feb-09 24-Nov-09 25-Feb-10 Plant RPM/ cm Phyto. / cm RPM/ cm Phyto. / cm RPM/ cm Phyto. / cm Cocos nucifera 0.59 0.25* 0 2.49 0.59* 0.03 0.01 1.62 0.03* 0.03 0.01 Phoenix canariensis 1.38 0.43* 0 3.58 1.1* 0 3.20 1.50* 0 Roystonea lenis 0.01 0.01 0 Arenga undulatifolia 0.28 0.14* 0.01 0.00 1.88 1.00* 0.01 0.00 Arenga tremula 0.65 0.38* 0.04 0.01 0.02 0.01 0.01 0.01 Arenga engleri 0.67 0.34* 0.03 .02 0 0 Arenga australasica 0.7 1 0.31* 0.03 0.01 0.20 0.07* 0.01 0.01 Attalea butyracea 0.01 0.01 0 0 0 Heliconia sp. 0.09 0.04* 0.02 0.00 0.01 0.01 0 Heterospathe elata var palauensis 0.50 0.22* 0.05 0.03 2.09 0.63* 0.04 0.01 Heterospathe intermedia 0.94 0.58* 0 0 0 Licuala spinosa 0.01 0.01 0 0 0 Livistona australis 0.38 0.17* 0.01 0.01 0 0 Livistona carinensis 0.32 0.18* 0.01 0.01 0 0 Livistona fulva 0.23 0.07* 0.01 0.00 0.32 0.12* 0.01 0.00 Livistona mariae 1.50 0.83* 0 3.07 1.10* 0.03 0.01 Livistona muelleri 3.31 1.58* 0.01 0.01 0.77 0.07* 0.03 0.02 Livistona rigida 0.44 0.17* 0 0.61 0.27* 0.01 0.01 Neoveitchia storckii 0.37 0.16* 0.03 0.01 0.18 0.05* 0.01 0.00 Musa acuminata Cavendish 0.39 0.10* 0 0.49 0.17* 0 Phoenix roebelenii 2.31 2.12* 0 1.93 0.28* 0 Allagoptera leucocalyx 0.01 0.02 0 0 0 Musa liukiuensis 0.01 0.01 0 0 0 Pritchardia affinis 0.01 0.01 0 0 0 Acanthophoenix rubra 0.21 0.05* 0 Caryota urens 0.90 0.17* 0.03 0.01 Gaussia princeps 3.85 0.43* 0 Brahea armata 0.10 0.08* 0 Phoenix sp 0.19 0.04* 0 Arenga microcarpa 0.38 0.09* 0 Phoenix reclinata 0.50 0.17* 0.01 0.01 Heterospathe elmeri 0.23 0.10* 0 Heterospathe negrosensis 0.03 0.01* 0 Guihaia grossefibrosa 2.14 0.89* 0 Rhapis excelsa 1.06 0.33* 0 Allagoptera arenaria 0.42 0.14* 0.01 0.01 reproductive host, plan ts with established R. indica colonies having all developmental stages (egg, larva, protonymph, deutonymph and adult).

PAGE 131

131 Table A-2. Mean number of red palm mites SE per cm2 and phytoseiid mites SE per cm2 found in palms located at the Botanical Gard ens, Trinidad. Samples were collected in the dry season (March 2008) and rainy season (June 2008). Date March 2008 (dry season) June 2008 (rainy season) Plant RPM/ cm Phytoseiids/ cm RPM/ cm Phytoseiids/ cm Cocos nucifera 5.76.75* 0.04.02 0.58* 0.29 0.01.8x-3 Arenga pinnata 0.32.24* 3.3x-3.3x-3 0 0 Pritchardia pacifica 1.05.52* 2.0x-3.8x-4 0.22* 0.11 1.9x-3.2x-3 Syagrus romanzoffiana 0.04.03* 5x-4.0x-4 *** *** Licuala spinosa 0.05.02* 5.7x-4.7x-4 *** *** Phoenix acaulis 1.38.34* 2.9x-3.2x-3 *** *** Phoenix roebelenii 3.89* 2.2x-3.2x-3 *** *** Ptychosperma macarthurii 1.47.68* 4.6x-3.1x-3 *** *** Washingtonia robusta 0.86.21* 6.3x-4.9x-4 *** *** Adonidia merrillii 1.65.61* 7.3x-3.0x-3 *** *** Latania sp ** 0.15* 6.0x-3 *** *** Rhapis excelsa ** 4.33* 0.02 *** *** Caryota urens ** 0.21* 0 *** *** Roystonea oleracea 0.003.0022.5x-4.5x-4 *** *** Sabal glaucescens 0.001.4x-40 *** *** Bactris spp.** 0.01 0 *** *** Elaeis guineensis 0 4x-4 0 5.2x-3 Chrysalidocarpus lutescens 0 0 *** *** Hyophorbe lagenicaulis 0 0 *** *** Scheelea urbaniana 0 0 0 0 Livistona rotundifolia *** *** 0.07.04* 1.3x-3.7x-4 Sabal umbraculifera *** *** 0.001.4x-1.5x-3.4x-4 Euterpe oleracea *** *** 0 0 Livistona chinensis *** *** 0 0 Thrinax floridana ( T. radiata) *** *** 0 0 reproductive host, plan ts with established R. indica colonies having all developmental stages (egg, larva, protonymph, deutonymph and adult). ** only one palm available. *** No samples taken Table A-3. Average counts of R. indica and phytoseiid mites on basil, bean and coconut six weeks after being infested with 50 R. indica females (Means SEM). Plant Average mite count/cm2 R. indica Predatory Mite Basil 0.00 0.00 3.50x-4 5.30x-5 Bean 0.00 0.00 4.20x-4 7.50x-5 Coconut 0.02 0.003 0.002 0.00

