<%BANNER%>

Biology and Host Specificity of Tectococcus ovatus (Hemiptera: Eriococcidae), a Potential Biological Control Agent of th...


PAGE 1

BIOLOGY AND HOST SPE CIFICITY OF Tectococcus ovatus (HEMIPTERA: ERIOCOCCIDAE), A POTENTIAL BI OLOGICAL CONTROL AGENT OF THE INVASIVE STRAWBERRY GUAVA, Psidium cattleianum (MYRTACEAE), IN FLORIDA By FRANCIS JAMES WESSELS IV A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Frank J. Wessels

PAGE 3

This document is dedicated to my parents, for their support and ge nerosity throughout my educational career. Without them, this work would not have been possible.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my major professor Dr. James P. Cuda for his invaluable guidance and help throughout my degree program. I also thank my other committee members, Dr. Kenneth A. Langeland and Dr. William A. Overholt, for their comments and suggestions on my research and this manuscript.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT....................................................................................................................... ix CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................7 Psidium cattleianum Sabine.........................................................................................7 Taxonomy..............................................................................................................7 Nomenclature........................................................................................................8 Scientific name...............................................................................................8 Common names..............................................................................................9 Morphology.........................................................................................................10 Distribution..........................................................................................................12 Native distribution........................................................................................12 Worldwide distribution................................................................................13 Distribution in the United States..................................................................14 Beneficial Uses....................................................................................................16 Invasive Properties..............................................................................................17 Control Methods..................................................................................................19 Mechanical control.......................................................................................19 Chemical control..........................................................................................20 Biological control.........................................................................................20 Tectococcus ovatus Hempel.......................................................................................23 Higher Classification...........................................................................................23 Taxonomy............................................................................................................24 Life History.........................................................................................................25 Gall description............................................................................................25 Morphology..................................................................................................25 Biology.........................................................................................................28 Nutritional Ecology.............................................................................................28

PAGE 6

vi Recorded Host Range..........................................................................................30 3 LABORATORY BIOLOGY OF Tectococcus ovatus ................................................32 Introduction.................................................................................................................32 Materials and Methods...............................................................................................34 Results and Discussion...............................................................................................38 Biology................................................................................................................38 Separation of Nymphal and Adult Stages of Tectococcus ovatus .......................40 Correlation of Gall Size to Ny mphal and Adult Stages of Tectococcus ovatus ..42 Acknowledgements.....................................................................................................45 4 HOST SPECIFICITY OF Tectococcus ovatus ...........................................................46 Introduction.................................................................................................................46 Materials and Methods...............................................................................................49 Results........................................................................................................................ .52 Discussion...................................................................................................................54 Acknowledgments......................................................................................................57 5 DISCUSSION AND CONCLUSIONS......................................................................58 APPENDIX A WORLDWIDE DISTRIBUTION OF Psidium cattleianum ......................................64 B FINAL TEST PLANT LIST FOR HO ST SPECIFICITY TESTING OF Tectococcus ovatus ..........................................................................................................................70 C CHANGES TO THE FINAL TEST PL ANT LIST FOR HOST SPECIFICITY TESTING OF Tectococcus ovatus .............................................................................76 LIST OF REFERENCES...................................................................................................80 BIOGRAPHICAL SKETCH.............................................................................................89

PAGE 7

vii LIST OF TABLES Table page 2-1. Potential biologic al control agents for strawberry guava...........................................22 3-1. r2 values for multiple regressions of separate leg measurements of T. ovatus vs. the gall size. ..................................................................................................................4 3 3-2. Mean and standard deviati on of the length of the fused prothoracic trochanter/femur segment for each deve lopmental instar....................................................................44 4-1. Results of T. ovatus host specificity testing...............................................................53

PAGE 8

viii LIST OF FIGURES Figure page 2-1. Morphology of P. cattleianum ....................................................................................11 2-2. Native distribution of strawberry guava in southeastern Brazil.................................12 2-3. Countries where strawberry guava has been reported................................................14 2-4. Florida counties where vouchered specimens of strawberry guava were collected...16 2-5. Phylogeny proposed by Bourgoin and Camp bell (2002) of the hi gher classification of the Hemiptera based on morphologi cal, molecular, and fossil data....................24 2-6. Cross sectional view of the galls of T. ovatus A) Female B) Male............................26 2-7. Ferris (1957) drawings of T. ovatus A) Adult female B) First instar crawler............27 2-8. T. ovatus adult female A) Dorsal B) Ventral..............................................................27 2-9. Life stages of T. ovatus ...............................................................................................29 3-1. Cross section depicting varia tion in female leaf galls of T. ovatus ............................36 3-2. Life stages of female T. ovatus. ..................................................................................39 3-3. T. ovatus adult male....................................................................................................39 3-4. Two dimensional representation of the FASTCLUS cluster analysis........................43 3-5. Linear regression analysis of the relationship be tween the diameter of the apical base of the plant gall and the fused prot horacic trochanter /femur length of T. ovatus ....44 4-1. Test arena for host specificity experiments................................................................51

PAGE 9

ix Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science BIOLOGY AND HOST SPE CIFICITY OF Tectococcus ovatus (HEMIPTERA: ERIOCOCCIDAE), A POTENTIAL BI OLOGICAL CONTROL AGENT OF THE INVASIVE STRAWBERRY GUAVA, Psidium cattleianum (MYRTACEAE), IN FLORIDA By Francis James Wessels IV December 2005 Chair: James P. Cuda Major Department: Entomology and Nematology Strawberry guava, Psidium cattleianum Sabine, is a woody tree or shrub native to coastal southeastern Brazil. Strawberry gua va was introduced into Florida in the late 1800s as an ornamental species. The plant escaped cultivation a nd is invading natural areas throughout the southern half of the state. In addition to negative effects on Florida’s native ecosystems, strawberry guava also is a preferred host of the Caribbean fruit fly, Anastrepha suspensa Loew (Diptera: Tephritidae). The Caribbean fruit fly is a common ag ricultural pest that affects several important fruit crops. The Caribbean fruit fly can cause direct yield loss, and its presence can affect shipments to quarantine sensitive markets. In Florida, various control techniques have been used with limited succe ss. A novel approach for reducing fruit fly populations is classical biological control of their preferred na turalized host plants, such as strawberry guava.

PAGE 10

x A survey of the entomofauna associated with strawberry guava identified five potential biocontrol agents. The most pr omising was a leaf-galling scale insect Tectococcus ovatus Hempel. Large infestations of T. ovatus cause premature leaf drop and inhibit fruiting, thereby reducing fruit fly br eeding sites. It is essential to understand as much as possible about a potential biological control agent prior to release. An analysis of a series of leg measurements wa s conducted in order to determine the number of nymphal stages and discern them from th e adult. Multivariate analyses of the measurements suggest the presence of two or possibly three instars. However, a greater sample size is necessary to de termine if there truly are thr ee nymphal stages or if the results are due to outlying data points. Prior to the release of any classical biologi cal control agent, th e host specificity of the agent needs to be demons trated. No choice tests we re conducted because of the rigorous nature of the tests. In total, 57 species of plants representing 21 families were included in the host range tests. Tectococcus ovatus first instars fed on two closely related guava species, Brazilian guava ( Psidium friedrichsthalianum O. Berg), and Costa Rican guava ( Psidium guineense Sw.). Incomplete gall formation was observed on the Costa Rican guava. However, none of the T. ovatus nymphs completed their development. These non-target effects were determined to be negligible relative to the target, because the two species attacked were not native and are rarely cultivated in Florida. The results of the hos t specificity test s suggest that T. ovatus is a suitable candidate for biological control of strawberry guava in Florida.

PAGE 11

1 CHAPTER 1 INTRODUCTION Prior to the age of explor ation, plant immigration to new habitats was primarily facilitated by seed dispersal adaptations. S eeds could be transported by natural elements such as wind and water or by animal moveme nt and migration, either in the digestive tract or externally. Although these proce sses were relatively commonplace, large natural boundaries such as oceans and mountains limite d the rate of long distance introductions. With the advent of human exploration and travel, these natural boundaries were readily crossed, and an explosion of plant intr oductions began. The majority of plant introductions have been inten tional-for agricultural, industria l, or ornamental purposes (Pimentel et al. 2000). Most non-native in troductions are sust ained solely through cultivation, and do not persist on their own. However, a sm all percentage of non-native plants become established, or naturalized in th eir new habitat. Some of these naturalized plants flourish and readily disperse, posing a th reat to native ecosystems. This is because the most important direct effect of non-native species is habitat modification (Simberloff 1997). Non-native species that alter nativ e habitats through direct competition or hybridization are termed exotic invasive sp ecies (Florida Exotic Pest Plant Council [FLEPPC] 2003). Although the percentage of ex otic invasive species is relatively low compared to the number of introduced species their negative impacts are widespread and can be extremely costly (Gordon and Thomas 1997). Florida’s climate and geography make th e state prone to non-indigenous species invasions (Invasive Species Working Gr oup [ISWG] 2003). The high number of

PAGE 12

2 invasive species proliferating in the state is understandable considering that 85% of all foreign plants imported into the United States arrive through Miami International Airport (US Congress, Office of Technology Asse ssment [OTA] 1993). Because of this, approximately 31% of Florida’s flora is of non-native origin (Wunderlin 1998). Of the 25,000 introduced species cultivated in the state, 925 have escaped cultivation and become naturalized in Florida’s native ecosystems (Frank and McCoy 1995). A smaller number of these plants have expanded thei r range and are consid ered to be exotic invasive species in Florida. The exact number of these species varies depending on which source is cited. As previously men tioned, the majority of plant introductions are intentional, and Florid a’s large ornamental nursery indus try may be responsible for the introduction of a large number of exotic i nvasive species. Not surprisingly, there has been a conflict of interest between envi ronmental and commercial groups regarding which species are truly invasive, and which are not. One organization that has been formed to inform resource managers of potential problems concerning exotic invasi ve plant species within the state is the Florida Exotic Pest Plant Council (Langeland 2002). To help resource managers make informed decisions about which plants to monitor and help set priorities fo r management, FLEPPC has compiled a list of 126 naturali zed plant species to be which they consider to be exotic invasive species (FLEPPC 2003, Langeland 2002). The FLEPPC invasive species list is divided into two categories, Category I represents the most harmful species which have been shown to be altering native plant comm unities; Category II represents species that are increasing in abundance but are not aff ecting native communities to the extent of Category I species. Although the FLEPPC list provides detailed and readily accessible

PAGE 13

3 information; it has been critici zed for being too liberal in its classification. This is due to the large discrepancy between the FLEPPC lis t and state and federa l regulations (24 of the 67 Category I species and 6 out of 59 Category II species are regulated by the government) and the lack of supporting evid ence and references (Fox et al. 2004a). Despite this controversy, the Florida Nu rsery, Growers, and Landscape Association (FNGLA) and FLEPPC asked FNGLA member s to stop selling 45 potentially invasive species (Wirth et al. 2003). Although this was a good start, the organizations could not agree upon the status of 14 Category I exotic in vasive species that are commonly sold as ornamentals. These 14 species represent only 2.8% of total nursery sales in Florida. However, considering the size of the ornament al nursery industry in the state, these species have a total combined economic imp act of $59 million (Wirth et al. 2003). The FNGLA (2005) called for the creation of a singl e invasive plant list that distinguishes between Florida’s geographic regions and intend ed plant use, both of which are not taken into account by the FLEPPC list. A new list was created by the University of Florida, Institute of Food and Agricultural Sciences (IFAS) called the IFAS assessment of the status of non-native plants in Florida’s natural areas (Fox et al. 2004b). The primary objective of this assessment is to direct re search and extension at the University of Florida to be focused in the proper direct ion (Fox et al. 2004b). Although, secondary objectives are to provide additi onal information to that avai lable on other state invasive lists and identify gaps in knowledge of inva sive species (Fox et al. 2004b). This assessment standardizes the classification of the status of selected pl ants in Florida based upon geographic location (north, central, or south Florida), economic importance, and effect on the environment. The conclusions of the IFAS assessment confirmed the

PAGE 14

4 invasive status of all 14 species designated invasive by FLEPPC (IFAS 2005). Although the IFAS assessment is not a re gulatory list (Fox 2005), this list is considered to be the most complete and appropriate for invasive plant species in the state of Florida. Therefore, the IFAS assessment will be used for the purposes of defining the invasive status of a plant in Florida throughout this manuscript. Although the environmental and economic thre at exotic invasive plants pose to native ecosystems is great, invasive plants may also have indirect effects on their new habitats. One example is strawberry guava, Psidium cattleianum Sabine. This woody invasive species was brought to Florida from southeastern Brazil for the ornamental and fruit trade in the late 1800s (Gordon and Thomas 1997). Strawberry guava is one of the 14 controversial species that is still commonly sold as an orna mental throughout the state. In addition to the environmental problems caused by strawberry guava invading natural areas, the plant also serves as a major host of the Caribbean fruit fly, Anastrepha suspensa Loew (Swanson and Baranowski 1972). Swanson and Baranowski (1972) determined the host specificity of A. suspensa based on specimens reared from 84 field collected fruits in 23 families. They identified six major hosts of A. suspensa : loquat, Eriobotrya japonica (Thunb.) Lindl.; Surinam cherry, Eugenia uniflora L.; rose apple, Syzygium jambos (L.) Alst.; tropical almond, Terminalia catappa L.; common guava, Psidium guajava L.; and strawberry guava. Becaus e of this study, Nguyen et al. (1992) counted the number of A. suspensa larvae present on the frui ts of three major hosts; loquat, Surinam cherry, and strawberry guava They found that the Caribbean fruit fly occurred on strawberry guava in larger num bers from July through October compared to loquat and Surinam cherry.

PAGE 15

5 The Caribbean fruit fly is native to th e West Indies; although it immigrated to Florida multiple times, finally becoming es tablished in 1965 (Nguyen et al. 1992). The larval stage of the fly feeds on a wide vari ety of tropical and subtropical fruits. The Caribbean fruit fly is extremely polyphagous, with nearly 100 recorded hosts (Weems et al. 2005). Fruit fly damage resulting from larval feeding renders fruit unmarketable. In addition to direct imp acts, the Caribbean fruit fly ha s the potential to infest post harvest commercial fresh citrus shipments de stined for quarantine sensitive domestic or foreign markets such as California, Texas, a nd Japan. For this reason, the Caribbean fruit fly is considered a major quarantine pest, a nd over the years multiple control strategies have been developed to reduce fly densities in Florida. Prior to 1983, shipments were fumigated with ethylene dibromide to elimin ate the presence of fr uit fly larvae (Nguyen et al. 1992). In 1983, the use of ethylene di bromide for this purpose was banned by the Environmental Protection Agency (Extensi on Toxicology Network [EXTOXNET] 2005). With the banning of this chemical, the Florid a Department of Agriculture and Consumer Services (FDACS) developed the Caribbean fru it fly free protocol to certify citrus crops as fly free. This protocol is an integrated approach involvi ng the use of monitoring traps, aerial pesticide sprays, and the removal of major hosts within 1.5 miles of designated groves (FDACS 2005). Removal of major host species such as strawberry guava costs citrus growers time and money. It is the gr ower’s responsibility to remove all major hosts, even if they are not pr esent on the grower’s property. Because of this, developing alternative methods of contro lling these weeds has become a priority, in order to help growers save money and prevent legal di sputes over removing pests on neighboring properties.

PAGE 16

6 Classical biological control of the majo r hosts would be preferential. Once released, classical biological control agents do no t require any physical input aside from monitoring and they do not re quire landowner permission. In 1991, researchers with the U.S. National Parks Service and the University of Hawaii collaborated with the Federal University of Paran in Curitiba, Brazil to investigate potential biological control agents for strawberry guava (Wikler et al. 2000) This project was undertaken because strawberry guava is a serious forest pest throughout the Hawaiian archipelago. The group discovered seven potential biocontrol agen ts. After preliminar y testing, the most promising of these agents was determined to be Tectococcus ovatus Hempel, a leaf galling scale insect in the family Erioco ccidae (Wikler et al. 2000). A colony of T. ovatus was established at the Institute of Pacifi c Islands Forestry in Volcano, Hawaii. At this laboratory, host specificity tests began to evaluate T. ovatus for release in Hawaii. In 2001, researchers at the University of Florid a obtained funding from the USDA CSREES T-STAR program to investigate th e biology and host specificity of T. ovatus in Florida. This manuscript is the result of this research. The data presented in the following chapters are the result s of initial investigati ons of the biology of T. ovatus and the host specificity testing of this insect.

PAGE 17

7 CHAPTER 2 LITERATURE REVIEW Psidium cattleianum Sabine Taxonomy The Myrtaceae is a large family including approximately 150 genera and 3600 species commonly found in trop ical and subtropical climat es worldwide, although some species are also established in temper ate Australia (Cronqui st 1981). The genus Psidium is one of the larger genera in the My rtaceae containing approximately 100 species (Cronquist 1981). The genus was first desc ribed by Linnaeus, derived from the Greek word sidion, meaning pomegranate ( Punica granatum L.) due to the similar shape of the fruits (Ellshoff et al. 1995). The complete classification of strawberry guava is as follows: Class: Magnoliophyta Subclass: Rosidae Order: Myrtales Family: Myrtaceae Subfamily: Myrtoidea Genus: Psidium Species: Psidium cattleianum Sabine

PAGE 18

8 Nomenclature Scientific name Psidium cattleianum was first described by Sabine (1821); it was named in honor of the botanist William Cattley (often misspelled cattleyanum in older literature). Cattley was the first person to successfully cultivate the plant in his conservatory in Britain, claiming that his plants were grown from s eeds he received from China (Bretschneider 1898). However, some botanists co nsider the correct name to be Psidium littorale giving priority to the description by Raddi in 1823. This conclusion was reinforced by Fosberg (1941) and Merrill and Perry (1938 ). The confusion arises fro m the difference in the date of publication and the date in the section of Raddi’s desc ription of strawberry guava (Ellshoff et al. 1995). Schroeder (1946) noted that because of the confusion surrounding the actual date of publication, Raddi’s descri ption cannot be proven to be earlier than Sabine’s. Since Sabine’s description can be dated definitely, the currently accepted name should remain Psidium cattleianum while Psidium littorale is considered a junior synonym (Ellshoff et al. 1995, Schroeder 1946) Synonyms include (from Fosberg 1941, Wikler 1999): Psidium littorale Raddi, Opusc. Sci. 4: 254. 1823 t. 7 f. 2. Psidium variabile Berg, Fl. Bras. 14(1): 400. 1857. Psidium coriaceum var. obovatum Berg, 1. c. 461 t. VI. 120. Psidium coriaceum var. grandifolium Berg, 1. c. 401. Psidium coriaceum var. longipes Berg, 1. c. 402. Psidium cattleianum var. coriaceum (Berg) Kiaerskou, Enum. Myrt. Bras. 28. 1893. Episyzygium oahuense Seuss. & A. Ludwig Psidium cattleianum var. cattleianum f. lucidum Degener

PAGE 19

9 Psidium cattleianum var. littorale (Raddi) Fosb. Psidium littorale var. lucidum (Degener) Fosb. There are two accepted varieties of strawb erry guava, distinguished solely by the color of the fruit, the red-fruited variety P. cattleianum var. cattleianum and the yellowfruited variety P cattleianum var. lucidum ( P. littorale var. lucidum ) (Wikler 2000b). Both varieties are found within the native range of southeastern Brazil. The yellow variety is much more common, while the red is restricted to highe r elevations (Hodges 1988). A similar situation occurs in Hawaii, with the yellow variety dominating lower elevations in Hawaii Volcanoes National Pa rk and the red variety dominating higher elevations (Tunison 1991). In Florida, both va rieties are present, de spite the lack of a major elevation gradient (L angeland and Hall 2000). In Hawaii, there are two different shapes of the fruit of the yellow variety of strawberry guava. One type is the typical round shape, while the other has ellipsoidobconical fruit (Wagner et al. 1999). Wagner et al. (1999) consider these to be two separate varieties ( P. cattelianum var. lucidum and P. c. var. littorale ). The ellipsoidobconical variety also is seen on Mauritius, although it has no t been reported from Brazil or Florida. Common names Strawberry guava has been widely intr oduced throughout the world, and for this reason, there are a considerable number of commo n names for the species. In the US and other English speaking countries, P. cattleianum is known as strawberry guava (presumably due to the strawberry-like flavor and possibly the red color of P c var. cattleianum ), purple guava in Jamaica, Chinese guava (due to Cattley’s assumption of Chinese origin), Cattley guava, and pineappl e guava. In Brazil, the plant is known as

PAGE 20

10 ara da praia, araazeiro coroa, ara ve rmelho, and araazeiro. Additional names are araz (Uruguay), cas dulce (Costa Rica), gua yaba japonesa (Guatemala), and guayaga peruana (Venezuela) (Gomes 1983, Morton 198 7, Popenoe 1920). Other names include goiave de L’Afrique (Dominican Republic ), ara-saiyu, and guayabo amarillo (Argentina), Calcutta guava (India), goyavier of St. Martin, goyavier fraise (Guadeloupe), guayabita fresa (Cuba), and goyavier prune (Martinique) (Roig y Mesa 1953, Wikler 2000b). In Hawaii, the red variety is called waiawi ulaula, while the yellow variety is simply called waiawi (Morton 1987). Morphology Strawberry guava is an evergreen shr ub or small tree between 2 and 6 m tall, although specimens of P. cattleianum var. lucidum have been reported growing up to 12 m (Morton 1987). Stems and branches are sm ooth, gray to reddish brown in color with bark that peels in thin sheets (Fig. 2-1A ) (Langeland and Burks 1998). Young branches are round and sparsely pubescen t. Leaves are alternate, obovate to elliptic-obovate between 3.5 and 13.5 cm long, petioles appr oximately 7-10 mm long (Wagner et al. 1999, Webb et al. 1988). Leaf surface is dark green glabrous and somewhat leathery in texture; lateral veins are sli ghtly elevated but inconspicuous (Fig. 2-1B). Flowers are white, 1.5-6 cm in diameter with prominen t stamens (Morton 1987); flowers are usually solitary and borne in almost all axils of upper leaves (Fig. 2-1C) (Langeland and Burks 1998, Webb et al. 1988). Fruits are sweet tasting, slightly acidic, round or elliptical, smooth and glabrous 2-3 cm in diameter. The fruits of P. cattleianum var. lucidum are yellow to white, while those of P. cattleianum var. cattleianum are reddish to purple in color (Fig. 2-1D). The red fruits also are reported to taste better, with a more subdued

PAGE 21

11 flavor (Dehgan 1998). Seeds ar e hard, flattened and triangul ar in shape approximately 2.5 mm long (Morton 1987). Strawberry guava is a relati vely hardy species that can grow in a wide variety of soils, although they perform best in rich sandy soils (Popenoe 1920). The red form seems to better withstand colder temperatures and can survive at temperatures as low as 22 F, whereas the yellow form is mo re susceptible to cold (Morton 1987). Both forms are drought resistant, although the yellow form also can withstand short periods of flooding (Morton 1987). Figure 2-1. Morphology of P. cattleianum ; A) characteristic bark; B) obovate leaves; C) flowers (Photo credit: Jeff Hutchinson); D) yellow fruits of P cattleianum var. lucidum

PAGE 22

12 Distribution Native distribution In Brazil, where the Myrtaceae is a wide spread and diverse family, the genus Psidium is represented by nine species (Wikler 2000a). Strawberry guava is located in the Atlantic forest ecosystem in the southeastern part of the country (Fig. 2-2). Within its native range, the plant is primarily a coastal species. Strawberry guava occurs in a coastal vegetation type known as “restinga”, although it can also be found in disturbed brush fields known as “capoeiras” (Hodges 1988). The northernmost range of strawberry guava in Brazil is in the state of Espirito Santo and extends south along the coast to the northern tip of Uruguay (Fi g. 2-2) (Hodges 1988). Hodges (1988) noted the plant growing within an elevation range of 5 – 100 m; although except for cultivated plants, he did not notice the red fruited form growing wild. Figure 2-2. Native distribution of strawberry guava in southeastern Brazil (adapted from Hodges 1988, Wikler 1995).

PAGE 23

13 Worldwide distribution Strawberry guava has been intentionally intr oduced in nearly all of the countries in which it is currently found. Its attractive fruit and leaves are generally desired more as an ornamental than a fruit crop. According to Popenoe (1920), strawberry guava was originally transported from its native range in Brazil to China at an “early period”, presumably by the Portuguese. Seeds were taken to Europe in 1818 by two Englishmen, Barr and Brookes, and described as origina ting from China (Bretschneider 1898). The only reference hinting to Eur opean introduction of this plan t in Hawaii is from Degener (1932) mentioning that live plants of both fo rms of strawberry guava may have been brought to the archipelago on board the “B londe” in 1825. Similar introductions may initially have been made by European or Portuguese ships resulting in the wide distribution of the plant. Distribution information obtained from lite rature records and herbarium specimens have been compiled (Appendix A) in order to create a worldwide ma p of the distribution of strawberry guava (Fig. 2-3). However, th e distribution of strawberry guava is most likely more extensive than mentioned in the lit erature. This plant is found in all 7 major world biogeographical regions; in the Afrotropica l region it is present in multiple coastal countries of tropical Africa, and on Mada gascar and surrounding islands. In the Australasian region, strawberry guava is found in Australia and New Zealand and surrounding island archipelagos. In the East Palearctic region it is found in southern China and Taiwan. Recorded distribution is most common in the Neotropical region where the plant is found throughout its native range of Brazil, surrounding South and Central American countries and throughout the Caribbean. In the Oriental region, strawberry guava has been reported on multip le islands in the Indian Ocean and the

PAGE 24

14 Pacific and in Japan, Malaysia, and the Philippines. In the Western Palearctic region it is found on the islands of Madeira, the Azores, the British Isles, the Cape Verde Islands, various Mediterranean islands a nd, in central France In the Nearctic region strawberry guava is found in the southern portion of th e continental United St ates and on Bermuda. Figure 2-3. Countries where strawberry guava has been reported (based on literature references and herbarium specimens, see Appendix A). Distribution in the United States In 1884, approximately 3000 trees were plante d as part of a co mmercial venture in La Mesa, California, and the trees were report ed as producing heavily a half century later (Morton 1987). Additionally, a herbarium specimen was collected in 1995 from a cultivated specimen at the Quail Botanical Gardens in San Diego County, California (New York Botanical Garden [NYBG] 2005). Although strawberry guava was

PAGE 25

15 introduced into southern Calif ornia, no records suggest that the species is considered invasive in the region. The California Invasi ve Plant Council (Cal-IPC) does not list the species as being invasive in any part of the stat e (Cal-IPC 2000). Although there are a few references confir ming the presence of strawberry guava in Texas, many books on the flora of Texas do not even include references to the family Myrtaceae. Strawberry guava is listed in a state checklist of vasc ular plants; however, there is no mention of locality or whether the specimen is cultivated or growing wild (Jones et al. 1997). Strawberry guava can grow in US Department of Agriculture plant zones 9-10, which suggests that the plant is capab le of surviving in th e southern parts of Texas and along the Gulf Coast. Oddly, one herbarium specimen shows the pr esence of strawberry guava in Boone County, Missouri in 1974. This record is curious, because it is well out of the known geographic range of the species in the Un ited States. The Boone county specimen was collected by D. B. Dunn at approximately at 39.02.00N, 92.20.00W at an elevation of 740 ft. (Missouri Botanical Garden [MBG] 2005). However, the collection data of the specimen are rather vague regarding whether or not this specimen was cultivated. The presence of strawberry guava in Hawaii and its subsequent effects on the native ecosystems of the islands has been well documented in the literature. The weed is considered the worst invasive plant in Hawaii and is present on almost all of the major islands (Smith 1985). The plant forms dense st ands in disturbed roadsides and in intact forests where it effectively out-competes nativ e species. Although the fruits are attractive to birds and mammals, the feral pig ( Sus scrofa L.) has been shown to be a major dispersal agent of the plant in Hawaii (Diong 1982).

PAGE 26

16 In Florida, strawberry guava was first listed as an ornamental in the 1887-1888 Catalog and Price List for Royal Palm Nurs eries (Langeland and Hall 2000). By 1956, it had escaped cultivation and wa s reported growing wild (Bar rett 1956). Strawberry guava is currently naturalized in the central and southern parts of the state. Voucher specimens indicate the presence of the plant in 18 counties (Fig. 2-4) (Wunderlin and Hansen 2004). However, the actual distribution is most likely more extensive, and needs to be confirmed with additional voucher specimens. Figure 2-4. Florida counties wh ere vouchered specimens of strawberry guava were collected (Wunderlin and Hansen 2004). Beneficial Uses The fruit of strawberry guava is said to be superior in taste to that of the common guava, Psidium guajava L., and is preferred by many (Popenoe 1920). The fruits can be eaten fresh, but are often contaminated by variou s fruit flies in Florida and Hawaii. Most often the fruit is made into jelly, jam, bu tter, paste, punch, and sherbet (Morton 1987). Although, the species is cultivated in many parts of the world, no distinct cultivars exist. In Florida, strawberry guava is primarily grown as an ornamental or in hedges around

PAGE 27

17 parking lots and other areas (Gilman a nd Watson 1994). In Brazil, the wood is sometimes collected for firewood (Hodges 1988, Wagner et al. 1999). There is little promise of commercial fruit production due to the delicate skin of the ripe fruit, which would require careful shi pping (Schroeder and Coit 1944). In addition, Schroeder and Coit (1944) note that the shelf life of the fresh fruit is too short for a commercial production, lasting only 3-4 days without refrigera tion. The largest recorded commercial planting was in La Mesa, Californi a. A local farmer planted approximately 3000 trees on a 5 acre plot. In 1943, produc tion was approximately 30 tons. The majority of the fruit was processed into pa ste for sale; however, some fresh fruit was marketed locally (Schroeder and Coit 1944) Despite the attempts at commercial production, strawberry guava seems to have a more marketable potential as an ornamental hedge or dooryard tree. Invasive Properties Plants that are invasive have a variety of characteristics that aid in their dominance of new habitats. Many authors have tried to generalize these characteristics in order to help predict which species may be pote ntial problems (Noble 1989, Reichard and Hamilton 1997). Although there are characteristics that many invasive plants share, generally these are poor predictors of whet her or not a species will be invasive. Determining the invasiveness of a species base d on biological attri butes is probably too simplistic. Sakai et al. (2001) argues that the invasiveness of a plant depends on a combination of ecological, genetic, and evolut ionary factors. A combination of factors most likely aids strawberry guava in its ab ility to out compete native plants in areas where it is introduced.

PAGE 28

18 Strawberry guava has a high tolerance to environmental heterogeneity, it is tolerant of moderate to highly acidic soils, and can tolerate th e heavy litter fall present in Hawaiian rainforests (Huenneke and Vitousek 19 90). The plant also can grow in a wide variety of soils including: ro cky soil, clay, sandy loam, and wet forest soils (Diong 1983). Strawberry guava is more cold tolerant th an the common guava, and has been reported withstanding temperatures as low as 22 degr ees (Schroeder and Co it 1944). Strawberry guava also is highly shade tolerant, a trait which can help seedlings grow up through the understory and overtake native vegetation (Tunison 1991). Another trait aiding to the invasiveness of strawberry guava is its ability to reproduce both sexually and asexually. Hu enneke and Vitousek (1990) found that asexual plants produced by clona l root suckering had larger leaves and produced a greater total leaf area than seedlings. This clona l behavior most likely contributes to the formation of dense monotypic stands of strawberry guava which are common in invaded habitats. These dense stands exclude native vegetation and may significantly alter native plant and animal communities. Strawberry gua va is a prolific fruiter with a high seed count. Germination rates also are very hi gh; Huenneke and Vit ousek (1990) recorded laboratory germination rates between 6080% and field germination rates of approximately 56%. The fruits of strawberry guava are at tractive to birds and other frugiverous animals, which aid in the dispersal of the pl ant. In Hawaii, an interesting mutualistic relationship between two invasi ve species occurs between strawberry guava and the feral pig, Sus scrofa L. (Diong 1983). Diong (1983) found that strawberry guava was a preferred food of the feral pig; during the fr uiting season, strawberry guava represented

PAGE 29

19 approximately 78.3% of the pig’s stomach conten ts. Strawberry guava benefits from the relationship by being carried to new habitats and deposited in nutrient rich pig droppings. Frequently, the pig droppings were deposited on soil disturbed by the rooting activity of the feral pig. Although seed vi ability was not affected by pa ssage through the pig’s gut, gut treatment did result in earlier germina tion time compared to untreated seeds (Diong 1983). In Hawaii, introduced agricultural livestock, such as goats, sheep, and cattle, also have been reported feeding on strawberry guava (MacCaughey 1917). In addition, a variety of birds have been reported as feedi ng on the fruits of strawberry guava, e.g. the laced-necked dove ( Streptopelia chinensis (Scopoli), mynah bird ( Acridotheres tristis L.), rice bird ( Lonchura punctulata L.), house sparrow ( Passer domesticus L.), the melodius laughing thrush ( Garrulax canorus Hwamei (Cheng), Japanese white eye ( Zosterops japonicus Temminck and Schlegel), and the red-billed leiothrix ( Leiothrix lutea Maier & Bowmaker), in addition to many others (Diong 1983, MacCaughey 1917). Other animals reported to consume strawberry guava fru its include mongooses, bats (Diong 1983), and squirrels (B. Overholt, University of Florida, IRREC – pers. comm.). Control Methods Mechanical control It is important for public land managers in charge of parks and preserves to remove invasive species from their properties. In many cases, the most cost effective method of removing small infestations is mechanical control (Tunison 1991). Young plants and saplings originating from seed can be uproot ed either by hand or with a weed wrench (Tunison 1991). The uprooted plants must be disposed of, because they may re-root if left on the ground. In extreme cases, mechanical control has been us ed to control large

PAGE 30

20 infestations. In Florida, at Jonathan Dickinson State Park, the strawberry guava infestation was so dense that it was necessary to bu lldoze and restore a three acre portion of the park (Langeland and Hall 2000). Chemical control The most common method of controlling w oody weed species is by using chemical herbicides, or a combination of mechanical and chemical control (cut stump, basal frill treatments). Strawberry guava is sensitive to picloram (Tordon, Grazon, Dow AgroSciences), dicamba (Banvel, Velsicol Chemical Corp.), glyphosate (Roundup, Monsanto Co.; Accord, Rodeo, Dow Agro Sciences), and triclopyr (Garlon, Dow AgroSciences) (Tunison 1991). Pratt et al. (1994 ) tested the efficacy of 4 herbicides for managing strawberry guava in Haleakala Nati onal Park, Hawaii. They found that soil treatments with Velpar (h exazinone, 90% active ingredie nt, DuPont) and Spike 20P (tebuthiuron, 20% active ingred ient, Elanco now Dow AgroSciences) were ineffective. The two most effective herbicides were 2 formulations of triclopyr produced by Dow AgroSciences, Garlon 3A (0.36 kg/l) and Garlon 4 (0.48 kg/l) applied with the cut stump method. These chemical control measur es work reasonably well for managing strawberry guava infestations within the ma intained borders of parks and preserves. However, chemical control methods are ofte n non-selective, can be expensive, because re-treatment is often necessary, and are not a long-term solution to the problem. Biological control A feasible long term control strategy for strawberry guava is biological control (Smith 1985). Biological control is defined as the use of natural enemies to reduce the numbers of pest organisms. Biological control is not an er adication technique but rather a management strategy. There are many t ypes of biological control, ranging from

PAGE 31

21 augmentative releases of natural enemies to the traditional approach of classical biological control. Classical biological control refers to the introduction of specialist natural enemies from the native range of an a dventive pest, with the intention of reducing the pest population density. For the remai nder of this manuscrip t, unless otherwise mentioned, the terms biological control and biocontrol will be considered synonymous with classical biological control. The tropical climate and geographically isol ated nature of the Hawaiian archipelago has made it an ideal habitat for introduced exotic plan t species to proliferate. In the early 1980s, the National Park Service and the Un iversity of Hawaii undertook a project to prioritize the potential of invasive plant spec ies for biological control (Gardner and Davis 1982). Initially, it was thought that strawberry guava was a poor candidate due to its close relationship to the common guava, for wh ich a small commercial market exists in Florida and Hawaii (Gardner and Davis 1982) In addition, the family Myrtaceae contains various genera whic h are commercially or ecologically important in both states; for example Eucalyptus Eugenia Pimenta and Syzygium Prior to 1988, the only mention of predators or potential pathogens fo r strawberry guava was a parasitic alga of the genus Cephaleuros which also attacks common guava (Marlatt 1980, Marlatt and Alfieri 1981). In the mid 1980s, the National Park Service and the University of Hawaii funded initial explorations for potential biologi cal control agents of strawberry guava in Brazil (Hodges 1988). The object ive of this study was to determine the native distribution of strawberry gua va, to make local contacts for future collaboration, and to determine kinds and relative impacts of pr edators and pathogens on strawberry guava. No promising pathogens were discovered, alt hough a wide variety of insects were found

PAGE 32

22 to attack strawberry guava (Hodges 1988). Due to the wide distribution of the species in Brazil, and the numerous insects found associ ated with the plant, it was recommended that local entomologists be c ontracted to perform the initial survey of biological control agents. As a result, the National Park Service and the University of Hawaii formed a collaboration with the Paran Forestry Founda tion and the Federal University of Paran, Brazil. The Brazilian entomologists undertook the initial explorati on for insects that attacked strawberry guava within its native range. Table 2-1. Potential biological control agents for strawberry guava (Wikler et al. 2000). AGENT TAXONOMY CONTROL POTENTIAL Dasineura gigantea Angelo and Maia (Diptera: Cecidomyiidae) Good– bud galling species Lamprosoma azureum Germar (Coleoptera: Chrysomelidae) Poornon-target effects Unidentified Psyllid (Hemiptera: Psy llidae) Goodleaf galling species Tectococcus ovatus Hempel (Hemiptera: Eriococcid ae) Goodleaf galling species Haplostegus epimelas Konow (Hymenoptera: Pergidae) Poornon-target effects Sycophilia sp. (Hymenoptera: Eurytomidae) Goodseed galling species Eurytoma sp. (Hymenoptera: Eurytomidae) Goodstem galling species The Brazilian researchers identified seven potential biological c ontrol agents (Table 2-1) (Wikler et al. 2000). Two of these species were determined unsuitable because of non-target effects. In in itial studies, the sawfly Haplostegus epimelas Konow attacked common guava, and the chrysomelid Lamprosoma azureum Germar attacked common guava and other myrtaceous species (Wikler et al. 2000). Of the remaining five agents, it was determined that the leaf galling eriococcid Tectococcus ovatus was the most promising agent because of the type of dama ge inflicted and the ease of handling (Wikler et al. 2000). A shipment of T. ovatus was then sent to the Institute of Pacific Islands Forestry in Volcano, Hawaii to establish a la boratory colony for host specificity testing.