PAGE 132

132 Table A-4. Reported host plant species of Raoiella indica Subfamily and tribe classi fication (Baker et al. 2009) Order Family Subfamily Tribe Scientific name Report Arecales Arecaceae Arecoideae Areceae Archontophoenix alexandrae (F. Muell.) H. Wendl. & Drude Welbourn (2009) Acanthophoenix rubra (Bory) H. Wendl. Fairchild, new report Areca catechu L. Nagesha-Chandra & Channabasavanna (1984) Areca sp. Pritchard & Baker (1958), Fairchild Neoveitchia storckii (Wendl) Fairchild, new report Dypsis decaryi (Jum.) Beentje & J. Dransf. Welbourn (2009) Dypsis lutescens (H. Wendl.) Beentje & J. Dransf. (= Chrysalidocarpus ) Kane et al. (2005) Adonidia merrillii (Becc.) Becc. (= Veitchia ) Fletchmann & Etienne (2004), Trinidad Ptychosperma elegans (R.Br.) Blume Welbourn (2009) Ptychosperma macarthurii (H. Wendl. ex HJ Veitch) H. Wendl. ex Hook. f. Etienne & Fletchmann (2006), Trinidad Ptychosperma sp. Cocco & Hoy (2009) Veitchia arecina Becc. Cocco & Hoy (2009) Veitchia sp. Welbourn (2009) Wodyetia bifurcata A.K. Irvine Welbourn (2009) Dictyosperma album (Bory) H. Wendl. & Drude ex Scheff. Moutia (1958) Heterospathe elmeri Becc. in Leafl. Fairchild, new report Heterospathe negrosensis Becc. Fairchild, new report Heterospathe elata Hough & Hubb Fairchild, new report Heterospathe intermedia (Becc.) Fernando Fairchild, new report Chamaedoreeae Ch amaedorea sp. Welbourn (2009) Gaussia princeps (Scott) Fairchild, new report C o c o s e a e Butia capitata (Mart.) Becc. Welbourn (2009) Cocos nucifera L. Hirst (1924), Welbourn (2009) Fairchild, Trinidad Syagrus schizophylla (Mart.) Glass. Welbourn (2009)

PAGE 133

133 Table A-4 Continuation Order Family Subfamily Tribe Scientific name Report Syagrus romanzoffiana (Cham.) Glass. Kane et al. (2005), Trinidad Allagoptera arenaria (Gomes) O. Kintze Fairchild, new report Beccariophoenix madagascariensis Jum. & H. Perrier Welbourn (2009) Aiphanes caryotifolia (Kunth) H. A. Wendl.* Welbourn (2009) Aiphanes sp. Kane et al. (2005) Bactris plumeriana Mart. Welbourn (2009) Elaeis guineensis Jacq. Welbourn (2009), NR-2 Roystoneeae Roystonea borinquena Cook Welbourn (2009) Roystonea regia (Kunth) Cook Welbourn (2009) Ceroxyloideae Cyclospatheae Pseudophoenix sargentii H. Wendl. Welbourn (2009) Pseudophoenix vinifera (Mart.) Becc. Welbourn (2009) Coryphoideae Borasseae Bismarckia nobilis Hildebr. & Wendl. Welbourn (2009) Latania sp. Trinidad, new report Caryoteae Arenga australasica (H. L. Wendl. & Drude) S. T. Blake Fairchild, new report Arenga engleri Becc. Fairchild, new report Arenga pinnata (Wurmb) Merrill Trinidad, new report Arenga tremula (Blanco) Becc. Fairchild, new report Arenga undulatifolia Becc. Fairchild, new report Arenga microcarpa Becc. Fairchild, new report Caryota mitis Lour Etienne & Fletchmann (2006), Fairchild, Trinidad Caryota urens L. Fl orida & Trinidad, new report Corypheae Corypha umbraculifera L. Welbourn (2009) L i v i s t o n e a e (Trachycarpeae) Licuala grandis Wendl. Etienne & Fletchmann (2006) Licuala spinosa Thunberg Trinidad, new report, NR-1 Livistona australis (R. Br.) Mart. Fairchild, new report Livistona carinensis (Chiov.) Dransf.& Uhl Fairchild, new report

PAGE 134

134 Table A-4 Continuation Order Family Subfamily Tribe Scientific name Report Livistona fulva A.N. Rodd Fairchild, new report Livistona mariae F.Muell Fairchild, new report Livistona muelleri F.Muell. Fairchild, new report Livistona rigida Becc. Fairchild, new report Livistona rotundifolia (Lam.) Mart. Trinidad, new report Livistona chinensis (Jacq.) R. Br. ex Mart. Welbourn (2009), NR-2 Pritchardia pacifica Seem. & H. Wendl. Etienne & Fletchmann (2006), Trinidad Pritchardia vuylstekeana H. Wendl. Cocco & Hoy (2009) Washingtonia filifera (Linden ex Andr) H. Wendl. Welbourn (2009) Washingtonia robusta H. Wendl. Etienne & Fletchmann (2006), Trinidad Acoelorrhaphe wrightii (Griseb. & H. Wendl.) H. Wendl. ex Becc. Welbourn (2009)* Brahea armata S. Watson Fairchild, new report Guihaia grossefibrosa J.Dransf., S.K.Lee & F.N.Wei Fairchild, new report Rhapis excelsa (Thunb.) A. Henry ex Rehder Welbourn (2009), Fairchild, Trinidad Phoeniceae Phoenix acaulis Roxb. Trinidad, new report Phoenix canariensis hort. ex Chabaud Etienne & Fletchmann (2006), Fairchild Phoenix dactylifera L. Sayed (1942) Phoenix reclinata Jacq. Welbourn (2009), Fairchild Phoenix roebelenii O'Brien Welbourn (2009), Fairchild, Trinidad Thrinaceae (Cryosophileae) Coccothrinax argentata (Jacq.) L. H. Bailey Co cco & Hoy (2009) Coccothrinax miraguama (Kunth) Becc. Welbourn (2009) Schippia concolor Burret Welbourn (2009) Thrinax radiata Loddiges ex J.A. & J.H. Schultes (= T. floridana) Welbourn (2009)

PAGE 135

135 Table A-4 Continuation Order Family Subfamily Tribe Scientific name Report Zingiberales Heliconiaceae Heliconia bihai L. Welbourn (2009) Heliconia caribaea Lam. Welbourn (2009) Heliconia psittacorum Sassy Welbourn (2009) Heliconia rostrata Ruiz & Pavon Etienne & Fletchmann (2006) Heliconia sp. Pea et al. (2006), Fairchild Strelitziaceae Ravenala madagascariensis Sonn. Welbourn (2009) Strelitzia reginae Ait. Etienne & Fletchmann (2006) Musaceae Musa acuminata Colla Kane et al. (2005), Fairchild Musa balbisiana Colla Kane et al. (2005) Musa corniculata Rumph Welbourn (2009) Musa sp. Etienne & Fletchmann (2006) Musa uranoscopus Lour. (= M. coccinea ) Kane et al. (2005) Musa paradisiaca L. Kane et al. (2005) Zingiberaceae Alpinia purpurata (Vieillard) K. Schumann Welbourn (2009) Alpinia zerumbet (Pers.) Burtt. & R. M. Sm. Welbourn (2009) Etlingera elatior (Jack) R. M. Sm. Etienne & Fletchmann (2006) Zingiber sp. Pea et al. (2006) Pandanales Pandanaceae Pandanus sp. Kane & Ochoa (2006) Pandanus utilis Bory Welbourn (2009) Plant names followed by are native to the Neot ropics according to Uhl and Dransfield 1987. NR= Non reproductive hosts.