PAGE 33

23 In 2001, the University of Florid a initiated a biocontrol progr am against strawberry guava in the state. Shipments of T. ovatus were sent from the Hawaii colony to the FDACS Division of Plant Industry, Florid a Biological Control Laboratory in Gainesville, Florida, to establish a colony for host specificity testing. Tectococcus ovatus Hempel Higher Classification The higher taxonomic classification of the Order Hemiptera has historically been one of great debate. The term higher classification is in reference to the ordinal and subordinal taxonomic level. The Order Hemi ptera often has been divided into two separate orders: the Hemiptera and the Hom optera (Borror et al. 1989). This earlier classification scheme was based solely on mor phological characteristics. With the advent of molecular systematics, many author s suggested that Homoptera was not a monophyletic group (Campbell et al. 1995, Schuh a nd Slater 1995, Sorensen et al. 1995). Recent molecular data supports this, showing that the order Homoptera is paraphyletic. Currently, most taxonomists agree and combin e the traditional Or ders Homoptera and Hemiptera into one order, the Hemiptera ( sensu lato ). This can be confusing, therefore unless otherwise mentioned in this manuscript; Hemiptera will be used in the broad sense ( s.l .). A debate also exists regarding the or ganization of the suborders. The Order Hemiptera is usually split into 4 or 5 s uborders; and different authors often suggest different names for the suborders which furthe r confuse the issue. To date, the most complete treatment of the higher classificat ion of the Hemiptera has been compiled by Bourgoin and Campbell (2002). Bourgoin a nd Campbell (2002) built their phylogeny, using a combination of morphological, molecu lar, and fossil data. They divide the

PAGE 34

24 Hemiptera into 5 suborders; the Sterno rrhyncha, Fulgoromorpha, Cicadomorpha, Coleorrhyncha, and the Heteroptera (Fig. 2-5). Tectococcus ovatus belongs to the family Eriococcidae. This family is grouped under the suborder Sternorrhyncha, which also contains the psyllids and aphids. Most au thors agree on the placement of the scales, psyllids, and aphids into the Sternorr hyncha (Bourgoin and Campbell 2002, Campbell et al. 1995, Schuh and Slater 1995, Sorensen et al. 1995). Figure 2-5. Phylogeny proposed by Bourgoi n and Campbell (2002) for the higher classification of the Hemiptera based on morphological, molecular, and fossil data. Taxonomy Members of the family Eriococcidae are commonly called felt scales or felted scales. The family contains approxima tely 50 genera and 350 species (Hoy 1963). Aside from the absence of paired anal plat es, there are very few defining characteristics of the family. In fact, the family seems to be comprised of a collection of unrelated groups (Rung et al. 2005). Recent molecu lar data using small subunit rDNA supports this conclusion, rendering the family paraphyletic (Cook et al. 2002). Although these findings have serious nomenclatural implicati ons, Cook et al. (2002) regard their data as

PAGE 35

25 preliminary and suggest that the group rema in intact until more extensive studies are conducted. The currently accepted classification scheme for T. ovatus is as follows: Class: Insecta Order: Hemiptera Suborder: Sternorrhyncha Superfamily: Coccoidea Family: Eriococcidae Genus: Tectococcus Species: Tectococcus ovatus Hempel Life History Gall description Leaf galls of T. ovatus each contain one insect and are visible on both sides of the leaf. There is one opening per gall and this is at the apical portion of the gall. The opening of the gall is formed on the side of the leaf upon which th e insect originally initiates feeding (Vitorino et al. 2000). Th e galls are generally acuminate on both sides of the leaf; occasionally the gall may be acumi nate only on the side of the leaf with the opening and convex on the other side. The inside of the gall is flat and covered with a fine powdery wax (Vitorino et al. 2000). The size of the gall is variable and depends on the developmental stage and sex of the insect. The galls are sexually dimorphic; the base of the female gall is much wider th an that of the male (Fig. 2-6). Morphology The genus Tectococcus is monotypic. Tectococcus ovatus was originally described by the Brazilian entomologist Hempel in 1900. Ferris (1957) described and illustrated the adult female and first instar “crawler” st age (Fig. 2-7). Adult males are typically not

PAGE 36

26 described in detail because they are rarely collected in the field and are not used for species identification. Tectococcus ovatus is a small species approximately 1.5 mm long, ovate in form with the caudal end acuminate (Fig. 2-8). Derm is membranous and pink to brown in color, dusted in a fine white powder (Vitorino et al. 2000). Legs are present and well developed although the adult female is relatively sessile, sp ending her entire life within the confines of a leaf gall. The an tennae are six segmented with the first joint being the longest. The major distinguishing feat ure of the species is the small, slightly sclerotized and hairless anal plates (Ferris 1957, Hempel 1901). The males are slight and narrow, either pink or light brown. Antennae are long and slender, approximately half of the length of the body. Males have narrow legs and one pair of wings; they are capable of weak flight. Figure 2-6. Cross sectional view of the galls of T. ovatus A) Female B) Male.

PAGE 37

27 Figure 2-7. Ferris (1957) drawings of T. ovatus A) Adult female B) First instar crawler. Figure 2-8. T. ovatus adult female A) Dorsal B) Vent ral (scale bars represent 1 mm).

PAGE 38

28 Biology The first instar of T. ovatus is the mobile stage of the insect and they are commonly called crawlers. Upon hatching, T. ovatus crawlers search the plant for suitable feeding sites, the ideal site usually being new flus h. Once a suitable site is found, the insect begins feeding and becomes sessile. Th e plant responds by forming a gall around the insect (Fig. 2-9 A, B, and C) Galls are typically formed on leaves, although they may also form on floral buds, young branches and developing fruit (Vitori no et al. 2000). Gall formation always begins on the same side of the leaf where the insect begins feeding. Females are facultatively part henogenetic, although there is at least one alternation of generations per year (Vitorino et al. 2000). The female remains in the gall for her entire life, whereas males are mobile in their adu lt stage. Once a female reaches the adult stage, she begins producing eggs within the ga ll. The eggs are then extruded in a cottony wax through the opening in the gall (Fig. 2-9 D). This wax pr obably aids in the dispersal of the eggs via wind. Nutritional Ecology The majority of phytophogous insects consume large amounts of plant tissues or fluids. To fulfill their nutri tional requirements, these insect s must move about their host, feeding at multiple sites. Gall forming insect s are unique in that they remain sessile, and feed only on specialized nutri tive cells that line the gall chamber (Dreger-Jauffret and Shorthouse 1992). The gall former stimulates an abnormal growth process in its host, from which it receives both nutrition a nd protection (Abrahamson and Weis 1987). Because of this highly specialized relationshi p between the herbivore and its host, most gall-inducing insects are highly host specific. The reduced risk of non-target effects

PAGE 39

29 makes gall forming insect’s ideal weed biolog ical control agents (Harris and Shorthouse 1996). Figure 2-9. Life stages of T. ovatus ; A) gall initiation around fi rst instar crawlers; B) close up of leaf galls; C) view of galls on multiple leaves; D) eggs extruded in cotton-like wax. (Photo credit: M. Tracy Johnson). Plant galls can be divided into two basic groups, organoid and hi stioid (Rohfritsch 1992). Organoid galls differ slightly from the normal plant growth pa ttern while histioid galls alter the basic growth patter ns of the host. The galls of T. ovatus are histioid, or more specifically they are categorized as pros oplasmic. Prosoplasmic refers to a histioid gall which is highly organized and displays tissue differentiation (Rohfritsch 1992). Galls of the Eriococcidae, how ever, typically have a less ch aracteristic nutritive tissue (Rohfritsch 1992).

PAGE 40

30 The induction of abnormal cell growth by gall forming insects results in a metabolic nutrient sink (Harris and Shorthouse 1996). In addition to depriving normal plant cells of nutrients, high densities of T. ovatus leaf galls may inhibit photosynthesis. Heavy infestations of T. ovatus result in premature leaf drop and in some cases complete defoliation (Vitorino et al. 2000). Th ere is anecdotal evidence that T. ovatus infestations also may reduce fruit production, reducing bot h the seed bank and potential fruit fly breeding sites (Vito rino et al. 2000). Recorded Host Range When first described by Hempel (1900), th e only reference to the host plant was that it belonged to the family Myrtaceae. The majority of the older literature also is vague regarding host species, mentioning onl y the family Myrtaceae (Da Costa Lima 1927, Hempel 1912, Lepage 1938, MacGillivray 1921). In a catalogue of Brazilian insects published in 1936, the host record s are more specific, although only common names are cited (Da Costa Lima 1936). Da Co sta Lima (1936) lists hosts as araazeiro (strawberry guava) and a plant known as “embir a”. Subsequent inve stigation revealed that the name “embira” refers to two diffe rent species, in two different families, Daphnopsis racemosa Griseb. (Thymelaeceae) and Rollinia salicifolia Schltdl. (Annonaceae). This reference most likely led to the inclusion of D. racemosa as a host by Hoy (1963) in his catalogue of the Erioco ccidae of the world. This reference is probably erroneous and may be due to confusion between T. ovatus and its relative Pseudotectococcus anonae Hempel (Johnson 2005). Whether the host record is erroneous or not, members of the genera Daphnopsis and Rollinia from Florida and the Caribbean would be appropriate for inclusion in a host specificity test plant list.

PAGE 41

31 In field observations, T. ovatus has been reported on ot her members of the genus Psidium in Brazil. Psidium longipetiolatum Legrand and Psidium spathulatum Mattos both have been reported as being attacked by T. ovatus (Vitorino et al. 2000). These two species are not present in the continental Unit ed States and therefor e are in no danger of being attacked by T. ovatus should it be approved for release in Florida.

PAGE 42

32 CHAPTER 3 LABORATORY BIOLOGY OF Tectococcus ovatus Introduction Tectococcus ovatus Hempel is a leaf galling scale insect (Eriococcidae) native to the coastal regions of southeas tern Brazil. The distributi on of the insect is closely correlated with the native range of it s primary host plant, strawberry guava, Psidium cattleianum Sabine (Myrtaceae); ranging from the Brazilian state of Espi rito Santo in the north to northern Uruguay in the south (Wik ler 1995). Strawberry guava was originally imported into Florida for the ornamental frui t trade in the late 1800s (Langeland and Hall 2000). However, it escaped cultivation and is now considered a highly invasive natural areas weed in the state (IFAS 2005). In addition to invading native plant communities and altering the natural ecological balance of plant and animal communities in Florida, strawberry guava also is considered one of the major hosts of the Caribbean fruit fly, Anastrepha suspensa Loew (Tephritidae) (Nguyen et al. 1992, Swanson and Baranowski 1972). Native to the West Indies, the Caribbean fruit fly is a highly polyphagous pest species with nearly 100 reco rded hosts (Weems et al. 2005). Populations of the Caribbean fruit fly eventually became establis hed in central and s outhern Florida in 1965 (Swanson and Baranowski 1972). This establishment was largely ignored until 1968, when the fly was discovered in commercial grapefruit, Citrus x paradisi Macfad. (Greany and Riherd 1993). To eliminate the spread of this pest, many domestic and foreign markets initiated quarantines on fresh citrus shipments from Florida. Multiple

PAGE 43

33 management strategies are currently being used to combat Caribbean fruit fly infestations in the state. These strategies include the sterile insect technique, classical biological control, and the development of the fruit flyfree protocol, an inte grated approach to certify citrus crops as fly-free (Baranowski et al. 1993). The fru it fly-free protocol involves a combination of tr apping, baiting, spraying, and th e removal of major hosts (including strawberry guava) from surr ounding areas (FDACS 2005). Compliance with this protocol can be problematic because the citrus grower is res ponsible for removal of major hosts from adjacent properties, ev en if they do not own the property (FDACS 2005). Controlling strawberry guava infestations mechanically and chemically are viable options for easily accessible plants. Howe ver, these methods are not practical in environmentally sensitive areas (e.g., in natu ral areas and state parks and preserves). Classical biological control of strawberry guava is ideal because once released, biological control agents are self sustaining, can locate less accessible plants, and do not require landowner permission. A colony of T. ovatus was established at the Florida Department of Agricultural and Consumer Services (FDAC S), Division of Plant Industr y (DPI) Florida Biological Control Laboratory (FBCL) in Gainesville, FL. Currently, this insect is under investigation for potential rel ease as a biological control agent for strawberry guava in Florida. High infestations of T. ovatus act as a nutrient sink. Diverting nutrients from plant growth and reproduction can cause pr emature leaf drop, may reduce photosynthesis, and inhibit fruit production, ultimately reduci ng fruit fly breeding sites (Vitorino et al. 2000).

PAGE 44

34 Prior to the release of any classical biol ogical control agent, it is necessary to understand the biology and ecology of the agent. The purpose of this study was to make detailed biological observations on T. ovatus in order to expand on the limited information available in the literature regard ing this species. An important factor in understanding the developmenta l biology of an insect is determining the number of instars and distinguishing these from the adul t. The soft bodied nature of this insect makes it difficult to determine the number of instars by traditional means. Because of this, multivariate analyses of multiple leg m easurements were conducted. The legs were selected for these analyses because they ar e sclerotized and readily identifiable. The majority of the life cycle of this species occurs inside of a pr otective plant gall. Therefore, in order to better understand the life history of T. ovatus a linear regression was constructed to determine if a correlation exists between the size of the leaf gall and the life stage of the insect (excluding the egg). This information may be useful for studies where destructive sampling of specimens is not a feasible option, for example, the construction of age-specific life tables. Materials and Methods Tectococcus ovatus specimens used in this study were obtained from the laboratory colony maintained at the FDACS DPI Florida Biological Control Laboratory in Gainesville, FL. All biological observati ons were made from insects reared in this colony. The colony was maintained in acrylic cylinders, 46 cm tall and 15 cm in diameter. These cylinders were placed over the host plants and the bottom was partially buried in the soil to prevent T. ovatus from escaping. The acrylic cylinders were ventilated by six holes 6 cm in diameter. Th e ventilation holes and the top of the cylinder were covered with a fine mesh, with a sc reen size of 150 x 150 (Green.tek Inc.,

PAGE 45

35 Edgerton, WI). The colony was maintained under natural light conditions in a quarantine greenhouse, supplemented with fluorescent lig ht (40 Watts) set on a 14:10 light: dark photoperiod. Average colony temperature in side of the cylinders was 28.88 1.61 C and the average humidity was 66.13 6.95 %. In sects were reared on the yellow fruiting variety of strawberry guava ( P. cattleianum var. lucidum ) either grown from seeds collected in the field or purch ased from nurseries within Flor ida. The yellow variety was selected because preliminar y studies indicated that T. ovatus preferred this to the red fruiting variety of strawberry guava ( P. cattleianum var. cattleianum ). Five specimens of T. ovatus were dissected from leaf galls every other day for 30 days during development, resulting in a to tal of 75 specimens. This method of sampling was chosen to ensure that every developmen tal stage was included in the analysis. During the slide mounting process, 13 specimens were lost or damaged. All 13 specimens that had missing data were elimin ated from this study, leaving a total of 62 specimens to be analyzed. Before specimens were dissected from thei r protective plant galls the width of the insect gall was measured using micrometer ca lipers. These measurements were taken in order to determine if a correlation exists between the gall size of T. ovatus and the various life stages after hatching. Due to va riation between galls, th e portion of the gall measured was the diameter at the base of the apical portion (Fig. 3-1). Preliminary investigation indicated that th is portion of the gall had the least amount of variation. The specimens were then slide mounted accordi ng to a protocol modified from Wilkey (1992). Prior to slide mounting the specimens were cleared in 10% KOH for 24 to 48 h. Specimens were then stained with #6379 doubl e stain (BioQuip Products, Inc. Rancho

PAGE 46

36 Dominguez, CA), and transferred into 95% Et OH (15 min) and then submerged in clove oil (30 min) (Ward’s Natural Science, Rocheste r, NY). Canada balsam (Fisher Scientific Co., Pittsburgh, PA) was used as the slide mount ing medium. The slides used in this study were 75 x 25 mm Fisherbrand plain pre-cleaned slides (Fisher Scientific Co., Pittsburgh, PA) used with Fisherbrand (Fisher Scientific Co., Pittsburgh, PA) 12 mm circular cover glass. Figure 3-1. Cross section depicting va riation in female leaf galls of T. ovatus Arrows indicate the diameter of the apical base of the gall where measurements were taken. In order to determine the number of ny mphal stages and distinguish them from the adult, multiple leg measurements were recorded for each insect. The legs were selected as the identifying feature because they are heavily sclerotized and readily identifiable; T. ovatus is a soft bodied insect and slid e mounting can alter the dimension of non-sclerotized parts. Due to the discont inuous nature of insect development, the number of instars and the adult can be dete rmined by comparing variation of multiple leg

PAGE 47

37 measurements. The length of the fused tr ochanter/femur segment, the tibia, and the tarsus was recorded for one the pairs of each leg (prothorcacic, mesothoracic, and metathoracic). In addition, the width of the fu sed trochanter/femur and tibia was recorded at their widest points on each leg. This proced ure resulted in a total of 15 measurements for each specimen. The slide mounted speci mens were digitally photographed with a JVC model KY-F70B 3-CCD digital camera (JVC Americas Corp.) mated to a Leica DMLB compound microscope (Lei ca Microsystems AG) with a Diagnostic Instruments T-49C 0.45x c-mount coupler (Diagnostic Inst ruments, Sterling Heights, MI). The Syncroscopy Auto-Montage software (S ynoptics Ltd., Frederick, MD) was used to measure the images at a resolution of 1360 x 1024 pixels. This system also was used to measure the eggs of T. ovatus. Measurements were recorded digitally in order to reduce error (the microscope only had to be ca librated once); this method also is less time consuming than using an ocular micrometer. Data were analyzed using the SAS statistical software (S AS Institute Inc., Cary, NC). Due to the number of dimensions measured (15 observations per insect; n = 62), a principal components analysis was performe d using PROC PRINCOMP This procedure reduced the dimensions of the data by derivi ng a small number of linear combinations (in this case 2 principal components) from the da ta. Next, a cluster an alysis was performed on the results of the principal components an alysis using PROC FASTCLUS in order to delineate distinct clusters of observations. Due to the discontinuous nature of insect growth, these data clusters were used to di fferentiate the number of instars and separate them from the adult stage. PROC FASTCLUS also calculated th e mean and standard deviation for all 15 morphometric parameters in each cluster. Finally, to obtain a better

PAGE 48

38 graphical representation of the clusters, a canonical analysis was performed using PROC CANDISC. This analysis transformed the da ta from the FASTCLUS analysis into two canonical variables (Can1 and Can2). In order to correlate gall size to a part icular life stage (excluding the egg), a regression analysis was performed using Microsoft Excel (Microsoft Corporation, Redmond, WA). The diameter at the base of th e apical portion of the gall was correlated with the length of the fused prothoracic trochanter/femur. The developmental instar can then be determined by comparing the prot horacic trochanter/femur length measurement with the mean length provided by the FASTCLUS analysis (see Table.3-2 in Results and Discussion). Results and Discussion Biology Tectococcus ovatus has a simple life cycle which has been previously described by Vitorino et al. (2000). Eggs are deposited in side the gall of the female, and are then extruded from the gall opening in a filament ous waxy secretion which may aid in their dispersal by wind. The eggs are oval in shape and range in color from nearly white to a light yellow (Fig. 3-2). Av erage egg length is 0.216 0.008 mm and average width is 0.115 0.006 mm (n = 20). Upon hatching, the mob ile first instar or “crawler” disperses on the plant in search of a su itable feeding site. Ideal f eeding sites are young flushes of leaf growth. Vitorino et al. (2000), however, mentions that galls also can form on floral buds, young branches, or developing fruit. On ce a suitable feeding site is found, feeding elicits a plant response to form a gall around th e sessile insect. The female will spend the rest of her life within the c onfines of this protective gall. However, the winged male is mobile as an adult (Fig. 3-3).

PAGE 49

39 Figure 3-2. Developmenta l stages of female T. ovatus ; A) egg (abnormal coloration due to preservation in 70% EtOH); B) first instar “crawler”; C) second instar; D) adult (possible third in star not depicted). Figure 3-3. T. ovatus adult male (scale bar represents 0.5 mm).