PAGE 136

136 Figure A-1. Monthly av erage counts of R. indica on the 3 native palms and coconut palm after being artificially infested.

PAGE 137

137 Average RPM count/cm2 0 5 10 15 20 25 Coconut Sabal Palmetto Saw Palmetto Florida Thatch Average RPM count/cm2 0 5 10 15 20 25 Average RPM count/cm2 0 5 10 15 20 25 Sampling Dates Dec 2008Feb 2009Apr 2009Jun 2009Aug 2009 Average RPM count/cm2 0 5 10 15 20 25 Sampling Dates Dec 2008Feb 2009Apr 2009Jun 2009Aug 2009 Site 1 Site 2 Site 3 Site 4 Site 5Site 6 Site 7 Site 8 Figure A-2. Average R. indica count on 3 native palms and coconut palms in Miami-Dade County (Sites 1-4), Broward County (Sites 56), and Palm Beach County (Sites 7-8).

PAGE 138

138 LIST OF REFERENCES Alberti G, Crooker, AR (1985) Internal anatomy. In: Helle W, Sabelis MW (eds) Spider mites: their biology, natural enemies, and control, vol A. Elsevier Amsterdam, Netherlands, pp 29 311 Agrawal AA (2000) Host-range evolution: adapta tion and trade-offs in fitness of mites on alternative hosts. Ecology 81(2):500 Andrewartha HG, Birch LC (1954) The distribution and abundance of animals. University of Chicago Press, Chicago, Illinois, pp 782 Askari M, Arbabi M, Golmohammad ZN (2002) Plant mite fauna of Sistan-Baluchestan and Hormozgan provinces. J Entomol Soc Iran 22(1):87 Badii MH, Hernandez-Ortiz E, Flores A, Landeros JN (2004) Prey stage preference and functional response of Euseius hibisci to Tetranychus urticae (Acari: Phytoseiidae). Exp Appl Acarol 34:263 Baker WJ, Savolainen V, Asmussen-Lange CB, Chas e MW, Dransfield J, Forest F, Harley MM, Uhl NW, Wilkinson M (2009) Complete generi c-level phylogenetic analyses of palms (Arecaceae) with comparisons of supertree and supermatrix approaches. Syst Biol 58(2):240 Bartlett, BR (1968) Outbreaks of two-spotted spid er mites and cotton aphids following pesticide treatment. I. Pest stimulation vs. natural en emies destruction as th e cause of outbreaks. J Econ Entomol 61:297 Bellotti AC, Smith L, Lapointe SL (1999) Recent advances in cassava pest management. Annu Rev Entomol 44:343 Bibbinger DJ, Weber DC, Hull LA (2009) Stethorini in biological control. Biol Control 51:268 283 Blackwood JS, Luh HK, Croft BA (2004) Evaluation of prey-stage preference as an indicator of life-style type in phytoseiid m ites. Exp Appl Acarol 33:261 Blackwood JS, Schausberger P, Croft BA (2001 ) Prey-stage preferen ce in generalist and specialist phytoseiid mites (Acar i: Phytoseiidae) when offered Tetranychus urticae (Acari: Tetranychidae) eggs and la rvae. Environ Entomol 30:1103 Bowman HM (2010) Molecular discrimination of phyt oseiids associated with the red palm mite Raoiella indica (Acari: Tenuipalpidae) from Mauritius and South Florida. MSc Thesis, University of Florida, Ga inesville, Florida, pp 136

PAGE 139

139 Braun AR, Guerrero, JM, Belloti AC, Wilson LT ( 1987) Relative toxicity of permethrin to Mononychellus progresivus Doreste and Tetranychus urticae Koch (Acari: Tetranychidae) and their predators Amblyseius limonicus Garman & McGregor (Acari: Phytoseiidae) and Oligota minuta Cameron (Coleoptera: Staphylinidae): bioassays and field validation. Environ Entomol 16:545 Bruce-Oliver SJ, Hoy MA (1990) Effect of prey stage on life-table attrib utes of a genetically manipulated strain of Metaseiulus occidentalis (Acari: Phytoseiidae) Exp Appl Acarol 9:201 CARDI (2010) Caribbean Agricultural Research and Development Institute. Natural resource management, invasive species. Available via DIALOG. http://www.cardi.org/default.asp?id=46 Accessed July 2010 Carrillo D, Pea JE, Hoy MA, Frank JH (2010) De velopment and reproduction of Amblyseius largoensis (Acari: Phytoseiidae) feeding on pollen, Raoiella indica (Acari: Tenuipalpidae), and other microarthropods inhabiting coconuts in Florida, USA. Exp Appl Acarol 52:119 129 Carrillo D, Amalin D, Hosein F, Roda A, Duncan R, Pea JE (2011a) Host plant range of Raoiella indica Hirst (Acari: Tenuipalpidae) in ar eas of invasion of the New World. Submitted to Exp Appl Acarol Carrillo D, Navia D, Ferragut F, Pea JE (2011b) First report of Raoiella indica HIRST (Acari: Tenuipalpidae) in Colombia Fla Entomol 94(2):370 Carrillo D, Pea JE (2011) Prey-stage preferences and functional numerical responses of Amblyseius largoensis (Acari: Phytoseiidae) to Raoiella indica (Acari: Tenuipalpidae). Submitted to Exp Appl Acarol Castagnoli M, Simoni S (1999) E ffect of long-term feeding hist ory on functional and numerical response of Neoseiulus californicus (Acari: Phytoseiidae). Ex p Appl Acarol 23:217 Chaudhri WM, Akbar S, Rasol A (1974) Taxonomic studies of the mites belonging to the families Tenuipalpidae, Tetranychidae, Tuckerellidae, Caligonellidae, Stigmaeidae and Phytoseiidae. University of Agriculture, Ly allpur, Pakistan, PL Project on Mites, pp 250 Chinnamade-Gowda C, Mallik B (2010) Fauna of Phytoseiid mites associ ated with plants in southern Karnataka, India. XIII International Congress of Acarology, Reci fe, Brazil, Abstract Book, pp 55 Cocco A, Hoy MA (2009) Feeding, reproduction, and developmen t of the red palm mite (Acari: Tenuipalpidae) on selected palms and banana cultivars in quarantine. Fla Entomol 92:276 291 Crooks JA, Soule ME (1999) Lag times in population explosions of invasive species: causes and implications. In: Sandlund OT, Schei PJ, Vike n A (eds) Invasive species and biodiversity management. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp 103

PAGE 140

140 Cuthbertson AGS, Bell AC, Murchie AK ( 2003) Impact of the predatory mite Anystis baccarum (Prostigmata: anystidae) on apple rust mite Aculus schlechtendali (Prostigmata: Eriophyidae) populations in Northern Ireland Bramley orchards. Ann Appl Biol 142(1):107 Daneshvar H (1980) Some predator mites from no rthern and western Iran. Entomol Phytopathol Appli 48(1):87 Daniel M (1981) Bionomics of the predaceous mite Amblyseius channabasavanni (Acari: Phytoseiidae) predaceous on the palm mite. In : Channabasavanna GP (ed) Contributions to acarology in India. Anubhava Printers, Bangalore, pp 167 De la Torre PE, Suarez A, Iris A (2010) Presencia del acaro Raoiella indica Hirst (Acari: Tenuipalpidae) en Cuba. Re v Proteccion Veg 25(1):1 Denmark HA, Muma MH (1989) A revision of the genus Amblyseius Berlese, 1914 (Acari: Phytoseiidae). Occasional papers of the Florida state collection of arthropods, USA, 4, pp 149 Dittrich VP, Streibert P, Bathe PA (1974) An ol d case reopened: mite stimulation by insecticide residues. Environ Entomol 3:534 Dunca RE, Carrillo D, Pea JE (2010) P opulation dynamics of the red palm mite, Raoiella indica (Acari: Tenuipalpidae) in Florida, USA. X III International Congress of Acarology, Recife, Brazil, Abstract Book. p 74 Elder RJ, Smith D (1995) Mass rearing of Aonidiella orientallis (Newstead) (Hemiptera: Diaspididae) on butternut gram ma. J Austral Entomol Soc 34:253 Elton CS (1958) The ecology of invasions by an imals and plants. Methuen and Co., Ltd, London, UK, pp 183 Elwan A (2000) A survey of the insect and mite pests associated with date palm trees in AlDakhliya region, Sultanate of Oman. Egypt J Agr Res 78:653 EPPO (2009) Raoiella indica (Acari: Tenuipalpidae). In: Euro pean and Mediterranean plant protection organization. Available via DIALOG. http://www.eppo.org/QUARANTINE/Alert_ List/insects/raoi ella_indica.htm Acce ssed Sept 2009 Etienne J, Flechtmann CHW (2006) First record of Raoiella indica (Hirst, 1924) (Acari: Tenuipalpidae) in Guadeloupe and Saint Ma rtin, West Indies. Int J Acarol 32:331 Fan Y, Petit F (1994) Parameter estimation of functional response. Environ Entomol 23:785 FDACS (2007 and 2011) Red palm mite infestation identified in Palm Gardens. In: Florida Department of Agriculture and Consumer Services. Available via DIALOG. http://www.doacs.state.fl.us/press/2007/12052007_2.html. Accessed Sept 2009

PAGE 141

141 Flechtmann CHW, Etienne J (2004) The red palm mite, Raoiella indica Hirst, a threat to palms in the Americas (Acari: Prostigmata: Te nuipalpidae). Syst Appl Acarol 9:109 Flores-Galano G, Montoya A, Rodriguez H (2010) Biologia de Raoiella indica Hirst (Acari: Tenuipalpidae) sobre Areca catechu L. Rev Proteccion Veg 25(1):11 Frank JH, Bennett FD, Cromroy HL (1992) Distribution and prey records for Oligota minuta (Coleoptera: Staphylinidae) a predator of mites. Fla Entomol 75(3):376 Gagn RJ (2004) A catalog of the Cecidomyiid ae (Diptera) of the world. Mem Entomol Soc Wash 25: 1 Gallego CE, Aterrado ED, Batomalaque CG ( 2003) Biology of the false spider mite, Rarosiella cocosae Rimando, infesting coconut palms in Ca miguin, northern Mindanao (Philippines). Philipp Entomol 17(2):187 Gallego CE, Batomalaque CG (2004) Biology, fecundity and predatory capacity of Telsimia ephippiger Chapin (Coleoptera Coccinellidae) false spider mite predator of the coconut. Philipp Entomol 18(2):166 Galvao AS, Gondim MGC, Moraes GJ, De Oliveira JV (2007) Biology of Amblyseius largoensis (Muma) (Acari: Phytoseiidae), a potential predator of Aceria guerreronis Keifer (Acari: Eriophyidae) on coconut tr ees. Neotrop Entomol 36:465 Gerson U, Venezian A, Blumberg D (1983) Phyt ophagous mites on date palms in Israel. Fruits 38:133 Gomes FP, Prado CHBA (2007) Ecophysiology of coconut palm under water stress. Braz J Plant Physiol 19(4):377 Gonzales AI, Ramos M (2010) Desarrollo y reproduccion de Raoiella indica Hirst (Acari: Tenuipalpidae) en laboratori o. Rev Proteccion Veg 25(1):7 Gould F (1979) Rapid host range evoluti on in a population of the phytophagous mite Tetranychus urticae Koch. Evolution 33(3):791 Gunn BF (2004) The phylogeny of the Cococeae (Arecaceae) with emphasis on Cocos nucifera Ann Missouri Bot Gard 91:505 Gupta SK (2001) A conspectus of natural enem ies of phytophagous mites and mites as potential biocontrol agents of agricultura l pests in India. In: Halliday RB, Walter DE, Proctor HC, Norton RA, Colloff MJ (eds) Proceedings of th e 10th International Congress of Acarology, Collingwood, Australia. CSIRO Publishing, pp 484 Gupta SK (1998) Present state of knowledge on bi oecological studies of plant mites including agricultural pests and predatory m ites in India. In: Sinha MP (ed) Advances in ecobiological research Vol 2. APH Publishing Corporation, New Deli, India, pp 117