PAGE 50

40 Vitorino et al. (2000) descri bes reproduction as facultativ ely parthenogenetic with at least one alternation of generations per ye ar. Prior to this study, copulation has never been observed. In this study, copulation was observed only once in the laboratory. Prior to copulation, the male inspected the opening of the gall with his antennae, presumably looking for a receptive female. The receptive fe male partially exposed her posterior end from the gall. The male turned around and rubbed the posterior portion of his abdomen around the anal area of the female in an a pparently random writhing motion. Copulation lasted for approximately 40 seconds. After copulation, the male co ntinued searching the leaf, possibly for other females, and the mated female re-entered her gall. Separation of Nymphal and Adult Stages of Tectococcus ovatus A traditional method of determining the num ber of instars of an insect is by constructing a frequency histogram of a particular body measurement (Kishi 1971). Typical measurements for this type of study are inter-ocular distance or some other head capsule measurement; instars are identified as distinct peaks in the histogram (Daly 1985). This method works reasonably well in ca ses where the data produce discrete nonoverlapping peaks. To check if any instars were missed during visu al inspection of the histogram, the logarithms of the means for each mode are plotted against the presumed number of instars. If all instars are included, a straight line should be observed (Daly 1985). This analysis is based on the predicti on of Dyar (1890) that the head capsule size increases by a constant geometric progre ssion every molt, commonly known as Dyar’s law. However, this type of analysis only looks at one dimension of growth, and it has been shown that the limited amount of dimens ions may result in the misinterpretation of the number of instars, particularly if the peak s are not discrete (Kishi 1971; Schmidt et al.

PAGE 51

41 1977). In addition, Dyar’s law does not necessa rily hold true for every insect, or even within a single family (Daly 1985, Gaines and Campbell 1935). Describing morphological vari ation in biological organi sms using a multivariate morphometric analysis is a re latively old technique that ha s been widely applied (Daly 1985). The term multivariate morphometric anal ysis is used to describe any multivariate statistical procedure used to describe relationships between measurements of a biological organism. Blair et al. (1964) indicates that multivariate morphometric analyses have potential for describing mo rphological variation in difficult groups such as the Coccoidea. Blair et al. ( 1964) looked at the variation in a homogenous population of Coccus hesperidum L. based upon the analysis of multiple measurements of the legs and antennae. Similarly, Boratynski and Davies (1971) analyzed multiple morphometric characters to describe taxonomic variati on in male coccids. Insects do not grow continuously but rather in disc ontinuous steps or instars. Th erefore, it should be possible to measure the average size increase of each in star of a particular species by using this type of analysis. By analyzing the 15 separate leg measur ements per specimen, it was possible to determine the number of life st ages (excluding the egg). Th e results of the FASTCLUS cluster analysis of the principal components indi cate that there are three distinct clusters. These clusters may represent two instars and the adult. However, the analysis also identified a weak fourth cluster. This may indicate the presence of a supernumerary instar in addition to the adu lt. The transformation produced from the canonical analysis provides a better display of these results (Fig. 3-4). The first three clusters are clearly visible; while the fourth cluster is shown by th e five star shaped points on the right of the

PAGE 52

42 graph. Typically, females in the family Erio coccidae have two instars in addition to the adult (Stehr 1991). There are a few possible explanations for the potential presence of a fourth cluster. There may be a wide range of size variation in th e size of the adult of T. ovatus In this case, the weak fourth cluster could be the result of outlying data from measurements of extremely large adults. Ou tlying data are more difficult to identify when comparing data sets with several dime nsions such as this (15 observations per specimen). The presence of a fourth cluster co uld also be due to va riation in the initial measurements. The measurements taken fr om the Auto-montage image were in two dimensions; slight errors in the measurements could have occurred if the leg segment of the slide mounted specimen was not perfectly horizontal. Both of these problems could be solved by taking a larger sample size (n > 62). Another explanati on is that there is a supernumerary instar; this could also be eluc idated by analyzing a gr eater sample size. The occurrence of supernumerary molts in laboratory reared colonies is not an uncommon observation (Chapman 1998). Correlation of Gall Size to Nymphal and Adult Stages of Tectococcus ovatus By comparing r2 values for multiple regressions of gall size vs. insect measurements, the length of the prothoracic tr ochanter/femur segment was determined to have the closest relationship with the gall wi dth (Table 3-1). Figur e 3-5 illustrates the close correlation betw een these two variab les; the best fit equation is y = 44.603x + 16.233 (r2 = 0.7126; df = 1, 65; p < 0.001).

PAGE 53

43 Table 3-1. r2 values for multiple regressions of separate leg measurements of T. ovatus vs. the gall size. Gall size is determined by the diameter of the base of the apical portion of the gall. Insect Measurement (Length) r2 Value Prothoracic Troch/Fem 0.7126 Mesothoracic Troch/Fem 0.7070 Metathoracic Troch/Fem 0.7114 Prothoracic Tibia 0.6718 Mesothoracic Tibia 0.6809 Metathoracic Tibia 0.7095 Figure 3-4. Two dimensional re presentation of the FASTCLUS cluster analysis. PROC CANDISC was used to generate two ca nonical variables (Can1 and Can 2) for graphing the results of the cluster analysis. The mean and standard deviation of the lengths for the prothoracic trochanter/femur segments of each cluste r were provided by the FASTCLUS cluster analysis (Table 3-2). The mean and standard deviation are provided for each of the four

PAGE 54

44 clusters because the possibility of a supernum erary instar could not be ruled out. In Table 3-2, the first and second clusters represen t the first and second instar respectively. Further sampling is necessary, but the third cl uster may represent either a supernumerary third instar or the adult stage. The fourth cl uster represents the adults with three instars or outlying data from measuremen ts of extremely large adults. y = 44.603x + 16.233 R2 = 0.7126 0 20 40 60 80 100 120 140 160 00.511.522.5Gall Size (mm)Troch/Fem Length (microns) Upper 95% CI Lower 95% CI Linear (Best Fit Line) Figure 3-5. Linear regression analysis of the relationship between the diameter of the apical base of the plant gall and the fu sed prothoracic trochanter/femur length of T. ovatus (r2 = 0.7126; df = 1, 65; p < 0.001). Table 3-2. Mean and standard deviation of the length of th e fused prothoracic trochanter/femur segment for each developmental instar. Cluster Mean Prothoracic Trochanter/ Femur Length () Standard Deviation 1 (First Instar) 35.748 5.706 2 (Second Instar) 55.315 5.118 3 (Third Instar or Adult) 101.706 6.823 4 (Adult) 116.914 4.390

PAGE 55

45 Acknowledgements The author would like to thank Greg Hodges and Yen Dao for instruction in preparing the slide mounted sp ecimens. The multivariate statistical analyses conducted for this study would not have been possibl e without the professi onal assistance of Meghan Brennan and Kenneth Portier. In a ddition, thanks are due to Alejandro Arevalo for his help with analyzing the regression data. This research was funded by the USDA CSREES Tropical/Subtropical Agriculture Re search program (T-STAR-Caribbean) grant No. 01062227.

PAGE 56

46 CHAPTER 4 HOST SPECIFICITY OF Tectococcus ovatus Introduction Strawberry guava, Psidium cattleianum Sabine, is a woody evergreen tree or shrub native to the coastal regions of southeastern Brazil. Closely related to common guava Psidium guajava L., strawberry guava was introduced to numerous countries worldwide because of its small edible fruit, attractive foliage, and broad environmental tolerances (Morton 1987). Strawberry guava was first introduced into Florida by the horticultural trade in the late 1800s (Langeland and Hall 2000) The delicate nature and short shelf life of the fresh fruit has inhibited the comm ercial potential of the plant (Schroeder and Coit 1944). There are no horticu ltural cultivars of strawber ry guava, although there are two, possibly three varieties th at are distinguished solely by the color and shape of the fruit. Two varieties are present in Florida, a yellow fruiting variety, P. cattleianum var. lucidum and a red fruiting variety, P. cattleianum var. cattleianum (Wikler 2000b). Strawberry guava is still commonly sold as an ornamental hedge and fruit tree in the state. This is despite the fact that th e plant has escaped cult ivation and is invading natural areas within the southe rn and central parts of Flor ida (Langeland and Hall 2000). Factors that aid in the spread of strawberry guava are its ability to grow in low light conditions, reproduce vegetatively, produce larg e amounts of fruit which are attractive to birds and other animals, and lack of natura l enemies (Vitorino et al. 2000, Huenneke and Vitousek 1990). Because of this, the Florida Exotic Pest Plant Council and the University of Florida, Institute of Food and Ag ricultural Sciences asse ssment of the status

PAGE 57

47 of non-native plants in Flor ida’s natural areas (IFAS A ssesment) has categorized strawberry guava as an exotic invasive species (FLEPPC 2003, IFAS 2005). In natural areas, strawberry guava can out-compete native plant species and form dense monospecific stands, which alter native plant and animal assemblages (Tunison 1991). In addition to the threat to native ecosystem s, strawberry guava also is a major host of the adventive Caribbean fruit fly, Anastrepha suspensa Loew (Nguyen et al. 1992). The Caribbean fruit fly, a native of the West Indies, is an extremely polyphagous species with almost 100 recorded hosts (Weems et al. 2005). In 1968, the Caribbean fruit fly was discovered in commercial grapefruit, which was previously thought not to be a host (Greany and Riherd 1993). To eliminate the potential spread of Caribbean fruit fly in fresh citrus shipments, the shipments were fu migated with ethylene dibromide. However, ethylene dibromide was banned by the Environm ental Protection Agency for this purpose in 1984 (Nguyen et al. 1992). This led to th e development of alternative methods to control Caribbean fruit fly populations, such as the sterile insect technique, classical biological control, and the Caribbean fruit fly-free protocol. Participation in the Caribbean fruit fly-free protocol is necessary if a grower inte nds to export fresh fruit to quarantine sensitive domestic and foreign ma rkets (FDACS 2005). The protocol involves a combination of trapping, baiting, spraying, and the establishment of a buffer zone around the designated grove. This buffer zone consists of an area free of major hosts extending 1.5 miles from the perimeter of th e designated grove; major hosts are common guava, strawberry guava, Surinam cherry ( Eugenia uniflora L.), rose apple ( Syzygium jambos L.), and loquat ( Eriobotrya japonica (Thunb.) Lindl.). Th e citrus grower is responsible for removal of major hosts from the buffer zone, and it is their responsibility

PAGE 58

48 to negotiate with property owners regardi ng removal (FDACS 2005). This can lead to disputes that can hinder citrus growers from exporting thei r product. Controlling major host plants with classical bi ological control may help so lve this problem by reducing plant populations with out active involvement from the citr us grower or property owners. The classical biological cont rol program against strawber ry guava began in Hawaii, where the plant is considered the worst inva sive weed in the archipelago (Smith 1985). In 1991, the U.S. National Park Service a nd the University of Hawaii formed a collaboration with the Federal University of Paran, Brazil to identify and evaluate potential biological control agents (Wikler et al. 2000). Five potential agents were identified and the leaf galling erioccocid, Tectococcus ovatus Hempel was determined to be the most promising based on the type of damage inflicted and the ease of handling (Wikler et al. 2000). Tectococcus ovatus is a scale insect that forms galls on the leaves, stems, and fruit of strawberry guava. Feedi ng and subsequent gall formation act as a nutrient sink depriving the pl ant of nutrients used for growth and reproduction. High infestations of T. ovatus can cause premature leaf drop, may inhibit fruit production, and may reduce photosynthesis. The damage caused by T. ovatus directly effects th e growth and sexual reproduction of strawberry guava, which may ultimately reduce fruit fly breeding sites (Vitorino et al. 2000). In addition, weakened plants will make removal easier in state parks and preserves. However, prior to the release of any biological control agent, the host range must be evaluated in order to en sure that the biocontrol agent will not harm non-target species. The purpose of this study was to determine if T. ovatus is suitably host specific in order to evaluate the potential for release in Florida.

PAGE 59

49 Materials and Methods The test plant list for this study was devel oped in accordance with the U.S. Dept. of Agriculture, Animal and Plant Health Insp ection Services Technical Advisory Group for Biological Control Agents of Weeds (TAG) guidelines. The TAG guidelines are based on the centrifugal-phylogenetic method develope d by Wapshere (1974). The test plant list is divided into 8 cate gories based on TAG guidelines and agricultural and economic concerns in Florida. These 8 categories and the division of the test plant list into the categories, as well as justification for inclus ion are available in Appendix B. Alterations made to the original test plant lis t are available in Appendix C. Specimens for establishing the T. ovatus colony used for host sp ecificity testing in Florida were shipped to the Di vision of Plant Industry quaran tine facility in Gainesville from the Hawaii Volcanoes National Park Qu arantine Facility in Hawaii (APHIS PPQ 526 permit 54024). Transport between the two st ates is more reliable and faster than receiving shipments from Brazil. In addition, the colony of T. ovatus in Hawaii was already free of predators and parasitoids. The insects reared at the Hawaii quarantine facility were obtained from an outdoor nur sery colony establis hed at the Federal University of Paran (APHIS PPQ 5 26 permits 47452, 69049) (Johnson 2005). This outdoor colony was obtained from field collected populations east of th e city of Curitiba (in the municipal districts of Piraquara, S o Jos dos Pinhais, and Colombo) (Lat 25.5167, Long -49.1667) (Johnson 2005). To transport the insects, strawber ry guava leaves containing mature T. ovatus galls were shipped in individua l containers. The leaves we re then placed on uninfested caged strawberry guava plants, to allow emergi ng crawlers to establish on new plants.

PAGE 60

50 Tectococcus ovatus colony production and host specificity experiments were conducted at the Florida Dept. of Agricultur e and Consumer Services, Division of Plant Industry, Florida Biological Control Laborat ory in Gainesville, Florida. Voucher specimens of T. ovatus were deposited in the Florida State Collection of Arthropods, Gainesville, Florida. Strawberry guava plants used to maintain the colony and as experimental controls as well as all other test plants were maintained at the University of Florida, Department of Entomology and Nematology, Ga inesville, Florida. Plants used for these experiments were either grown from seed, purchased, or collected from the field. If there was any question regarding plant taxonomy or identifi cation, a qualified botanis t was consulted. Test plants were not treated with systemic insecticides to eliminate the chance of these chemicals altering the results. All control and colony plan ts were potted with Fafard middleweight mix # 4 potting soil (Conrad Fafard Inc., Agawam, MA), test plants were potted with soil mixtures appropriate for each species. Plants that required fertilization were fertilized with Dynamite Plant Food (Florikan E.S.A. Corp., Sarasota, FL) 6 month time-release pellets with a 13:13:13 (N:P:K). No-choice host specificity tests were conducte d because this type of test is rigorous in nature (Heard 1997). Tests were replicated three times and the yellow fruiting form of strawberry guava was used as a control (Heard 1997). When testing Rhexia lutea (Melastomataceae) and Punica granatum (Punicaceae), only 2 replications were conducted due to death of plants prior to testi ng. Control plants we re set up at the same time as the test plants. Being a leaf gall former, T. ovatus requires new flush to produce a gall. Therefore, prior to te sting, test plants were pruned to induce the growth of new

PAGE 61

51 flush. To break plant dormancy, Prunus angustifolia and P. persica were maintained in a growth chamber with a 12:12 lig ht: dark photoperiod at -1.11 C for one month and then placed outside under ambient conditions. Figure 4-1. Test arena for host specificity experiments. Twenty first instar nymphs or “crawlers” were placed on the new growth of each test plant; no more than 5 insects were placed on one individual leaf Test plants ranged in height from 25 – 45 cm, and were plante d in 3.8, 7.6, or 11.4 L pots. Insects were transferred individually from a colony plant using fine for ceps. Once the insects were

PAGE 62

52 transferred onto the test plan t, an acrylic cylinder 46 cm ta ll and 15 cm in diameter was placed over the plant and the bottom was part ially buried in the soil to prevent the crawlers from escaping. The acrylic cylinders were vent ilated by six holes 6 cm in diameter. The ventilation holes and the top of the cylinder were covered with a fine mesh, with a screen size of 150 x 150 (Gr een.tek Inc., Edgerton, WI) (Fig 4-1). Once the test arenas were assembled, the plants were placed in a quarantine greenhouse. Test plants were exposed to both natural and artificial light condi tions. Supplemental fluorescent lighting in the greenhouse was se t on a 14:10 light:dark photoperiod. The average temperature inside test cylinder s was 28.88 1.61 C and the average humidity was 66.13 6.95 %. Tests were conducted for a duration of 2 weeks, after which the plants were inspected for the presence of T. ovatus or gall development. If surviving T. ovatus were found, then the tests were extended for another 2 weeks a nd subsequently reexamined. Results In total, 57 species representing 21 families were tested. Tectococcus ovatus only survived and formed viable galls on stra wberry guava (Table 4-1). However, T. ovatus survived longer than the 2 week test period on three species; strawberry guava, Brazilian guava ( Psidium friedrichsthalianum O. Berg), and Costa Rican guava ( Psidium guineense Sw.). Because of this, the Brazilian guava a nd Costa Rican guava tests were extended for an additional 2 weeks. Tectococcus ovatus fed on Brazilian guava although no gall was formed and all insects died within 4 weeks. Feeding on Costa Rican guava induced a weak gall, which was poorly formed; the gall s were cuplike in shape and did not fully cover the insect, as a normal ga ll would. In total, 5 of these galls were formed and the insects did not survive longer than 4 weeks on Costa Rican guava.