PAGE 142

142 Gupta YN (1984) On a collection of tetranychoid m ites from Tamil Nadu with description of a new species of Aponychus (Acari: Tetranychidae). Bu ll Zool Surv India 6: 237 Gutirrez B, Rodrigues N, Garcia A (2007) Situacin actual del co cotero in al municipio Valdez del estado Sucre. In: Produccin y negocio. Available via DIALOG. http://www.produccionynegocio.com /edicion_20/cocotero.htm Accessed 29 Sept 2009 Hastie E, Benegas A, Rodrguez H (2010) Inve ntario de caros depredadores asociados a fitocaros en plantas de las familias Arecaceae y Musaceae Rev Prot Veg 25:17 Helle W, Bolland HR, Haitmans WRB (1980) Chromo somes and types of parthenogenesis in the false spider mites (Acari: Te nuipalpidae). Genetica 54(1):45 Herren HR, Neuenschwander P (1991) Biological c ontrol of cassava pests in Africa. Annu Rev Entomol 36:257 Hirst S (1924) On some new species of red spider. Ann Mag Nat Hist 14:522 Hokkanen H., Pimentel D (1984) Ne w approach for selecting biol ogical control agents. Can Ent 116:1109 Holling C (1959a) The components of predation as revealed by a study of small mammal predation of european pine sawfly. Can Entomol 91:293 Holling C (1959b) Some characteristics of simple types of predation and parasistism. Can Entomol 91:385 Holling C (1961) Principles of inse ct predation. Annu Rev Entomol 6:163 Holling C (1965) The functional res ponse of predators to prey density and its role in mimicry and population regulation. Mem Entomol Soc Can 45:5 Hoy MA, Flaherty D, Peacock W, Culver D ( 1979) Vineyard and labor atory evaluations of methomyl, dimethoate, and permethrin for gr ape pest management program in the San Joaquin Valley of California. J Econ Entomol 72:250 Hoy MA, Smilanick JM (1981) Non-random pr ey location by the phytoseiid predator Metasieulus occidentalis: differential responses to several spidermite species. Entomol Exp Appl 29:241 Hoy MA (2011) Agricultural acarology: introdu ction to integrated mite management. CRC Press, Boca Raton, Florida, p 129 Huffaker CB, van de Vrie M, McMurtry JA (1970) Ecology of tetranychid mites and their natural enemies: a review. II. Tetranychi d populations and thei r possible control by predators: an evaluation. Hilgardia 40:391

PAGE 143

143 Jagadish PS, Nageshachandra BK (1981) Biology of Typhlodromips tetranychivorous (Acari: Phytoseiidae) on red palm mite Raoiella indica (Acari: Tenuipalpidae). In: Channabasavanna GP (ed) 1st Indian Symposium in Acarology. April 23, Bangalore, India, p 173 Jai F, Margolies DC, Boyer JE, Charlton RE (20 02) Genetic variation in foraging traits among inbred lines of a predat ory mite. Heredity 89:371 James DG (1990) Biol ogical control of Tetranychus urticae (Koch) (Acari: Tetranychidae) in New South Wales peach orchards: the role of Amblyseius victoriensis (Acari: Phytoseiidae). Aust J Zool 37:645 Jones VP (1983) Pest resurgence of, and sa mpling plans for, the citrus red mite, Panonychus citri (McGregor) (Acari: Tetranychidae). PhD dissertation, Univ of California, Riverside, California, pp 104 Juliano SA (2001) Nonlinear curve fitting: predati on and functional response curves. In: Cheiner SM, Gurven J (eds) Design and analysis of eco logical experiments, 2nd ed Chapman & Hall, New York, pp 178 Kamburov SS (1971) Feeding, deve lopment, and reproduction of Amblyseius largoensis on various food substances. J Econ Entomol 64:643 Kane EC, Ochoa R (2006) Detection and Identification of Raoiella indica Hirst (Acari: Tenuipalpidae). Agency Report. USDA ARS Syst ematic Entomology Laboratory Beltsville, MD. Available via DIALOG. http://entnemdept.ufl.edu/pestalert/Raoiella_indica_Guide.pdf Accessed May 2011 Kane EC, Ochoa R, Mathurin G, Erbe EF (2005) Raoiella ind ica Hirst (Acari: Tenuipalpidae): An island-hopping mite pest in th e Caribbean. Available via DIALOG. http://www.sel.barc.usda.gov/acari/PDF /Raoiella indica-Kane et al.pdf Accessed July 2010 Kapur AP (1961) A new species of Stethorus W eise (Coleoptera: Coccinellidae), feeding on Arecanut palm mites in Kerala, Sout hern India. Entomophaga 6(1):35 Koehler HH (1999) Predatory mites (Gamasina, Mesostigmata). Agric Ecosyst Environ 74:395 410 Koveos DS, Broufas GD (2000) Functional response of Euseius finlandicus and Amblyseius andersoni to Panonychus ulmi on apple and peach leaves in th e laboratory. Exp Appl Acarol 24:247 Krantz GW, Walter DE (2009) A manual of acar ology, third edition. Texas Tech University press, Lubbock Texas, pp 807 Krebs CJ (1978) Ecology, the expe rimental analysis of distribu tion and abundance, 2nd edition, Harper and Row, New York, USA, pp 694

PAGE 144

144 Laing JE, Osborn JAL (1974) The effect of prey density on the functional and numerical response of three species of predatory mites. Entomophaga 19:267 Lawson-Balagbo LM, Gondim MGC, Moraes GJ, Ha nna R, Schausberger P (2008) Exploration of the acarine fauna on coconut palm in Brazil with emphasis on Aceria guerreronis (Acari : Eriophyidae) and its natural enemies. Bull Entomol Res 98:83 Lesna I, Sabelis MW (1999) Diet dependent fema le choice for males with good genes. Nature 401:581 Luck RF (1988) Experimental methods for ev aluating arthropod natural enemies. Annu Rev Entomol 33:367 Maggi VL, Leigh TF (1983) Fecund ity response of the twospotted sp ider mite to cotton treated with methyl parathion or phos phoric acid. J Econ Entomol 76:20 Maia AHN, Luiz AJB, Campanhola C (2000) Statis tical inference on associated fertility life table parameters using Jackknife technique: co mputational aspects. J Econ Entomol 93:511 518 Manly BFJ, Miller P, Cook LM (1972) Analysis of a selective predati on experiment. Am Nat 106:719 Marsaro Jr AL; Navia D, Gondim Jr MG, Silva FR Moraes GJ (2009) Chegou ao Brasil-o caro vermelho das palmeiras Raoiella indica Cultivar, Hortalias e Frutas 57:31 McMurtry JA (1992) Dynamics and potential imp act of generalist phyto seiids in agroecosystems and possibilities for estab lishment of exotic species Exp Appl Acarol 14:371 McMurtry JA, Croft BA (1997) Life-styles of phyt oseiid mites and their role in biological control. Annu Rev Entomol 42:291 McMurtry JA, Moraes GJ (1984) Some phytose iid mites from the South Pacific, with descriptions of new species and a definition of the Amblyseius largoensis group. Int J Acarol 10(1):27 McMurtry JA, Morse JG, Johnson HG (1992) Studies of the impact of Euseius species (Acari: Phytoseiidae) on citrus mites us ing predator exclusion and pred ator release experiments. Exp Appl Acarol 15:233 Mesa NC, Ochoa R, Welbourn WC, Evans GA, Moraes GJ (2009) A catalog of the Tenuipalpidae (Acari) of the world w ith a key to genera. Zootaxa 2098:185 Meyerdirk DE, Khasimuddin S, Bashir M (1988) Importation, colonization and establishment of Anagyrus indicus (Hymenoptera: Encyrtidae) on Nipaecoccus viridis (Homoptera: Pseudococcidae) in Jordan. Entomophaga 33:229

PAGE 145

145 Mitrofanov VI, Strunkova ZI ( 1979) Ofredelitel kleshcheiploskot elok (a key to false spider mites). Donish, pp 148 (in Russian) Moraes GJ, Kreiter S, Lofego AC (2000) Plant mites (Acari) of th e French Antilles. 3. Phytoseiidae (Gamasida). Acarologia 40(3):237 Moraes GJ, McMurtry JA, Denmark HA, Campos CB (2004) A revised catalog of the mite family Phytoseiidae. Zootaxa 434:494 Mori H, Chant D (1966) The influence of prey density, relative humidity and starvation on the predaceous behavior of Phytoseiulus persimilis Athias-Henriot (Acarina: Phytoseiidae). Can J Zool 44:483 Moutia LA (1958) Contribution to study of some phytophagous Acarina a nd their predators in Mauritius. Bull Entomol Res 49:59 Muma MH (1955) Phytoseiidae (Acarina) associated with citrus in Florida. Ann Entomol Soc Am 48(4):262 Muma MH (1975) Mites associated with citrus in Florida. Bulletin No. 640A. Agricultural Experiment Stations, Institute of Food and Agri cultural Sciences, University of Florida, p 92 Nageshachandra BK, ChannaBasavanna GP (1983a) Stethorus keralicus (Coleoptera: Coccinellidae) feeding on Raoiella macfarlanei infesting roseapple. Ac arol Newsletter 13:3 Nageshachandra BK, Channabasavanna GP (1983b) Studies on seasonal fluctuation of the population of Raoiella indica Hirst (Acari: Tenuipalpidae) on coconut with reference to weather parameters. Indian J Acarol 8:104 Nageshachandra BK, Channabasavanna GP (1984) Development and ecology of Raoiella indica Hirst (Acari: Tenuipalpidae) on coconut. In: Gr iffiths DA, Bowman CE (eds), Acarology VI. Ellis Horwood Publishers, Chicester, UK, pp 785 Nangia N, ChannaBasavnna GP (1989) Feeding potential of Amblyseius tetranychivorus (Acari: Phytoseiidae), an indigenous predator on select ed tetranychid and tenuipa lpid mites. Indian J Acarol 10:75 NAPPO (2009) North American Plant Protection Or ganization. Detection of the red palm mite ( Raoiella indica ) in Cancun and Isla Mujeres, Qu intana Roo, Mexico. Available via DIALOG: http://www.pestalert.org/ oprDetail.cfm ?oprID=406 Accessed 30 December 2010 Navia D, Marsaro AL, Silva FR, Gondim MGC, Mo raes GJ (2011) First report of the red palm mite, Raoiella indica Hirst (Acari: Tenuipalpidae), in Br azil. Neotrop Entomol (accepted 29 Nov 2010) Navia D, Ochoa R, Welbourn C, Ferragut F (2010 ) Adventive Eriophyoid mites: a global review of their impact, pathways, prevention and challenges. Exp Appl Acarol 51:225

PAGE 146

146 Noronha ACDS, Moraes, GJ (2004) Reproductive compatibility between mite populations previously identified as Euseius concordis (Acari: Phytoseiidae). Exp Appl Acarol 32:271 279 Ochoa R, Beard JJ, Bauchan GR, Kane EC, Dow ling APG, Erbe EF (2011) Herbivore exploits chink in armor of Host. Am Entomol 57(1):26 Pea JE, Rodrigues JCV, Roda A, Carrillo D, Osborne LS (2009) Predator-prey dynamics and strategies for control of the red palm mite ( Raoiella indica ) (Acari: Tenuipalpidae) in areas of invasion in the Neotropics. Proceedings of the 2nd Meeting of IOBC/WPRS, Work group integrated control of plant feeding m ites. Florence, Italy 9 March 2009, pp 69 Pea JE, Mannion CM, Howard FW, Hoy MA (2006) Raoiella indica (Prostigmata: Tenuipalpidae): the red palm mite: a potential in vasive pest of palms and bananas and other tropical crops in Florida. In: University of Florida IFAS ex tension. Available via DIALOG. http://edis.ifas.ufl.edu/IN681. Accessed Sept 2009 Pim entel D, Zuniga R, Morrison D (2005) Upda te of the environmental and economic costs associated with alien-invasive specie s in the United States. Ecol Econ 52:273 Pritchard AE, Baker EW (1958) The false spider mites (Acarina: Tenuipalpidae). University of California publications in entomology 14(3):175 Puttarudriah M, ChannaBasavanna GP (1956) Some beneficial coccinellids of Mysore. J Bombay Nat Hist Soc 54:156 Puttaswamy, Rangaswamy HR (1976) Stethorus keralicus Kapur (Coleoptera: Coccinellidae), a predator of the areca palm mite. Curr Res 5:27 RADA (2007) The red palm mite in Jamaica. Rural Agricultural Development Authority of Jamaica. Available via DIALOG. http://www.moa.gov.jm/food/files/THE% 20RE D%20PALM%20MITE%20IN%20JAMAIC A.pdf Accessed May 2011 Rahmani H, Hoffman D, Walzer A, Shausber ger P (2009) Adaptive learning in the foraging behaviour of the predatory mite Phytoseiulus persimilis Behav Ecol 20:46 Rajagopal V, Kasturi-Bai KV, Voleti SR (1990) Screening of coconut genotypes for drought tolerance. Oleagineux 45:215 Ramos M, Gonzales AI, Gonzales M (2010) Management strategy of Raoiella indica Hirst in Cuba, based on biology, host plants, seasona l occurrence and use of acaricide. XIII International Congress of Acarology, Recife, Brazil, Abstract Book, pp 218 Reis A, Gondim JM, Moraes GJ, Hanna R, Scha usberger P, Lawson-Balagbo L, Barros R (2008) Population dynamics of Aceria guerreronis Keifer (Acari: Eriophyidae) and associated predators on coconut fruits in north eastern Brazil. Neotrop Entomol 37:467