PAGE 63

53 Table 4-1. Results of T. ovatus host specificity testing. A “+” indicates feeding damage and gall development, whereas a “-“ indi cates a lack of feeding damage and gall development. Test Plant Family Results Replications Psidium cattleianum var lucidum Sabine Myrtaceae + 50 Psidium cattleianum var cattleianum Sabine Myrtaceae + 3 Psidium friedrichsthalianum O. Berg Myrtaceae -a 3 Psidium guineense Sw. Myrtaceae +b 3 Psidium guajava L. Myrtaceae 3 Acca sellowiana (O. Berg) Burret Myrtaceae 3 Eugenia axillaris (Sw.) Willd. Myrtaceae 3 Eugenia foetida Pers Myrtaceae 3 Eugenia uniflora L Myrtaceae 3 Myrciaria cauliflora (C. Martius) O. Berg Myrtaceae 3 Pimenta dioica (L.) Merr. Myrtaceae 3 Pimenta racemosa (P. Mill.) J.W. Moore Myrtaceae 3 Syzygium malaccense (L.) Merr. & Perry Myrtaceae 3 Syzygium paniculatum Gaertner Myrtaceae 3 Callistemon citrinus (Curtis) Staph Myrtaceae 3 Callistemon viminale (Gaertn.) G.Don ex Loudon Myrtaceae 3 Eucalyptus camaldulensis Dehnhardt Myrtaceae 3 Leptospermum scoparium J.R. & G. Forst. Myrtaceae 3 Melaleuca quinquenervia (Cav.) Blake Myrtaceae 3 Calyptranthes pallens Griseb Myrtaceae 3 Calyptranthes zuzygium (L.) Sw Myrtaceae 3 Eugenia confusa DC Myrtaceae 3 Eugenia rhombea Krug & Urban Myrtaceae 3 Mosiera longipes (Berg) McVaugh Myrtaceae 3 Myrcianthes fragrans (Sw.) McVaugh Myrtaceae 3 Ammannia coccinea Rottb. Lythraceae 3 Cuphea hyssopifolia Kunth Lythraceae 3 Cuphea micropetala Humb., Bonpl. & Kunth Lythraceae 3 Decodon verticillatus (L.) Ell. Lythraceae 3 Lagerstroemia indica L. Lythraceae 3 Lythrum alatum Pursh Lythraceae 3 Rhexia lutea Walt Melastomataceae 2 Rhexia mariana L. Melastomataceae 3 Rhexia nashii Small Melastomataceae 3

PAGE 64

54 Table 4-1. Continued. Test Plant Family Results Replications Tetrazygia bicolor (P. Mill.) Cogn. Melastomataceae 3 Rollinia mucosa (Jacq.) Baill. Annonaceae 3 Punica granatum L. Punicaceae 2 Conocarpus erectus L Combretaceae 3 Chrysobalanus icaco L Chrysobalanaceae 3 Nyssa sylvatica var biflora Walt Nyssaceae 3 Daphnopsis americana (P. Mill.) J.R Thymelaeaceae 3 Ilex cassine L. Aquifoliaceae 3 Ilex x attenuata Ashe Aquifoliaceae 3 Delonix regia (Bojer ex Hook) Raf. Fabaceae 3 Quercus hemisphaerica Bartr. ex Willd Fagaceae 3 Persea americana P. Mill. Lauraceae 3 Ficus aurea Nutt. Moraceae 3 Myrica cerifera (L.) Small Myricaceae 3 Saccharum officinarum L. Poaceae 3 Eriobotrya japonica (Thunb.) Lindl. Rosaceae 3 Prunus angustifolia Marsh. Rosaceae 3 Prunus persica (L.) Batsch Rosaceae 3 Pyrus x lecontei ‘Hood’ Rosaceae 3 Citrus limon (K.) Burm. F. Rutaceae 3 Citrus x paradisi Macfad Rutaceae 3 Citrus sinensis (L.) Osbeck Rutaceae 3 Taxodium distichum (L.) L.C. Cupressaceae 3 Pinus elliottii Engelm. Pinaceae 3 Podocarpus macrophyllus (Thunb.) Sweet Podocarpaceae 3 a T. ovatus survived longer than the 2 week test pe riod; test was extended to 4 weeks, but no damage or gall formation was observed. b T. ovatus survived longer than the 2 week test period; test was extended to 4 weeks, weak leaf gall formation was observed. Discussion The results of the host specificity tests show that T. ovatus is highly host specific; feeding and weak gall formation were only obs erved on two species closely related to the target weed, Costa Rican guava and Brazilian guava. These results are not surprising because T. ovatus has been reported to attack a close relative, Psidium spathulatum

PAGE 65

55 Mattos in Brazil (Vitorino et al. 2000). This species is not na tive to North America or the Caribbean and therefore is in no danger, should T. ovatus be released. Costa Rican guava is found throughout S outh and Central America (Morton 1987). There is no commercial production but the fruit can be used to make jellies or fruit drinks (Morton 1987). This guava is occasionally grown as a minor ornamental species in Florida, although it is not comm only listed by most nurseries. Tectococcus ovatus was able to feed on Costa Rican guava, and w eak gall formation was observed. This interaction may be explained by the conservativ e nature of no-choice testing because this association has never been reported to occur in the wild. However, Costa Rican guava is not commonly grown as an or namental or fruit crop within the United States, therefore damage inflicted by T. ovatus may not be of concern. Tectococcus ovatus also was observed feeding on Brazilian guava, although feeding damage was not noticeable and no gall formation was observed. Because feeding by T. ovatus did not appear to have any noticeable adverse effect on the plant, and all insects died within 4 weeks, this behavior should not pose a risk to Brazilian guava if the insect were approved for release in Florida. Most importantly, T. ovatus did not attack common guava, which is grown as a fruit crop in south Florida. Originally, the biological control of strawberry guava was thought to be impossible because of the cl ose relationship between common guava and strawberry guava (Wikler et al. 2000). Thes e results have been confirmed in Brazil, where the two species are found within the sa me range, and in host specificity tests in Hawaii (Vitorino et al. 2000; Johnson 2005).

PAGE 66

56 In a literature search of additional hosts of T. ovatus Daphnopsis racemosa Griseb. (Thymelaceae) was listed as a host in a worl dwide catalog of the family Eriococcidae (Hoy 1963). The references from this catalog were obtained, and the reference to this host plant association was traced back to a catalog of Brazilian insects (Da Costa Lima 1936). Da Costa Lima (1936) lists T. ovatus as producing galls on the leaves of strawberry guava and another plant called “embi ra”. Subsequent investigation revealed that the common name embira refers to two plants in two different families D. racemosa and Rollinia salicifolia Schltdl. (Annonaceae). These par ticular species do not occur in North America or the Caribbean and theref ore are in no danger of being attacked by T. ovatus However, there are members of the two genera present in the Caribbean, including some endangered members of the genus Daphnopsis Although this host association is most likely erroneous, one representative of each genus ( D. americana and R. mucosa ) was tested and were found not to be attacked by T. ovatus. An advantage of choosing a gall forming insect as a biological control agent is they tend to have narrow host ranges (Harris and Shorthouse, 1996). Th is is due to the complex co-evolutionary relationship that gall forming insects have with their host plants. The results of this study support this observation. The most pressing issue that will need to be addressed prior to release of this insect is the continued sale of strawberry guava as an ornamental in Florida. Despite evidence that this species is invasive; the ornamental industry is relu ctant to phase out strawberry guava because of its economic value. The FD ACS, DPI is the state agency charged with implementing and enforcing laws regarding inva sive plants. Their fo cus has historically been on agricultural threats, mo st of which are not yet presen t in the state (FDACS 2004).

PAGE 67

57 However, the IFAS assessment focuses on envi ronmentally relevant invasive species. The IFAS assessment is not a regulatory list; the primary goal of this assessment is to direct research and extension at the Univer sity of Florida. A possible solution for nurseries would be to substitute invasive spec ies with native plants that have similar desirable characteristics. A native plant s ubstitution guide for Fl orida was developed by FLEPPC (Ferriter 2003). Recommended substitu tions were based on the aesthetic values of the plants, and similarity of fruit char acteristics. Three plants recommended as substitutions for strawberry guava are Simpson’s stopper, Myrcianthes fragrans (Sw.) McVaugh, Guianese colicwood, Rapanea punctata (Lam.) Lundell, and Jamaican caper, Capparis cynophallophora L. Based on the results of this study, T. ovatus is highly host specific and would make a suitable biological control agent for the cont rol of strawberry guava in Florida. The non-target effects observed on Costa Rican guava and Brazilain guava were minimal, and the insect was unable to complete its de velopment on these guavas. However, the continuing conflict with the nur sery industry regarding the sale of guavas as ornamentals in Florida needs to be resolved prio r to the release of this organism. Acknowledgments The author would like to thank the botanis ts Mark A. Garland and Richard Weaver for their professional assistan ce with nomenclature, identifi cation, and collection of many specimens used throughout this project. Ge nerous plant donations were made by the Chicago Botanical Garden and Ornamental Plants & Trees Inc. Thanks are also due to Judy Gillmore for her assistance in procuring an d maintaining test plants. This research was funded by the USDA CSREES Tropical/Sub tropical Agriculture Research program (T-STAR-Caribbean) grant No. 01062227.

PAGE 68

58 CHAPTER 5 DISCUSSION AND CONCLUSIONS Prior to this study, little was known about the biology of T. ovatus Literature references were limited to mainly tax onomic descriptions (F erris 1957) and catalog citations (Hoy 1963). This is most likely due to the fact that T. ovatus is not an economically important species and is lim ited in its geographical and host range. Therefore, it is not surprising that prior to an outbreak of strawberry guava, the primary host plant of this species, the main scientific value of T. ovatus was taxonomic in nature. The identification of strawberry guava as a major natural areas weed in Hawaii and its association with anothe r invasive species, the fera l pig attracted attention to strawberry guava and its natural enemies (Wik ler et al. 2000). Initia l investigations for biological control were directed towards plant pathogens in hopes of developing a bioherbicide. These initial explorations we re not successful and the focus shifted to highly specific phytophagous insects (Hodge s 1988, Wikler 2000b, Wikler et al. 2000). Previous studies on the biology of T. ovatus were preliminary in nature and published in non-refereed proceedings (Wikle r et al. 2000, Vitorino et al. 2000). These two papers were the first published biological studies on T. ovatus Wikler et al. (2000) studied seven different natural enemies of strawberry guava. They identified T. ovatus as being the most promising agent for biological control. Additionall y, the paper included general descriptions of the l eaf gall, the male and female T. ovatus a note on distribution, records of two parasitoids, Metaphycus flavus Howard (Hymenopter a: Encyrtidae) and Aprostocetus sp. Westwood (Hymenoptera: Eu lophidae), and one predator, Hyperaspis

PAGE 69

59 delicate Massuti and Vitorino (Wikler et al. 2000) Wikler et al. (2000) reported that T. ovatus is found more frequently on the red frui ted variety of strawberry guava, whereas Vitorino et al. (2000) stated that the cocci d is more common on the yellow fruiting variety. Vitorino et al. (2000) provides a much more detailed account of biological observations than Wikler et al. (2000). Vito rino et al. (2000) recorded mean diameters on both sides of the leaf gall and height measur ements. The gall sizes were divided into three groups, based on the median for all of th e measurements and one standard deviation from the mean in both directions. Vitorino et al. (2000) also conducte d tests to determine the best method for transferring the first inst ar crawlers, and conducted preliminary host specificity tests. Most biological observati ons were morphological in nature with a brief description of life cycl e. However, these two papers provided the basis for further investigations into the biology of T. ovatus Investigating the developmental biology of the female T. ovatus was chosen for a couple of reasons. Female coccids are typi cally used in taxonomic descriptions of species (Ferris 1957). This is because they are more persistent and usually sessile; therefore, they are encountered more frequen tly in nature. Additionally, the life history and physiology of female T. ovatus makes them more important for biological control. Females are much larger because of their ab ility to produce progeny. Unlike males, they must continue to feed in the adult stage in order to produce eggs. According to Stehr (1991), female members of the Family Erioco ccidae typically have two instars and one adult stage. However, preliminary molecular investigations by Cook et al. (2002) suggest

PAGE 70

60 that the family is not monophyletic. Therefor e, researchers should be cautious of making any general assumptions about the family. Females of T. ovatus spend the majority of their life inside a protective plant gall. This makes investigating the development of the insect difficult without dissecting the gall and disrupting the normal growth proce ss. Therefore, an attempt was made to correlate gall size to a particular nymphal stag e or the adult, in order to make assumptions about development without destructive samp ling. However, before this could be accomplished, the number of life stages (excludi ng the egg) had to be confirmed. This turned out to be more diffic ult than anticipated. Typical methods of determining the number of instars are by constructing a frequency hist ogram of a particular body measurement (typically a head capsule meas urement) (Daly 1985). However, this was not possible for T. ovatus because it is a soft bodied insect and slide mounting can alter the shape of the soft integument, resulting in excessive variation in the measurements. Measurements needed to be made of scleroti zed structures that w ould not be altered by the slide mounting process. Th e two sclerotized portions of T. ovatus are the mouthparts and the legs; the legs were chosen because th ey are more distinct and easily identified and measured. Multivariate morphometric analyses (principal components analysis, cluster analysis, and canonical analysis) we re conducted on 15 measurements of female T. ovatus leg. The results indicated the presence of two or possibly three instars. The presence of supernumerary instars is not uncommon in laboratory colonies; however, the possible third instar also could have been produced by outlying data points. It was determined that increasing the sample size wi ll be necessary to definitively determine the correct number of developmental instars.

PAGE 71

61 A regression analysis was performed to sa tisfy the original concept of correlating instar to gall size. A significant correlat ion was found between the length of the fused prothoracic trochanter/fem ur segment and the diameter at th e base of the apical portion of the leaf gall (r2 = 0.7126; doff. = 1, 65; p < 0.001). The length of the fused prothoracic trochanter/femur can be used to correlate th e life stage (excluding e gg) with the gall size (based on the morphometric analysis). The original purpose of studying T. ovatus was to evaluate the potential of the insect as a classical biological control agent of strawberry guava in Florida. The primary concern of researchers attempti ng to evaluate classical biolog ical control agents is the safety of cultivated or socially important plants from attack by the agent (Wapshere 1979). The test plant list for this study wa s developed in accordance with the Technical Advisory Group for Biological Control Agents of Weeds guidelines. These guidelines are based on the centrifugal -phylogenetic method develope d by Wapshere (1974). The analysis of 57 species of plants representi ng 21 families resulted in only two relatively minimal non-target effects. Tectococcus ovatus fed on two closely related Psidium species, Brazilian guava and Costa Rican guava Additionally, incomplete gall formation was observed on the Costa Rican guava. Bo th of these tests were extended for 2 additional weeks, and no T. ovatus specimens survived on either plant species. Due to the limited value of these two guavas in Florid a, these non-target effects were determined to be negligible and not a concern if T. ovatus were approved for release in Florida. The most important result of the host range test was the lack of damage to common guava, which is closely related to the target weed and commercially produced on a small scale in Florida. Tectococcus ovatus did not attack common guava in this experiment and

PAGE 72

62 these results have been confirmed by observa tions in Brazil and host specificity tests in Hawaii (Vitorino et al. 2000, Johnson 2005). These studies have elucidated a portion of the biology of T. ovatus, although the continued evaluation of this insect in Flor ida is important. Fu rther research on the developmental biology of the insect is needed to establish the length or duration of each instar. Additionally, data from these studies could be used to furt her correlate the gall size of T. ovatus with instar number. This correlat ion would make it possible to follow multiple cohorts of individuals throughout thei r life cycle without disrupting the integrity of the gall of the insect. The cumulative da ta would be useful for the constructing of multiple age-specific life tables. These life tables could help researchers better anticipate peaks in natural populations of T. ovatus should the insect be rele ased in Florida. These tables also may assist in timing subsequent releases to coincide with natural population spikes, increasing the chances of a successful release. If T. ovatus is approved for released in Florida, studies also could be conducted on the distribution rates of mobile life stages. This may aid researchers in predicting the spread of T. ovatus under field conditions. The efficacy of T. ovatus as a biological control agent also could be be tter analyzed in the field. This could help researchers determine if further studies on additional biol ogical control agents for strawberry guava should be pursued. This brings up the importa nce of evaluating the effect of generalist predators and parasitio ids on populations of T. ovatus in the field. Vito rino et al. (2000) recorded a high parasitism rate of 49% within the native range of T. ovatus Understanding how a biological control agen t interacts within the new habitat is fundamental to the continued progress of weed biological cont rol as a science.

PAGE 73

63 An important aspect of this study was the aim of reducing Caribbean fruit fly populations by reducing breeding si tes in strawberry guava. If T. ovatus is approved for released in Florida, studies should be conduc ted on the subsequent effect on populations of the Caribbean fruit fly. Field studies co uld be analyzed along with state fruit fly trapping data collected before and after field releases of T. ovatus A reduction of Caribbean fruit fly populations that is corre lated with reduction of breeding areas in strawberry guava could justify the implemen tation of biological control programs against other invasive hosts such as Surinam cherry (IFAS 2005).

PAGE 74

64 APPENDIX A WORLDWIDE DISTRIBUTION OF Psidium cattleianum The following table, lists the worldwide distribution of P. cattleianum The table is divided into seven sections based on biogeogr aphic regions. The records are organized by country or island chain and locations such as cities or particular islands are provided.

PAGE 75

65COUNTRY/ ISLAND CHAIN LOCATION SOURCEa REFERENCEb STATUS AFROTROPICAL Gana HS Wikler 1999 Kenya Nairobi HS Wikler 1999 Madagascar Antananarivo HS MBG Fianarantsoa HS MBG Toamasina HS MBG Toliara HS MBG Tamatave HS Wikler 1999 Mascarenes Islands LR Weber 2003 Invasive Mauritius HS MBG, Wikler 1999 Invasive LR Wyse-Jackson 1990 Invasive Reunion HS Wikler 1999 Rwanda Butare HS Wikler 1999 Seychelles HS Wikler 1999 Invasive Mah LR Gerlach 2004 Invasive Silhouette Island LR Gerlach 2004 Invasive Sierra Leone HS Wikler 1999 South Africa Natal, Durban HS Wikler 1999 Invasive Natal, Durban LR Henderson 1989 Invasive Tanzania Amani HS Wikler 1999 Lushoto HS Wikler 1999 AUSTRALASIAN Australia Lord Howe Island HS Wikler 1999 Queensland HS Wikler 1999 Lord Howe Island HS Wikler 1999 Micronesia LR Weber 2003 Introduced Melanesia LR Weber 2003

PAGE 76

66COUNTRY/ ISLAND CHAIN LOCATION SOURCEa REFERENCEb STATUS New Caledonia St. Louis HS MBG, Wikler 1999 New Zealand LR Webb et al. 1988 Norfolk Island HS Wikler 1999 Invasive New Hebrides Pentecost Island HS Wikler 1999 Polynesia LR Weber 2003 Invasive Samoa Upolu HS NYBG Solomon Islands HS Wikler 1999 EAST PALEARCTIC China LR Bretschneider 1898 Cultivated/ Possibly Naturalized Taiwan LR Li and Huan 1979 Cultivated/ Possibly Naturalized NEARCTIC Bermuda HS Wikler 1999 Naturalized LR Britton 1918 Naturalized Mexico LR Weber 2003 United States Arizona HS ASU Herbarium California HS NYBG Cultivated/ Possibly Naturalized Florida HS MBG, NYBG, Wikler 1999 Invasive Missouri HS MBG Texas LR Jones et al. 1997 NEOTROPICAL Argentina LR Wikler 2000 Bahamas LR Morton 1987 Brazil Bahia HS NYBG, Wikler 1999 Native

PAGE 77

67COUNTRY/ ISLAND CHAIN LOCATION SOURCEa REFERENCEb STATUS Brazil Espirito Santo HS NYBG Native Parana HS NYBG Native Rio de Janeiro HS NYBG, Wikler 1999 Native Rio Grande do Sul HS NYBG, Wikler 1999 Native Santa Catarina HS MBG, NYBG, Wikler 1999 Native Sao Paulo HS MBG, NYBG, Wikler 1999 Native Chile LR Weber 2003 Colombia Antioquia HS MBG Costa Rica LR McVaugh 1963 Cartago LR Standley 1937 Cultivated/ Possibly Naturalized San Jose LR Standley 1937 Cultivated/ Possibly Naturalized Cuba LR Roig and Mesa 1953 Dominican Republic LR Wikler 2000 Galapagos Islands LR Weber 2003 Guatemala LR McVaugh 1963 Guayana LR McVaugh 1969 Honduras Francisco Morazan HS MBG Jamaica Clarendon HS MBG, NYBG Naturalized LR Fawcett and Rendle 1926 Naturalized Lesser Antilles LR Wikler 2000 Cultivated/ Possibly Naturalized Guadeloupe LR Howard 1989 Martinique LR Howard 1989 Montserrat LR Howard 1989 Nevis LR Howard 1989

PAGE 78

68COUNTRY/ ISLAND CHAIN LOCATION SOURCEa REFERENCEb STATUS Puerto Rico LR Liogier a nd Martorell 1982 Naturalized Uruguay LR Lombardo 1964 Native Venezuela LR Pittier 1926 ORIENTAL Christmas Island HS Wikler 1999 Fiji Taunovo HS Wikler 1999 Invasive Viti Levu HS Wikler 1999 Invasive Viti Levu LR Greenwood 1944, 1949 Invasive Hong Kong HS Wikler 1999 India LR Morton 1987 Japan Bonin Islands HS Wikler 1999 Malaysia Sabah HS Wikler 1999 Selangor HS Wikler 1999 Philippines LR Morton 1987 Singapore LR Morton 1987 Sri Lanka Seethaganguala HS Wikler 1999 LR Fosberg 1971 Tahiti HS NYBG, Wikler 1999 United States Hawaii HS MBG, NYBG, Wikler 1999 Invasive WESTERN PALEARCTIC Azores LR Weber 2003 Introduced British Isles LR Weber 2003 Cape Verde Islands LR Weber 2003 France, Central LR Weber 2003 Mediterranean Isl. LR Weber 2003

PAGE 79

69COUNTRY/ ISLAND CHAIN LOCATION SOURCEa REFERENCEb STATUS Madeira LR Lowe 1868 aHS denotes that the source is a herbarium specime n, while LR denotes a literature reference. bMBG refers to the Missouri Botanical Garden and NYBG refers to the New York Botanical Garden