PAGE 147

147 Reis PR, Sousa EO, Teodoro AV, Neto MP (2003) Effect of prey densities on the functional and numerical response of two speci es of predaceous mites (Aca ri: Phytoseiidae). Neotrop Entomol 32:461 Reis PR, Teodoro AV, Neto MP, Silva EA (2007) Life history of Amblyseius herbicolus (Acari: Phytoseiidae) on coffee plant. Neotrop Entomol 36:282 Ricciardi A, Palmer ME, Yan D (2011) Should biological invasions be managed as natural disasters? BioScience 61(4):312 Roda A, Dowling A, Welbourn C, Pea JE, Rodr igues JCV, Hoy MA, Ochoa R, Duncan RA, De Chi W (2008) Red palm mite situation in the Caribbean and Florida. Proc Caribbean Food Crops Soc 44(1):80 Rodrigues JCV, Kitajima EW, Childers CC, Chagas CM (2003) Citrus lepr osis virus vectored by Brevipalpus phoenicis (Acari: Tenuipalpidae) in citrus in Brazil. Exp Appl Acarol 30(1/3):161 Rodrigues JCV, Ochoa R, Kane EC (2007) First report of Raoiella indica Hirst (Acari: Tenuipalpidae) and its damage to coconut palm s in Puerto Rico and Culebra Island. Int J Acarol 33:3 Rodriguez H, Ramos M (2004) Biology and feeding behavior of Amblyseius largoensis (Muma) (Acari: Phytoseiidae) on Polyphagotarsonemus latus (Banks) (Acari: Tarsonemidae). Rev Proteccion Veg 19:73 Rogers D (1972) Random search and insect population models. J Animal Ecol 18:1 Rosen D, Huffaker CB (1983) An overview of desire d attributes of effective biological control agents, with particular emphasis on mites. In: Hoy MA, Cunningham GL, Kuntson L (eds) Biological control of pests by mites. Div Agr Sci, Univ Calif Publ 3304, pp 2 Roush RT, Hoy MA (1978) Relative toxi city of permethrin to a predator, Metaseiulus occidentalis and its prey, Tetranychus urticae Environ Entomol 7(2):287 Ryoo M (1996) Studies on the basic component of the predation of Phytoseiulus persimilis (Acari: Phytoseiidae). Res Pop Ecol 28:17 Sabelis MW (1985) Predatorprey interaction: pr edation on spider mites. In: Helle W, Sabelis MW (eds) Spider mites: their biology, natura l enemies and control. Elsevier, Amsterdam, Netherlands, pp 103 Sabelis MW, Lesna I (2010) Does artificial select ion for fixed prey preference affect learning in a predatory mite? Experiments to unrav el mechanisms underlying polyphagy in Hypoaspis aculeifer Acarologia 50(2):257 Sandness JN, McMurtry JA (1970) Functional re sponse of three species of Phytoseiidae (Acarina) to prey density. Can Entomol 102:692

PAGE 148

148 Sandness JN, McMurtry JA (1972) Pr ey consumption behavior of Amblyseius largoensis in relation to hunger. Can Entomol 104:461 Santos MA (1975) Functional and numeri cal responses of the predatory mite, Amblyseius fallacis to prey density. Environ Entomol 4:89 Sarkar PK, Somchoudhury AK (1989) Influence of major abiotic factors on the seasonal incidence of Raoiella indica and Tetranychus fijiensis on coconut. In: Channabasavanna GP, Viraktamath CA (eds) Progress in Acarology. Ox ford and IBH, New Delhi, India, pp 59 SAS Institute Inc (2003) SAS 9.1 for Windows. Cary, NC, USA SAS Institute Inc SAS 9.2 for Windows. Cary, NC, USA Sayed MT (1942) Contribution to the knowledge of Acarina in Egypt: 1. The genus Raoiella Hirst (Pseudotetranychidae: Tetranyc hidae). Bull Soc Ent Egypte 26:81 Sayed MT (1950) On the taxonomy of Tetranychid ae and allied genera. A new family and two new sub-families in Acarina. Proceedings of the 8th International Congress of Entomology, Stockholm, Sweden, 1012 Schausberger P, Croft BA(1999) Activity, feedin g, and development among larvae of specialist and generalist phytoseiid m ite species (Acari: Phytosei idae). Environ Entomol 28:322 Schausberger P, Walzer A, Hoffman D, Rahm ani H (2010) Food imprinting revisited: early learning in foraging predator y mites. Behaviour 147:883 Sepulveda F, Carrillo R (2008) Functiona l response of the predatory mite Chileseius camposi (Acarina: Phytoseiidae) on densities of it prey, Panonychus ulmi (Acarina: Tetranychidae). Int J Trop Biol 56(3):1255 Sheeja UM, Ramani N (2009) Feeding potential of Lasioseius sp. (Acari Mesostigmata), a promising predator of the red palm mite Raoiella indica (Acari: Tenuipalpidae). Karnataka J Agric Sci 23(3):698 Skirvin DJ, Fenlon JS (2001) Plant species modifies the functional response of Phytoseiulus persimilis ( Acari:Phytoseiidae ) to Tetranychus urticae ( Acari: Tetranychidae ): implications for biological control. Bull Entomol Res 91:61 Smith JC, Newsom LD (1970) The biology of Amblyseius fallacis (Acarina: Phytoseiidae) at various temperature and photoperiod regimes. Ann Entomol Soc Am 63:460 Solomon ME (1949) The natural control of animal populations. J Anim Ecol 18:1 Somchoudhry AK, Sarkar PK (1987) Observations on natural enemies found in association with coconut mite, Raoiella indica Hirst. Bulletin of Entomology 28:104

PAGE 149

149 Systat Software Inc (2006) SigmaPlot 10 users manual. Systat Software Inc, Point Richmond, CA. Taylor B, Rahman PM, Murphy ST, Sudheendr akumar VV (2011) Population dynamics of the red palm mite ( Raoiella indica) and phytoseiid predators on two host palm species in southwest India. Submitted to Exp Appl Acarol TDA (2008) Texas Department of Agriculture. Red Palm Mite, Raoiella indica Hirst. Available via DIALOG. http://www.agr.state.tx.us/agr/ m ain_render/0,1968,1848_27454_0_0,00.html?channelId=274 54. Accessed May 2011 Tixier M-S, Ferrero M, Okassa M, Guichou S, Kreiter S (2010) On the specific identity of specimens of Phytoseiulus longipes Evans (Mesostigmata: Phytoseiidae) showing different feeding behaviours: morphological and molecular analyses. Bull Entomol Res doi:10.1017/S0007485309990617 Thompson JN (1988) Evolutionary ecology of the relationship between oviposition preference and performance of offspring in phytopha gous insects. Entomol Exp Appl 47: 3 Uhl NW, Dransfield J (1987) Gene ra Palmarum: a classification of palms based on the work of HE Moore Jr Lawrence Kansas: Internationa l Palm Society and L. H Bailey Hortorium. Van Driesche RG, Bellows TS Jr. (1996) Biologi cal control. Kluwer Academic Publishers, Massachusetts, USA, pp 539 Van Rijn PCJ, Tanigoshi LK (1999) Po llen as food for the predatory mites Iphiseius degenerans and Neoseiulus cucumeris (Acari: Phytoseiidae): dietary range and life history. Exp Appl Acarol 23:785 Vsquez C, Quirs GM, Aponte O, Sandoval DMF (2008) First report of Raoiella indica Hirst (Acari: Tenuipalpidae) in Sout h America. Neotrop Entomol 37:739 Vet LEM, Dicke M (1992) Ecology of infochemi cal use by natural enemies in a tritrophic context. Annu Rev Entomol 37:141 Ward DB, Austin DF, Coile NC (2003) Endangered and threatened plants of Florida, ranked in order of rarity. Castanea 68:160 Watson DM, Du TY, Li M, Xiong JJ, Liu DG, Huang MD, Rae DJ, Beattie GAC (2000a) Life hystory and feeding biology of the predatory thrips, Aleurodothrips fasciapennis (Thysanoptera: Phlaeothripid ae). Bull Entomol Res 88:351 Watson DM, Du TY, Li M, Xiong JJ, Liu DG, Huang MD, Rae DJ, Beattie GAC (2000b) The effect of two prey species, Chrysomphalus aonidum and Corcyra cephalonica on the quantity of the predatory thrips, Aleurodothrips fasciapennis reared in the laboratory. BioControl 45:45

PAGE 150

150 Welbourn C (2009) Pest alert: red palm mite, Raoiella indica Hirst (Acari: Tenuipalpidae). In: Florida Department of Agriculture and C onsumer Services. Available via DIALOG. http://www.doacs.state.fl.us /pi/enpp/ento/r.indica.htm l Accessed May 2011 Xiao Y, Fadamiro HY (2010) Functional responses and prey-stage preferen ces of three species of predacious mites (Acari: Phytoseiidae) on citrus red mite, Panonychus citri (Acari: Tetranychidae). Biol Control 53:345 YadavBabu RK, Manjunatha M (2007) Seasonal incidence of mite popul ation in arecanut. Karnataka J Agric Sci 20(2):401 Yaninek JS, Herren HR (1988) Introduction and spread of the cassava green mite, Mononychellus tanajoa (Bondar) (Acari: Tetranychidae), an exotic pest in Africa, and the search for appropriate control methods: a review. Bull Entomol Res 78:1 Yue BS, Tsai JH (1996) Development, survivorship, and reproduction of Amblyseius largoensis (Acari: Phytoseiidae) on selected plant polle ns and temperatures. Environ Entomol 25:488 494 Zaher MA, Wafa AK, Yousef AA (1969) Biological studies on Raoiella indica Hirst and Phyllotetranychus aegyptiacus Sayed infesting date palm trees in U.A.R. (Acarina: Tenuipalpidae). Z Angew Entomol 63:406 Zannou ID, Negloh K, Hanna R, Houadakpode S, Sa belis MW (2010) Mite diversity in coconut habitat in West and East Af rica. XIII International Congress of Acarology, Recife, Brazil, Abstract Book, p 295 Zhang Y, Zhang Z, Lin J, Liu Q (1999) Biology of Typhlodromus bombusae (Acari: Phytoseiidae) a predator of Schizotetranychus nanjingensis (Acari: Tetranychidae) injurous to bamboo in Fujian, China. Syst Appl Acarol 4:57

PAGE 151

151 BIOGRAPHICAL SKETCH Daniel Carrillo was born in Pereira, Colo mbia. He attended high school at San Carlos School at Bogota, Colombia. Upon graduati on, he began his underg raduate studies at Universidad Nacional de Colombia in Bogota. He received his bachelo rs degree in agronomic engineering in 2005. Before graduating, he wa s employed (2003) at the Horticultural Research Center, Jorge Tadeo Lozano University (Colombia). He then moved to Florida, and obtained a Master of Science degree in entomo logy from the University of Florida in 2007. In 2008 he enrolled in a PhD. program under the supervis ion of Dr. Jorge E. Pea in the Department of Entomology and Nematology at the University of Florida, Tropical Research and Education Center, Homestead, FL. He currently is a memb er of the Entomological Society of America (ESA), Florida Entomological Society (FES), Acarological Society of America (ASA), Colombian Entomological Society (SOCOLEN), International organization for Biological Control Nearctic Regional Section (IOBC/NRS).