PAGE 80

70 APPENDIX B FINAL TEST PLANT LIST FOR HO ST SPECIFICITY TESTING OF Tectococcus ovatus Category 1 Genetic types of the target weed species (varieties, races, forms, genotypes, apomicts, etc.) found in North America. Category 2 – Species present in North America in the same genus as the target weed, divided by subfamily (if applicable). Category 3 North American species in other genera in the same family as the target weed, divided by subfamily (if applicable). Category 4 Threatened and endangered species in the same family as the target weed divided by subgenus, genus, and subfamily Category 5 North American species in other families in the same order that have some phylogenetic, morphological or biochemi cal similarities to the target weed. Category 6 North American species in other orders that have some morphological, or biochemical sim ilarities to the target weed Category 7 Any plant on which the biological control agent or its close relatives (within the same genus) have been previous ly found or recorded to feed and/or reproduce Category 8 Plants not closely related to weed, which have economic significance and are grown in the same range as the weed in North America

PAGE 81

71PLANTS AUTHOR COMMON NAME REASON FOR INCLUSION CATEGORY 1 Family Myrtaceae Subfamily Myrtoideae Psidium cattleianum var. cattleianum Sabine strawberry guava target weed Psidium cattleianum var. lucidium Sabine strawberry guava target weed CATEGORY 2 Family Myrtaceae Subfamily Myrtoideae Psidium friedrichsthalianum O. Berg Costa Rican guava closey re lated to target, grown as an minor ornamental in FL Psidium guajava L. common guava closey related to target, grown as a minor fruit cropl in south FL Psidium guineense Sw. Brazilian guava closely relate d to target, possibly grown as an ornamental in FL CATEGORY 3 Family Myrtaceae Subfamily Myrtoideae Acca sellowiana (O. Berg) O. Berg feijoa, pineapple guava commonly grown as an ornamental in FL Eugenia axillaris (Sw.) Willd. white stopper native Eugenia, also cultivated as an ornamental Eugenia foetida Pers. Spanish Stopper native Eugenia, also cultivated as an ornamental Eugenia uniflora L. Surinam cherry not native but commonly sold as ornamental, preferred host of Caribfly Myrciaria cauliflora (C. Martius) O. Berg jaboticaba not native, orname ntal fruit in south FL

PAGE 82

72PLANTS AUTHOR COMMON NAME REASON FOR INCLUSION Pimenta dioica (L.) Merr allspice not nati ve, invasive in HI, minor ornamental in FL Pimenta racemosa (P. Mill) J.W. Moore bay-rum tree native in Caribbean, minor ornamental in FL, extracts used in perfumes Syzygium malaccense (L.) Merr. & Perry malay apple common ornamental in FL Syzygium paniculatum Gaertner Australian brush cherry ornamental in FL Subfamily Leptospermoidiae Callistemon citrinus (Curtis) Staph crimson bottlebrush not nati ve, but cultivated as an ornamental in FL Callistemon viminale (Gaertn.) G. Don ex Loudon weeping bottlebrush not native, but cultivated as an ornamental in FL Eucalyptus camaldulensis Dehnhardt red river gum ornamental in US Leptospermum scoparium J.R. & G. Forst. broom teatree invasive in HI, cultivated in FL Melaleuca quinquenervia (Cav.) Blake melaleuca invasive in FL, had on hand CATEGORY 4 Family Myrtaceae Subfamily Myrtoideae Calyptranthes pallens Griseb. spicewood native threatened species Calyptranthes zuzygium (L.) Sw. myrtle of the rive r native endangered species Eugenia confusa DC. redberry stopper native endangered species Eugenia rhombea Krug & Urban spiceberry eugenia native endangered species Mosiera longipes (Berg) McVaugh mangroveberry native threatened species Myrcianthes fragrans (Sw.) McVaugh Simpson's stopper native threateded species CATEGORY 5 Family Lythraceae Ammannia coccinea Rottb valley redstem native species

PAGE 83

73PLANTS AUTHOR COMMON NAME REASON FOR INCLUSION Cuphea hyssopifolia Kunth Mexican false heather introdu ced species, common ornamental Cuphea micropetala Humb., Bonpl. & Kunth tall cigar plant cultivated species Decodon verticillatus (L.) Ell. swamp loosestrif e native endangered species Lagerstroemia indica L. crapemyrtle introduced species, commercially important Lythrum alatum Pursh winged Lythrum native species Family Melastomataceae Rhexia lutea Walt. yellow meadowbeauty native species Rhexia mariana L. Maryland meadowbeauty native species Rhexia nashii Small maid Marian native species Tetrazygia bicolor (P. Mill.) Cogn flordia cover ash native threatened species Family Punicaceae Punica granatum L. Pomegranite introduced, commercially important, minor fruit crop Family Combretaceae Conocarpus erectus L. button mangrove native, economically and environmentally important CATEGORY 6 Family Chrysobalanaceae Chrysobalanus icaco L. icaco coco plum native species also sold as ornamental Family Nyssaceae Nyssa sylvatica v biflora Walt. swamp tupelo native specie s also sold as ornamental

PAGE 84

74PLANTS AUTHOR COMMON NAME REASON FOR INCLUSION CATEGORY 7 Family Annonaceae Rollinia mucosa (Jacq.) Baill wild sugar apple cl osely related to recorded host of T. ovatus Family Thymelaeaceae Daphnopsis americana (P. Mill.) J.R. burn nose closel y related to recorded host of T. ovatus CATEGORY 8 Agriculturally Important Plants Aquifoliaceae Ilex cassine L. dahoon holly native, and common ornamental, as are many Illex species Ilex x attenuata Ashe topal holly native, a nd common ornamental, as are many Illex species Fabaceae Delonix regia (Bojer ex Hook) Raf. royal poinciana ornamental tree in S Florida Fagaceae Quercus hemisphaerica Bartr. Ex Willd. darlington oak native, common hardwood tree Lauraceae Persea americana P. Mill avocado introduced crop tree, common in S Florida Moraceae Ficus aurea Nutt. Florida strangler fig native, common ornamental in S Florida Myricaceae Myrica cerifera (L.) Small wax myrtle native ornamental Poaceae Saccharum officinarum L. sugarcane introduced, common crop

PAGE 85

75PLANTS AUTHOR COMMON NAME REASON FOR INCLUSION Rosaceae Eriobotrya japonica (Thunb.) Lindl. loquat introduced, common ornamental/ fruit tree Prunus angustifolia Marsh. chicksaw plum native, ornamental Prunus persica (L.) Batsch peach introduced crop tree Pyrus x lecontei 'Hood' Rehd. Hood pear introduce d, cultivated crop tree Rutaceae Citrus limon (L.) Burm. F. lemon introduced crop tree Citrus x paradisi Macfad. grapefruit introduced crop tree Citrus sinensis (L.) Osbeck sweet orange introduced crop tree Cupressaceae Taxodium distichum (L.) L.C. cypress native, common ornamental species Pinaceae Pinus elliottii Engelm. slash pine native, common ornamental species Podocarpaceae Podocarpus macrophyllus (Thunb.) Sweet southern yew introdu ced, common ornamental species

PAGE 86

76 APPENDIX C CHANGES TO THE FINAL TEST PL ANT LIST FOR HOST SPECIFICITY TESTING OF Tectococcus ovatus

PAGE 87

77Plants Removed From the Original Test Plant List SCIENTIFIC NAME AUTHOR COMMON NAME REASON FOR INCLUSION Albizia julibrissin Durazz silktree Invasive acco rding to IFAS Assessment Calyptranthes thomasiana Griseb. Thomas. Lidflower endangered in PR, unable to obtain, tested 2 native threatened Calypthranthes species Daphnopsis helleriana Urban Heller's cieneguillo native to PR, could not obtain, tested D. americana Daphnopsis philippiana Krug & Urban emajagua de sierra nativ e to PR, could not obtain, tested D. americana Dirca palustris L. eastern leatherwood native to PR, could not obtain Eucalyptus cinera F. Muell. Ex Benth silver dollar tree introduced to Hawaii Eucalyptus grandis W. Hill ex Maid grand eucalyptus introduced in FL, not common as ornamental Eugenia aggregata (Vell.) Kiaersk aggregate eugenia not nati ve, plenty of Eugenias represented in test plant list Eugenia brasiliensis Lam Brazil cherry not native, plen ty of Eugenias represented in test plant list Eugenia haematocarpa Alain. luquillo mountain stopper endangered in PR, unable to obtain, tested many native Eugenias Eugenia koolauensis O. Deg. koolau eugenia endangere d in HI, in no danger from proposed agent Eugenia reinwardtiana Blume mountian stopper native to Hawaii, in no danger from proposed agent Eugenia woodburyana Alain. Woodbury's stopper endangered in PR, unable to obtain, tested many native Eugenias Pseudanamomis umbellulifera (Kunth) Kausel monos plum introdu ced, not common as ornamental Senna pendula (Humb. & Bonpl. ex Willd.) Irwin & Barneby valamuerto invasive in sout h Florida according to IFAS Assessment Syzygium cumini (L.) Skeels java plum invasive, tested 2 Syzygium species, none native to continental US

PAGE 88

78SCIENTIFIC NAME AUTHOR COMMON NAME REASON FOR INCLUSION Syzygium jambos (L.) Alston rose apple invasive, tested 2 Syzygium species, none native to continental US Syzygium samarangense (Blume) Merr. & Perry wax apple not native, tested 2 Syzygium species, none native to continental US

PAGE 89

79Plants Substituted on the Original Test Plant List SCIENTIFIC NAME AUTHOR COMMON NAME REASON FOR INCLUSION Ammannia robusta Heer & Regel grand redstem not pres ent in release area, decided to substitute more appropriate A. coccinea Cuphea aspera Chapman tropical waxweed highly e ndangered, collecti ng to test may harm remaining population, testing 2 surrogate species Ficus benjamina L. weeping fig not native, decided to test a native Ficus aurea Lythrum curtissii Fern. Curtiss' loosestrife not gr own as ornamental tested native ornamental species L. alatum Lythrum flagellare Shuttlw. ex. Chapman Florida loosestrife not grown as ornamental tested native ornamental species L. alatum Prunus caroliniana (P. Mill.) Ait chrry laurel substituted closeley related Prunus angustifolia because of availability Pyrus communis L. pear tested a variety cultivated to grow in FL, Pyrus x lecontei Rhexia parviflora Chapman white meadowbeauty tested 3 native surrogates R. lutea R. mariana and R. nashii Rhexia salicifolia Kral & Bostick panhandle meadowbeauty tested 3 native surrogates R. lutea R. mariana and R. nashii

PAGE 90

80 LIST OF REFERENCES Abrahamson, W.G., Weis, A.E., 1987. Nutriti onal ecology of arthr opod gall makers. In. Slansky, F., Rodriguez, T.A., (eds.), Nutrit ional Ecology of Insects, Mites, Spiders and Related Invertebrates John Wile y and Sons Inc., New York, 235-258. ASU (Arizona State University) Vasc ular Plant Herbarium, May 2005. url. http://lsvl.la.asu.edu/herbarium/. Baranowski, R., Glenn, H., Sivinski, J., 1993. Biological control of the Caribbean fruit fly, Anastrepha suspensa (Loew). Florida Entomologist. 76, 245-250. Barrett, M.F., 1956. Common Exotic Trees of Sout h Florida (dicotyledons). University of Florida Press, Gainesville, 414 pp. Blair, C.A., Blackith, R.E., Bora tynski, K.L., 1964. Variation in Coccus hesperidum L. (Homoptera: Coccidae). Proceedings of the Royal Entomological Society of London. 39, 129-134. Boratynski, K., Davies, R.G., 1971. The taxonomic value of male Coccoidea (Homoptera) with an applic ation and evaluation of some numerical techniques. Biological Journal of the Linnean Society. 3, 57-102. Borror, D.J., Triplehorn, C.A., Johnson, N. F., 1989. An Introduction to the Study of Insects, Sixth Edition. Harcourt Brace Publishers, Orlando, 875 pp. Bourgoin, T.H., Campbell, B.C., 2002. Inferring a phylogeny for Hemiptera: falling into the “autapomorphic tra p.” Denisia. 176, 67-82. Bretschneider, E., 1898. History of European Botanical Discoveries in China. Marston and Co. Ltd., London, 1167 pp. Britton, N.L., 1918. Flora of Bermud a. Scribner, New York, 585 pp. Cal-IPC (California Invasive Plant Council), September 2005. Cal-IPC invasive plant inventory. url. http://www.cal -ipc.org/pest_plant_list/ Campbell, B.C., Steffen-Campbell, J.D., Sorenson, J.T., Gill, R.J., 1995. Paraphyly of Homoptera and Auchenorrhyncha inferred from 18s rDNA nucleotide sequences. Systematic Entomology. 20, 175-194.

PAGE 91

81 Chapman, R.F., 1998. The Insects: Structure and Function. 4th ed. Cambridge University Press, Cambridge. 770 pp. Cook, L.G., Gullan, P.J., Trueman, H.E. 2002. A preliminary phylogeny of the scale insects (Hemiptera: Sternorrhyncha: Co ccoidea) based on nuclear small-subunit ribosomal DNA. Molecular Phylog enetics and Evolution. 25, 43-52. Cronquist, A., 1981. An Integrated System of Classification of Flowering Plants. Columbia University Press, New York, 1262 pp. Da Costa Lima, A., 1927. Archivos da Esco la Superior de Agricultura e Medicina Veterinaria. 8, 161. Da Costa Lima, A., 1936. Terceiro Catal ogo dos Insectos que Vivem nas Plantas do Brasil Directoria da Estatistica da Producca o, Seccao de Publicidade, Rio de Janeiro, 460 pp. Daly, H.V., 1985. Insect morphometrics. Annual Review of Entomology. 30, 415-438. Degener, O., 1932. Flora Hawaiiensis. Honolulu, 273 pp. Dehgan, B., 1998. Landscape Plants for Subtropi cal Climates. Univ. Press of Florida, Gainesville, 638 pp. Diong, C.H., 1983. Population ecology and management of the feral pig ( Sus scrofa L.) in Kipahulu Valley, Haleakala National Par k, Maui, Hawaii. University of Hawaii, Ph.D. Diss. Honolulu. 408 pp. Dreger-Jauffret, F., Shorthouse, J.D., 1992. Di versity of gall inducing insects and their galls. In. Shorthouse, J.D., Rohfritsch, O ., (eds.), Biology of Insect-Induced Galls. Oxford University Press, New York, 8-33. Dyar, H.G., 1890. The number of molts of lepidopterous larvae. Psyche. 7, 189-191. Ellshoff, Z.E., Gardner, D.E., Wikler, C., Smith, C.W., 1995. Annotat ed bibliography of the genus Psidium with emphasis on P. cattleianum (strawberry guava) and P. guajava (common guava), forest weeds in Hawai'i. Cooperative National Park Resources Studies Unit, University of Hawai'i, Dept. of Botany, Honolulu. Technical Report 95. 102 pp. EXTOXNET (Extension Toxicology Network) February 2005. Pesticide information profiles: ethylene dibromide (EDB), dibromoethane. url. http://extoxnet.orst.edu/pips/edb.htm Fawcett, W., Rendle, A.B., 1926. Flora of Jamaica. British Museum, London, 121 pp.

PAGE 92

82 FDACS (Florida Department of Agricultur al and Consumer Services), December 2004. Minutes of the noxious weeds and invasi ve plant review committee meeting. Division of Plant Industry. url. http://www.doacs.state.fl.us/p i/enpp/botany/noxweed.html. FDACS, June 2005. Caribbean fruit fly-free protocol. url. http://www.doacs.state.fl.us/onestop/plt/cfffprotocol.html Ferris, G.F., 1957. A review of the fam ily Eriococcidae (In secta: Coccoidea). Microentomology. 22, 81-89. Ferriter, A., 2003. Explore your a lter-natives: a plant substitu tion guide for south Florida. Wildland Weeds 6(4), insert. FLEPPC (Florida Exotic Pest Plant Council), November 2003. List of invasive species. url. http://www.fleppc.org/Plantlist/03list.htm. FNGLA (Florida Nursery, Growers, and Lands cape Association), May 2005. State issue: invasive plants. url. www.fngla.org/ fngla-action/doc/Invasive%20Plants.pdf Fosberg, F.R., 1941. Varieties of the strawber ry guava. Proceedings of the Biological Society of Washington. 54, 179-180. Fosberg, F.R., 1971. Psidium (Myrtaceae) in Ceylon. Ceylon Journal of Science, Biological Science. 9, 58-60. Fox, A.M., Gordon, D.R., Dusky, J.A., Tyson, L., Stocker, R.K., 2004a. IFAS assesment of the status of non-native plants in Fl orida's natural areas. Florida Cooperative Extension Service, Institute of Food a nd Agricultural Sciences, University of Florida. SS-AGR-225. 27 pp. Fox, A.M., Gordon, D.R., Dusky, J.A., Tyson, L., Stocker, R.K., 2004b. The story behind the IFAS assesment of non-native plants in Florida's natural areas. Florida Cooperative Extension Service, Institut e of Food and Agricultural Sciences, University of Florida. SS-AGR-86. 6 pp. Fox, A.M., 2005. Invasive plant lists of the southeast: a primer. Wildland Weeds. 8, 1820. Frank, J.H., McCoy, E.D., 1995. Introduction to insect behavioural ecology: the good, the bad, and the beautiful: non-i ndigenous species in Florida. Florida Entomologist. 78, 1-15. Gaines, J.C., Campbell, F.L., 1935. Dyar’s rule as related to the number of instars of the corn ear worm, Heliothis obsoleta (Fab.), collected in the field. Annals of the Entomological Society of America. 28, 445-461.

PAGE 93

83 Gardner, D.E., Davis, C.J., 1982. The prospect s for biological control of nonnative plants in Hawaiian national parks. Cooperative National Park Resources Studies Unit, University of Hawaii. Honolul u. Technical Report 45. 55 pp. Gerlach, J., 2004. A 10-year study of changes in forest vegetation on Silhouette Island, Seychelles. Journal for Na ture Conservation. 12, 149-155. Gilman, E.F., Watson, D.G., 1994. Psidium littorale, Cattley guava. Florida Cooperative Extention Service, Institute of Food a nd Agricultural Sciences: University of Florida. Fact Sheet ST-529. 3 pp. Gomes, P., 1983. Fruticultura Brasileira. 9th Edio, Livraria Nobel, So Paulo. 448 pp. Gordon, D.R., Thomas, K.P., 1997. Florida's invasion by nonindigenous plants: history, screening, and regulation. In. Simberloff, D., Smith, D.C., Brown, T.C., (eds.) Strangers in Paradise: Impact and Mana gement of Nonindigenous Species in Florida. Island Press, Washington D.C., 22-37. Greany, P.D., Riherd, C., 1993. Caribbean fru it fly status, economic importance, and control (Diptera: Te phritidae), Florida Entomologist. 76, 209-211. Greenwood, W., 1944. Supplementary notes on th e adventive and weed flora of the leeward coasts of Fiji. Journal of the Arnold Arboretum. 25, 397-405. Greenwood, W., 1949. Notes on some Fijian weeds and introduced plants. Journal of the Arnold Arboretum. 30, 75-84. Harris, P., Shorthouse, J.D., 1996. Effectiven ess of gall inducers in weed biological control. The Canadian Entomologist. 128, 1021-1055. Heard, T., 1997. Host range testing of ins ects. In. Julien, M., White, G., Biological Control of Weeds: Theory and Practical Application. ACIAR Monograph Series, Canberra, Australia, 192 pp. Hempel, A., 1900. As Coccidas Brasileiras. Revista do Museu Paulista. 4, 378-380. Hempel, A., 1901. Descriptions of Brazilian Co ccidae. Annals and Magazine of Natural History. 7, 118-119. Hempel, A., 1912. Catologos da Fauna Brazileir a: Vol. 3. As Coccidas do Brazil. Museu Paulista, So Paulo, 4v. Henderson, L., 1989. Invasive alien woody plants of Natal and northeastern Orange Free State (South Africa). Bothalia. 19, 237-262. Hodges, C.S., 1988. Preliminary exploration fo r potential biological control agents for Psidium cattleianum Cooperative National Park Resources Studies Unit, University of Hawaii, Dept. of Bo tany, Honolulu. Technical Report 66. 32 pp.

PAGE 94

84 Howard, R.A., 1989. Flora of the Lesser Anti lles: Leeward and Windward Island. Arnold Arboretum, Harvard Univer sity. Jamaica Plain, 6v. Hoy, J.M., 1963. A catalogue of the Eriococcid ae (Homoptera: Coccoidea) of the world. New Zealand Department of Scientific and Industrial Research Bulletin. 150, 1260. Huenneke, L.F., Vitousek, P.M., 1990. Seedling and clonal recruitment of the invasive tree Psidium cattleianum : implications for management of native Hawaiian forests. Biological Conservation. 53, 199-211. IFAS (University of Florida, Institute of Food and Agricultural Sciences), 2005. Conclusion from the IFAS assessment of the status of non-native plants in Florida's natural areas. Florida Coope rative Extension Service, Institute of Food and Agricultural Sciences, Univ ersity of Florida. 5 pp. ISWG (Invasive Species Working Group), D ecember 2003. Statewide invasive species strategic plan for Florida. url. http://www.dep.state.fl.us/la nds/invaspec/2ndlevpgs/ISWG.htm Johnson, M.T., 2005. Petition for field release of Tectococcus ovatus (Homoptera: Eriococcidae) for classical biologica l control of strawberry guava, Psidium cattleianum Sabine (Myrtaceae) in Hawaii. Inst itute of Pacific Islands Forestry, USDA Forest Service Pacific Southw est Research Station. Volcano, HI. Unpublished. Jones, S.D., Wipff, J.K., Montgomery, P. M. 1997. Vascular Plants of Texas: a Comprehensive Checklist Including Synony my, Bibliography, and Index. The University of Texas Press, Austin, 404 pp. Kishi, Y., 1971. Reconsideration of the method to measure the larval instars by use of the frequency distribution of head-capsule widths or lengths. The Canadian Entomologist. 103, 1011-1015. Langeland, K.A., Burks, K.C., (eds.), 1998. Identification & Biology of Non-native Plants in Florida’s Natural Areas. Univ ersity of Florida, Gainesville, 165 pp. Langeland, K., Hall, D., 2000. Exotic guavas in Florida: so delicious but-. Wildland Weeds. 3, 4-11. Langeland, K. 2002. Proper uses of FLEPPC’s lis t of invasive species. Wildland Weeds. 5. 16-17. Lepage, H.S., 1938. Catologo dos Coccideos do Brasil (Homoptera: Coccoidea). Revista Do Museu Paulista. 23, 389-389. Li, H.L., Huan, T.C. 1979. Flora of Taiwan: Vo l. 6. Vascular Plants Checklist and Index. Epoch Publishing Co., Taiwan, 6v.

PAGE 95

85 Liogier, A.H., Martorell, L.F., 1982. Flora of Puerto Rico and Adjacent Islands: a Systematic Synopsis. Editorial de la Un iversidad de Puerto Rico, San Juan. Lombardo, A., 1964. Flora Arborea y Arboresce nte del Uruguay. Museo Nacional de Historia Natural, Montevideo, Uruguay. 151 pp. Lowe, R.T., 1868. A Manual of Flora of Ma deira and the Adjacent Porto Santo and Desertas. Jon van Voorst, London. 366 pp. MacCaughey, V., 1917. The guavas of the Hawa iian islands. Bulletin of the Torrey Botanical Club. 4, 513-524. MacGillivray, A.D., 1921. The Coccidae: Tables for the Identification of the Subfamilies and Some of the More Important Genera a nd Species Together w ith Discussions of their Anatomy and Life History. Scarab Co., Urbana, 502 pp. Marlatt, R.B., 1980. Susceptibility of Psidium guajava selections to injury by Cephaleuros sp. Plant Disease, 64, 1010-1011. Marlatt, R.B., Alfieri, J.R., 1981. Hosts of a parasitic alga, Cephaleuros Kunze, in Florida. Plant Disease. 65, 520-522. MBG (Missouri Botanical Garden Herbarium), June 2005. url. http://www.mobot.org/MOBOT/Re search/herbarium.shtml McVaugh, R., 1963. Flora of Guatemala, pa rt vii. Fieldiana: Botany. 24, 283-405. McVaugh, R., 1969. Myrtaceae. In. Maguire, B. (ed.) The Botany of the Guayana Highland – Part VIII. Memoirs of the New York Botanical Garden. 18, 55-286. Merrill, E.D., Perry, L.M., 1938. The Myrta ceae of China. Journal of the Arnold Arboretum. 19, 197-199. Morton, J.F., 1987. Fruits of Warm C limates. Morton, J.F., Miami, 505 pp. Nguyen, R.U., Poucher, C., Brazzel, J.R., 1992. Seasonal occurrence of Anastrepha suspensa (Diptera: Tephritidae) in Indi an River County, Florida, 1984-1997. Journal of Economic Entomology. 85, 813-820. Noble, I.R., 1989. Attributes of invaders and th e invading process: terr estrial and vascular plants. In. Drake, J.A., Mooney, H.A., (eds.), Biological Invasions: A Global Perspective. John Wiley and Sons Ltd., Chichester, 301-313. NYBG (New York Botanical Garden Herbarium), June 2005. url. http://www.nybg.org/bsci/herb/

PAGE 96

86 OTA (Office of Technology Assessment, US Congress), 1993. Harmful non-indigenous species in the United States. OTA-F565, US Government Printing Office, Washington, DC. 397 pp. Pimentel, D., Lach, L., Zuniga, R., Mo rrison, D., 2000. Environmental and economic costs of nonindigenous species in the United States. BioScience. 50, 53-65. Pittier, H., 1926. Manual de las Plantas Usuale s de Venezuela. Litografia del Comerico, Caracas, Venezuela, 458 pp. Popenoe, W., 1920. Manual of Tropical and Subtropical Fruits. Macmillan Co., New York, 474 pp. Pratt, L.W., Santos, G.L., Stone, C.P., 1994. A test of four herbicides for use on strawberry guava ( Psidium cattleianum Sabine) in Kipaiiulu valley, Haleakala National Park. Cooperative Na tional Park Resources Studi es Unit, University of Hawaii, Dept. of Botany. Honolulu. Technical Report 90. 24 pp. Reichard, S.H., Hamilton, C.W., 1997. Predic ting invasions of woody plants introduced into North America. Conservation Biology. 11, 193-203. Rohfritsch, O., 1992. Patterns in gall development. In. Short house, J.D., Rohfritsch, O., (eds.), Biology of Insect-Induced Galls. Oxford University Press, New York, pp. 60-86. Roig y Mesa, J.T., 1953. Diccionario Botanico de Nombres Vulgares Cubanos. Tomo II. 2da. Edicion Ampliada y Corregida. Seoane, Fernandez y Cia, Havana. 1140 pp. Rung, A., Venable, G.L., Miller, D.R., Gill, R.J., September 2005. Scale families, family: Eriococcidae. url. http://www.sel.barc .usda.gov/scalekeys/scalefamilies/Index.html. Sabine, J., 1821. Account of a new Psidium in a letter to the secretary. by William Cattley, Esq. F.H.S. Transactions of the Horticultural Society of London. 4, 315317. Sakai, A.K., Allendorf, F.W., Holt, J.S ., Lodge, D.M., Molofsky, J., With, K.A., Baughman, S., Cabin, R.J., Cohen, J.E., Ells trand, N.C., McCauley, D.E., O’Neil, P., Parker, I.M., Thompson, J.N., Welle r, S.G., 2001. The population biology of invasive species. Annual Review of Ecology and Systematics. 32, 305-332. Schmidt, F.H., Campbell, R.K., Trotter, S.J ., 1977. Errors in determining instar numbers through head capsule measurements of a lepidopteran – a laboratory study and critique. Annals of the Entomol ogical Society of America. 70, 750-756. Schroeder, C.A., Coit, J.E., 1944. The cattley (commonly known as strawberry guava). California Avocado Society Yearbook. 29, 44-47.

PAGE 97

87 Schroeder, C.A., 1946. Priority of the species Psidium cattleianum Sabine. Journal of the Arnold Arboretum. 27, 314-315. Schuh, R.T., Slater, J.A., 1995. True Bugs of the World (Hemiptera: Heteroptera); Classification and Natural History. Corn ell University Press, Ithica, 336 pp. Simberloff, D., 1997. The biology of invasions. In. Simberloff, D., Smith, D.C., Brown, T.C., (eds.) Strangers in Paradise: Im pact and Management of Nonindigenous Species in Florida. Island Press, Washington D.C., 3-17. Smith, C.W., 1985. Impact of alien plants on Hawa i'i's native biota. In: Stone, C.P., Scott, J.M., (eds.), Hawai'i's Terrestrial Ec osystems Preservation and Management Cooperative National Park Resource s Studies Unit, Honolulu. pp. 180-250. Sorensen, J.T., Campbell, B.C., Gill, R.J., Steffen-Campbell, J.D., 1995. Non-monophyly of Auchenorryncha ('Homoptera' ) based upon 18s rDNA phylogeny: ecoevolutionary and cladistic implications w ithin pre-Heteropterodea Hemiptera (s.l.) and a proposal for new, monophyletic subor ders. Pan-Pacific Entomologist. 71, 3160. Standley, P.C., 1937. Flora of Costa Rica. Fi eld Museum of Natural History, Chicago, 4v. Stehr, F.W., 1991. Immature Insects. 2 vols. Kendall/ Hunt Publishing Co. Dubuque. 992 pp. Swanson, R.W., Baranowski, R.M., 1972. Host range and infestation by the Caribbean fruit fly, Anastrepha suspensa (Diptera: Tephritidae), in south Florida. Proceedings of the Florida State Horticultural Society. 85, 271-274. Tunison, T., 1991. Element st ewardship abstract for Psidium cattleianum The Nature Conservancy. url. http://theweeds.ucdavi s.edu/esadocs/documents/psidcat.pdf, last accessed January 2004. Vitorino, M.D., Pedrosa-Macedo, J.H., Smith, C.W., 2000. The biology of Tectococcus ovatus Hempel (Heteroptera: Eriococcidae) a nd its potential as a biocontrol agent of Psidium cattleianum (Myrtaceae). Proceedings of the X International Symposium on Biological C ontrol of Weeds, 4-14 July, Bozeman, Montana. 651657. Wagner, W.L., Herbst, D.R., Shomer, S.H ., 1999. Manual of the Flowering Plants of Hawai'i. Bishop Museum and Universi ty of Hawaii Press, Honolulu, 1919 pp. Wapshere, A.J., 1974. A strategy for evaluating the safety of organisms for biological weed control. Annnals of Applied Biology. 77, 201-211. Wapshere, A.J., 1979. Recent progress in the biological control of weeds. EPPO Bulletin. 9, 95-105.

PAGE 98

88 Webb, C.J., Sykes, W.R., Garnock-Jones, P.J., 1988. Flora of New Zealand. Botany Division DSIR, Christchurch, 867 pp. Weber, E., 2003. Invasive Plant Species of the World: A Reference Guide to Environmental Weeds. CABI Publishing, Cambridge, 548 pp. Weems, H.V., Heppner, J.B., Fasulo, T.R ., Nation, J.L., October 2005. Caribbean fruit fly Anastrepha suspensa Featured creatures. Univer sity of Florida IFAS and Florida Department of Agriculture and Consumer Services, Division of Plant Industry. url. http://creatures.ifas.ufl. edu/fruit/tropical/c aribbean_fruit_fly.htm Wikler, C., 1995. Aspectos bioecologicos de Eurytoma sp. causador de galha-do-ramo do araazeiro. Universidade Federal do Pa rana, M.S. Thesis, Curitiba. 63 pp. Wikler, C., 1999. Distribuicao geografica mundial de Psidium cattleianum Sabine e um cecidogeno com possibilidades de utilizacao em controle biologico. Universidade Federal do Parana. Ph.D. Dissertation Curitiba.135 pp. Wikler, C., 2000a. Gall former as a potential biological control for strawberry guava – Psidium cattleianum Proceedings of the X International Symposium on Biological Control of Weeds, 4-14 July, Bozeman, Montana, 667-671. Wikler, C., 2000b. Psidium cattleianum deliciously dangerous in Hawaii. Wildland Weeds. 3, 5-10. Wikler, C., Pedrosa-Macedo, J.H., Vitori no, M.D., Caxamb, M.G., Smith, C.W., 2000. Strawberry guava ( Psidium cattleianum ) prospects for biological control. Proceedings of the X International Symposium on Biological Control of Weeds, 414 July, Bozeman, Montana, 659-665. Wilkey, R.F., 1962. A simplified technique for clearing, staining, and permanently mounting small arthropods. Annals of th e Entomological Society of America. 55, 606. Wirth, F.F., Davis, K.J., Wilson, S.B., 2003. Ma rket economics of 14 potentially invasive ornamental plant species in Florida. De partment of Food and Resource Economics, Economic Analysis Research Report, Un iversity of Florida. EIR 03-1. 29 pp. Wunderlin, R.P., 1998. Guide to the Vascular Pl ants of Florida. University Press of Florida, Gainesville, 806 pp. Wunderlin, R.P., Hansen, B.F., December 2003. A tlas of Florida vascular plants. Institute for Systematic Botany, Univ. of South Florida, Tampa. url. http://www.plantatlas.usf.edu. Wyse-Jackson, P.S., 1990. Nesocodon mauritianus (Campanulaceae). Kew Magazine. 7, 113-117.

PAGE 99

89 BIOGRAPHICAL SKETCH Frank Wessels was born in 1980 in Indian apolis, IN, and moved to Minnesota before finally settling in Columbus, OH, where he grew up. From a young age Frank developed an interest in bi ology by exploring the inhabitant s of a creek and small woods near his house. This eventually led him to Tampa, FL, where he pursued a double major of marine science and biology at the Univ ersity of Tampa. In May of 2002, Frank received his BS degree. Inte rnships at the Department of Environmental Protection and with Dow AgroSciences in Tampa piqued his interest in entomol ogy. Frank decided to pursue his masters degree in entomology at The University of Flor ida and has spent the last two and a half year s in Gainesville, FL.


Permanent Link: http://ufdc.ufl.edu/UFE0013223/00001

Material Information

Title: Biology and Host Specificity of Tectococcus ovatus (Hemiptera: Eriococcidae), a Potential Biological Control Agent of the Invasive Strawberry Guava, Psidium cattleianum (Myrtaceae), in Florida
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013223:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013223/00001

Material Information

Title: Biology and Host Specificity of Tectococcus ovatus (Hemiptera: Eriococcidae), a Potential Biological Control Agent of the Invasive Strawberry Guava, Psidium cattleianum (Myrtaceae), in Florida
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013223:00001


This item has the following downloads:


Full Text











BIOLOGY AND HOST SPECIFICITY OF Tectococcus ovatus (HEMIPTERA:
ERIOCOCCIDAE), A POTENTIAL BIOLOGICAL CONTROL AGENT OF THE
INVASIVE STRAWBERRY GUAVA, Psidium cattleianum (MYRTACEAE), IN
FLORIDA















By

FRANCIS JAMES WESSELS IV


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Frank J. Wessels

































This document is dedicated to my parents, for their support and generosity throughout my
educational career. Without them, this work would not have been possible.















ACKNOWLEDGMENTS

I would like to thank my major professor Dr. James P. Cuda for his invaluable

guidance and help throughout my degree program. I also thank my other committee

members, Dr. Kenneth A. Langeland and Dr. William A. Overholt, for their comments

and suggestions on my research and this manuscript.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

L IS T O F T A B L E S ...................................................................... .............. ................ .. v ii

LIST OF FIGURES ............................... .................... .... ...... .. viii

ABSTRACT .............. .......................................... ix

CHAPTER

1 IN TR O D U C TIO N ......................................................................... .... .. ........

2 LITER A TU R E R EV IEW ............................................................... ...................... 7

P sidium cattleianum Sabine ...................................................................................7
T ax on om y .................................................................. ............................. . 7
N om en clatu re ....................................................... 8
S cien tific n am e ................................................................... ............... 8
C o m m o n n am e s ............. .... .................................. ................ ........ ..... .. .
M orphology ..................................... ........................... .... 10
D istrib u tio n ..................................................................................................... 12
N ativ e distribution ...................................................................... ........... 12
W orldw ide distribution ........................................... ........................... 13
Distribution in the United States .............. .........................................14
B beneficial U ses .................................................................................. 16
Invasive Properties ...................... .................. .. .. .... ......... ....... 17
C control M eth o d s ................................................................. .... ....... .. .... .19
M ech anical control ........................................................... ... .... ..... ... .. 19
C hem ical control ...................... ...... ................ .. ..... .. .......... ..20
B biological control ........................................................ .. .. ............ .... 2 0
Tectococcus ovatus H em pel ................................................ ............................. 23
Higher Classification .................. ....................... .. ........ ................ 23
Taxonom y ................................................................... .... ...... ........ 24
L ife H isto ry .................................................................................................... 2 5
G all description ................................................. ............... 25
M o rp h o lo g y ................................................................................. 2 5
B biology ..................................... ..........................2 8
N u tritio n al E co lo g y ........................................................................................ 2 8


v









R recorded H ost R ange .............................................. .............................. 30

3 LABORATORY BIOLOGY OF Tectococcus ovatus .............................................. 32

Introduction.................................... ................................ ......... 32
M materials and M methods ....................................................................... ..................34
R results and D discussion .............................. ...... .............................. .... ........ ..38
Biology .................................................... ...............38
Separation of Nymphal and Adult Stages of Tectococcus ovatus......................40
Correlation of Gall Size to Nymphal and Adult Stages of Tectococcus ovatus..42
A know ledgem ents .................................................. ........ .. ........ .... 45

4 HOST SPECIFICITY OF Tectococcus ovatus................................ ...............46

Introdu action ...................................... ................................................. 4 6
M materials and M methods ....................................................................... ..................49
R e su lts ...................................... .......................................................5 2
D isc u ssio n ............................................................................................................. 5 4
A know ledgm ents ........................ .................. .. .. ........... .... ....... 57

5 DISCU SSION AND CON CLU SION S ........................................... .....................58

APPENDIX

A WORLDWIDE DISTRIBUTION OF Psidium cattleianum....................................64

B FINAL TEST PLANT LIST FOR HOST SPECIFICITY TESTING OF Tectococcus
o va tu s ...............................................................................................7 0

C CHANGES TO THE FINAL TEST PLANT LIST FOR HOST SPECIFICITY
TESTIN G OF Tectococcus ovatus ........................................ ......... ............... 76

L IST O F R E FE R E N C E S ....................................................................... ... ...................80

B IO G R A PH IC A L SK E TCH ..................................................................... ..................89
















LIST OF TABLES


Table page

2-1. Potential biological control agents for strawberry guava. ........................................22

3-1. r2 values for multiple regressions of separate leg measurements of T ovatus vs. the
gall size. ................................................................................43

3-2. Mean and standard deviation of the length of the fused prothoracic trochanter/femur
segment for each developmental instar. ...................................... ............... 44

4-1. Results of T ovatus host specificity testing. ................................... ............... 53
















LIST OF FIGURES


Figure page

2-1. M orphology of P. cattleianum .................. ....................................... ............... 11

2-2. Native distribution of strawberry guava in southeastern Brazil ...............................12

2-3. Countries where strawberry guava has been reported..............................................14

2-4. Florida counties where vouchered specimens of strawberry guava were collected...16

2-5. Phylogeny proposed by Bourgoin and Campbell (2002) of the higher classification
of the Hemiptera based on morphological, molecular, and fossil data. ...................24

2-6. Cross sectional view of the galls of T ovatus A) Female B) Male............................26

2-7. Ferris (1957) drawings of T ovatus A) Adult female B) First instar crawler ..........27

2-8. T. ovatus adult female A) Dorsal B) Ventral............... ........................................... 27

2-9. L ife stages of T. ovatus ............................................................................... ..... .. 29

3-1. Cross section depicting variation in female leaf galls of T. ovatus. ...........................36

3-2. Life stages of fem ale T. ovatus. .............. ............ ........................ ...................... 39

3-3. T. ovatus adult m ale................................... ............. ................ 39

3-4. Two dimensional representation of the FASTCLUS cluster analysis.....................43

3-5. Linear regression analysis of the relationship between the diameter of the apical base
of the plant gall and the fused prothoracic trochanter/femur length of T. ovatus. ...44

4-1. Test arena for host specificity experiments. ..................................... ............... 51
















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

BIOLOGY AND HOST SPECIFICITY OF Tectococcus ovatus (HEMIPTERA:
ERIOCOCCIDAE), A POTENTIAL BIOLOGICAL CONTROL AGENT OF THE
INVASIVE STRAWBERRY GUAVA, Psidium cattleianum (MYRTACEAE), IN
FLORIDA

By

Francis James Wessels IV

December 2005

Chair: James P. Cuda
Major Department: Entomology and Nematology

Strawberry guava, Psidium cattleianum Sabine, is a woody tree or shrub native to

coastal southeastern Brazil. Strawberry guava was introduced into Florida in the late

1800s as an ornamental species. The plant escaped cultivation and is invading natural

areas throughout the southern half of the state. In addition to negative effects on

Florida's native ecosystems, strawberry guava also is a preferred host of the Caribbean

fruit fly, Anastrepha suspense Loew (Diptera: Tephritidae).

The Caribbean fruit fly is a common agricultural pest that affects several

important fruit crops. The Caribbean fruit fly can cause direct yield loss, and its presence

can affect shipments to quarantine sensitive markets. In Florida, various control

techniques have been used with limited success. A novel approach for reducing fruit fly

populations is classical biological control of their preferred naturalized host plants, such

as strawberry guava.









A survey of the entomofauna associated with strawberry guava identified five

potential biocontrol agents. The most promising was a leaf-galling scale insect

Tectococcus ovatus Hempel. Large infestations of T. ovatus cause premature leaf drop

and inhibit fruiting, thereby reducing fruit fly breeding sites. It is essential to understand

as much as possible about a potential biological control agent prior to release. An

analysis of a series of leg measurements was conducted in order to determine the number

of nymphal stages and discern them from the adult. Multivariate analyses of the

measurements suggest the presence of two or possibly three instars. However, a greater

sample size is necessary to determine if there truly are three nymphal stages or if the

results are due to outlying data points.

Prior to the release of any classical biological control agent, the host specificity of

the agent needs to be demonstrated. No choice tests were conducted because of the

rigorous nature of the tests. In total, 57 species of plants representing 21 families were

included in the host range tests. Tectococcus ovatus first instars fed on two closely

related guava species, Brazilian guava (Psidiumfriedrichsthalianum O. Berg), and Costa

Rican guava (Psidium guineense Sw.). Incomplete gall formation was observed on the

Costa Rican guava. However, none of the T ovatus nymphs completed their

development. These non-target effects were determined to be negligible relative to the

target, because the two species attacked were not native and are rarely cultivated in

Florida. The results of the host specificity tests suggest that T. ovatus is a suitable

candidate for biological control of strawberry guava in Florida.














CHAPTER 1
INTRODUCTION

Prior to the age of exploration, plant immigration to new habitats was primarily

facilitated by seed dispersal adaptations. Seeds could be transported by natural elements

such as wind and water or by animal movement and migration, either in the digestive

tract or externally. Although these processes were relatively commonplace, large natural

boundaries such as oceans and mountains limited the rate of long distance introductions.

With the advent of human exploration and travel, these natural boundaries were readily

crossed, and an explosion of plant introductions began. The majority of plant

introductions have been intentional-for agricultural, industrial, or ornamental purposes

(Pimentel et al. 2000). Most non-native introductions are sustained solely through

cultivation, and do not persist on their own. However, a small percentage of non-native

plants become established, or naturalized in their new habitat. Some of these naturalized

plants flourish and readily disperse, posing a threat to native ecosystems. This is because

the most important direct effect of non-native species is habitat modification (Simberloff

1997). Non-native species that alter native habitats through direct competition or

hybridization are termed exotic invasive species (Florida Exotic Pest Plant Council

[FLEPPC] 2003). Although the percentage of exotic invasive species is relatively low

compared to the number of introduced species, their negative impacts are widespread and

can be extremely costly (Gordon and Thomas 1997).

Florida's climate and geography make the state prone to non-indigenous species

invasions (Invasive Species Working Group [ISWG] 2003). The high number of









invasive species proliferating in the state is understandable considering that 85% of all

foreign plants imported into the United States arrive through Miami International Airport

(US Congress, Office of Technology Assessment [OTA] 1993). Because of this,

approximately 31% of Florida's flora is of non-native origin (Wunderlin 1998). Of the

25,000 introduced species cultivated in the state, 925 have escaped cultivation and

become naturalized in Florida's native ecosystems (Frank and McCoy 1995). A smaller

number of these plants have expanded their range and are considered to be exotic

invasive species in Florida. The exact number of these species varies depending on

which source is cited. As previously mentioned, the majority of plant introductions are

intentional, and Florida's large ornamental nursery industry may be responsible for the

introduction of a large number of exotic invasive species. Not surprisingly, there has

been a conflict of interest between environmental and commercial groups regarding

which species are truly invasive, and which are not.

One organization that has been formed to inform resource managers of potential

problems concerning exotic invasive plant species within the state is the Florida Exotic

Pest Plant Council (Langeland 2002). To help resource managers make informed

decisions about which plants to monitor and help set priorities for management, FLEPPC

has compiled a list of 126 naturalized plant species to be which they consider to be exotic

invasive species (FLEPPC 2003, Langeland 2002). The FLEPPC invasive species list is

divided into two categories, Category I represents the most harmful species which have

been shown to be altering native plant communities; Category II represents species that

are increasing in abundance but are not affecting native communities to the extent of

Category I species. Although the FLEPPC list provides detailed and readily accessible









information; it has been criticized for being too liberal in its classification. This is due to

the large discrepancy between the FLEPPC list and state and federal regulations (24 of

the 67 Category I species and 6 out of 59 Category II species are regulated by the

government) and the lack of supporting evidence and references (Fox et al. 2004a).

Despite this controversy, the Florida Nursery, Growers, and Landscape Association

(FNGLA) and FLEPPC asked FNGLA members to stop selling 45 potentially invasive

species (Wirth et al. 2003). Although this was a good start, the organizations could not

agree upon the status of 14 Category I exotic invasive species that are commonly sold as

ornamentals. These 14 species represent only 2.8% of total nursery sales in Florida.

However, considering the size of the ornamental nursery industry in the state, these

species have a total combined economic impact of $59 million (Wirth et al. 2003). The

FNGLA (2005) called for the creation of a single invasive plant list that distinguishes

between Florida's geographic regions and intended plant use, both of which are not taken

into account by the FLEPPC list. A new list was created by the University of Florida,

Institute of Food and Agricultural Sciences (IFAS) called the IFAS assessment of the

status of non-native plants in Florida's natural areas (Fox et al. 2004b). The primary

objective of this assessment is to direct research and extension at the University of

Florida to be focused in the proper direction (Fox et al. 2004b). Although, secondary

objectives are to provide additional information to that available on other state invasive

lists and identify gaps in knowledge of invasive species (Fox et al. 2004b). This

assessment standardizes the classification of the status of selected plants in Florida based

upon geographic location (north, central, or south Florida), economic importance, and

effect on the environment. The conclusions of the IFAS assessment confirmed the









invasive status of all 14 species designated invasive by FLEPPC (IFAS 2005). Although

the IFAS assessment is not a regulatory list (Fox 2005), this list is considered to be the

most complete and appropriate for invasive plant species in the state of Florida.

Therefore, the IFAS assessment will be used for the purposes of defining the invasive

status of a plant in Florida throughout this manuscript.

Although the environmental and economic threat exotic invasive plants pose to

native ecosystems is great, invasive plants may also have indirect effects on their new

habitats. One example is strawberry guava, Psidium cattleianum Sabine. This woody

invasive species was brought to Florida from southeastern Brazil for the ornamental and

fruit trade in the late 1800s (Gordon and Thomas 1997). Strawberry guava is one of the

14 controversial species that is still commonly sold as an ornamental throughout the state.

In addition to the environmental problems caused by strawberry guava invading natural

areas, the plant also serves as a major host of the Caribbean fruit fly, Anastrepha

suspense Loew (Swanson and Baranowski 1972). Swanson and Baranowski (1972)

determined the host specificity ofA. suspense based on specimens reared from 84 field

collected fruits in 23 families. They identified six major hosts of A. suspense: loquat,

Eriobotryajaponica (Thunb.) Lindl.; Surinam cherry, Eugenia uniflora L.; rose apple,

Syzygiumjambos (L.) Alst.; tropical almond, Terminalia catappa L.; common guava,

Psidium guajava L.; and strawberry guava. Because of this study, Nguyen et al. (1992)

counted the number of A. suspense larvae present on the fruits of three major hosts;

loquat, Surinam cherry, and strawberry guava. They found that the Caribbean fruit fly

occurred on strawberry guava in larger numbers from July through October compared to

loquat and Surinam cherry.









The Caribbean fruit fly is native to the West Indies; although it immigrated to

Florida multiple times, finally becoming established in 1965 (Nguyen et al. 1992). The

larval stage of the fly feeds on a wide variety of tropical and subtropical fruits. The

Caribbean fruit fly is extremely polyphagous, with nearly 100 recorded hosts (Weems et

al. 2005). Fruit fly damage resulting from larval feeding renders fruit unmarketable.

In addition to direct impacts, the Caribbean fruit fly has the potential to infest post

harvest commercial fresh citrus shipments destined for quarantine sensitive domestic or

foreign markets such as California, Texas, and Japan. For this reason, the Caribbean fruit

fly is considered a major quarantine pest, and over the years multiple control strategies

have been developed to reduce fly densities in Florida. Prior to 1983, shipments were

fumigated with ethylene dibromide to eliminate the presence of fruit fly larvae (Nguyen

et al. 1992). In 1983, the use of ethylene dibromide for this purpose was banned by the

Environmental Protection Agency (Extension Toxicology Network [EXTOXNET] 2005).

With the banning of this chemical, the Florida Department of Agriculture and Consumer

Services (FDACS) developed the Caribbean fruit fly free protocol to certify citrus crops

as fly free. This protocol is an integrated approach involving the use of monitoring traps,

aerial pesticide sprays, and the removal of major hosts within 1.5 miles of designated

groves (FDACS 2005). Removal of major host species such as strawberry guava costs

citrus growers time and money. It is the grower's responsibility to remove all major

hosts, even if they are not present on the grower's property. Because of this, developing

alternative methods of controlling these weeds has become a priority, in order to help

growers save money and prevent legal disputes over removing pests on neighboring

properties.









Classical biological control of the major hosts would be preferential. Once

released, classical biological control agents do not require any physical input aside from

monitoring and they do not require landowner permission. In 1991, researchers with the

U.S. National Parks Service and the University of Hawaii collaborated with the Federal

University of Parana in Curitiba, Brazil to investigate potential biological control agents

for strawberry guava (Wikler et al. 2000). This project was undertaken because

strawberry guava is a serious forest pest throughout the Hawaiian archipelago. The group

discovered seven potential biocontrol agents. After preliminary testing, the most

promising of these agents was determined to be Tectococcus ovatus Hempel, a leaf

galling scale insect in the family Eriococcidae (Wikler et al. 2000). A colony of T

ovatus was established at the Institute of Pacific Islands Forestry in Volcano, Hawaii. At

this laboratory, host specificity tests began to evaluate T. ovatus for release in Hawaii. In

2001, researchers at the University of Florida obtained funding from the USDA CSREES

T-STAR program to investigate the biology and host specificity of T. ovatus in Florida.

This manuscript is the result of this research. The data presented in the following

chapters are the results of initial investigations of the biology of T ovatus and the host

specificity testing of this insect.














CHAPTER 2
LITERATURE REVIEW

Psidium cattleianum Sabine

Taxonomy

The Myrtaceae is a large family including approximately 150 genera and 3600

species commonly found in tropical and subtropical climates worldwide, although some

species are also established in temperate Australia (Cronquist 1981). The genus Psidium

is one of the larger genera in the Myrtaceae containing approximately 100 species

(Cronquist 1981). The genus was first described by Linnaeus, derived from the Greek

word sidion, meaning pomegranate (Punica granatum L.) due to the similar shape of the

fruits (Ellshoff et al. 1995). The complete classification of strawberry guava is as

follows:

Class: Magnoliophyta

Subclass: Rosidae

Order: Myrtales

Family: Myrtaceae

Subfamily: Myrtoidea

Genus: Psidium

Species: Psidium cattleianum Sabine









Nomenclature

Scientific name

Psidium cattleianum was first described by Sabine (1821); it was named in honor

of the botanist William Cattley (often misspelled cattleyanum in older literature). Cattley

was the first person to successfully cultivate the plant in his conservatory in Britain,

claiming that his plants were grown from seeds he received from China (Bretschneider

1898). However, some botanists consider the correct name to be Psidium littorale giving

priority to the description by Raddi in 1823. This conclusion was reinforced by Fosberg

(1941) and Merrill and Perry (1938). The confusion arises from the difference in the date

of publication and the date in the section of Raddi's description of strawberry guava

(Ellshoff et al. 1995). Schroeder (1946) noted that because of the confusion surrounding

the actual date of publication, Raddi's description cannot be proven to be earlier than

Sabine's. Since Sabine's description can be dated definitely, the currently accepted name

should remain Psidium cattleianum while Psidium littorale is considered a junior

synonym (Ellshoff et al. 1995, Schroeder 1946). Synonyms include (from Fosberg 1941,

Wikler 1999):

Psidium littorale Raddi, Opusc. Sci. 4: 254. 1823 t. 7 f 2.

Psidium variabile Berg, Fl. Bras. 14(1): 400. 1857.

Psidium coriaceum var. obovatum Berg, 1. c. 461 t. VI. 120.

Psidium coriaceum var. grandifolium Berg, 1. c. 401.

Psidium coriaceum var. longipes Berg, 1. c. 402.

Psidium cattleianum var. coriaceum (Berg) Kiaerskou, Enum. Myrt. Bras. 28. 1893.

Episyzygium oahuense Seuss. & A. Ludwig

Psidium cattleianum var. cattleianum f lucidum Degener









Psidium cattleianum var. littorale (Raddi) Fosb.

Psidium littorale var. lucidum (Degener) Fosb.

There are two accepted varieties of strawberry guava, distinguished solely by the

color of the fruit, the red-fruited variety P. cattleianum var. cattleianum and the yellow-

fruited variety P. cattleianum var. lucidum (P. littorale var. lucidum) (Wikler 2000b).

Both varieties are found within the native range of southeastern Brazil. The yellow

variety is much more common, while the red is restricted to higher elevations (Hodges

1988). A similar situation occurs in Hawaii, with the yellow variety dominating lower

elevations in Hawaii Volcanoes National Park and the red variety dominating higher

elevations (Tunison 1991). In Florida, both varieties are present, despite the lack of a

major elevation gradient (Langeland and Hall 2000).

In Hawaii, there are two different shapes of the fruit of the yellow variety of

strawberry guava. One type is the typical round shape, while the other has ellipsoid-

obconical fruit (Wagner et al. 1999). Wagner et al. (1999) consider these to be two

separate varieties (P. cattelianum var. lucidum and P. c. var. littorale). The ellipsoid-

obconical variety also is seen on Mauritius, although it has not been reported from Brazil

or Florida.

Common names

Strawberry guava has been widely introduced throughout the world, and for this

reason, there are a considerable number of common names for the species. In the US and

other English speaking countries, P. cattleianum is known as strawberry guava

(presumably due to the strawberry-like flavor and possibly the red color ofP. c. var.

cattleianum), purple guava in Jamaica, Chinese guava (due to Cattley's assumption of

Chinese origin), Cattley guava, and pineapple guava. In Brazil, the plant is known as









aracg da praia, aracazeiro coroa, aracg vermelho, and aracazeiro. Additional names are

araza (Uruguay), cas dulce (Costa Rica), guayaba japonesa (Guatemala), and guayaga

peruana (Venezuela) (Gomes 1983, Morton 1987, Popenoe 1920). Other names include

goiave de L'Afrique (Dominican Republic), aracg-saiyu, and guayabo amarillo

(Argentina), Calcutta guava (India), goyavier of St. Martin, goyavier fraise (Guadeloupe),

guayabita fresa (Cuba), and goyavier prune (Martinique) (Roig y Mesa 1953, Wikler

2000b). In Hawaii, the red variety is called waiawi ulaula, while the yellow variety is

simply called waiawi (Morton 1987).

Morphology

Strawberry guava is an evergreen shrub or small tree between 2 and 6 m tall,

although specimens of P. cattleianum var. lucidum have been reported growing up to 12

m (Morton 1987). Stems and branches are smooth, gray to reddish brown in color with

bark that peels in thin sheets (Fig. 2-1A) (Langeland and Burks 1998). Young branches

are round and sparsely pubescent. Leaves are alternate, obovate to elliptic-obovate

between 3.5 and 13.5 cm long, petioles approximately 7-10 mm long (Wagner et al.

1999, Webb et al. 1988). Leaf surface is dark green glabrous and somewhat leathery in

texture; lateral veins are slightly elevated but inconspicuous (Fig. 2-1B). Flowers are

white, 1.5-6 cm in diameter with prominent stamens (Morton 1987); flowers are usually

solitary and borne in almost all axils of upper leaves (Fig. 2-1C) (Langeland and Burks

1998, Webb et al. 1988). Fruits are sweet tasting, slightly acidic, round or elliptical,

smooth and glabrous 2-3 cm in diameter. The fruits ofP. cattleianum var. lucidum are

yellow to white, while those of P. cattleianum var. cattleianum are reddish to purple in

color (Fig. 2-1D). The red fruits also are reported to taste better, with a more subdued









flavor (Dehgan 1998). Seeds are hard, flattened and triangular in shape approximately

2.5 mm long (Morton 1987).

Strawberry guava is a relatively hardy species that can grow in a wide variety of

soils, although they perform best in rich sandy soils (Popenoe 1920). The red form seems

to better withstand colder temperatures and can survive at temperatures as low as 220 F,

whereas the yellow form is more susceptible to cold (Morton 1987). Both forms are

drought resistant, although the yellow form also can withstand short periods of flooding

(Morton 1987).


SA B




I i




















Figure 2-1. Morphology of P. cattleianum; A) characteristic bark; B) obovate leaves; C)
flowers (Photo credit: Jeff Hutchinson); D) yellow fruits of P. cattleianum var.
lucidum.









Distribution

Native distribution

In Brazil, where the Myrtaceae is a widespread and diverse family, the genus

Psidium is represented by nine species (Wikler 2000a). Strawberry guava is located in

the Atlantic forest ecosystem in the southeastern part of the country (Fig. 2-2). Within its

native range, the plant is primarily a coastal species. Strawberry guava occurs in a

coastal vegetation type known as restingga, although it can also be found in disturbed

brush fields known as "capoeiras" (Hodges 1988). The northernmost range of strawberry

guava in Brazil is in the state of Espirito Santo and extends south along the coast to the

northern tip of Uruguay (Fig. 2-2) (Hodges 1988). Hodges (1988) noted the plant

growing within an elevation range of 5 100 m; although except for cultivated plants, he

did not notice the red fruited form growing wild.






BRAZIL








Legend
Brasil
Uruguay t
M Approximate
Distribution of N
P. cattleianumn
URUGUAY


Figure 2-2. Native distribution of strawberry guava in southeastern Brazil (adapted from
Hodges 1988, Wikler 1995).









Worldwide distribution

Strawberry guava has been intentionally introduced in nearly all of the countries in

which it is currently found. Its attractive fruit and leaves are generally desired more as an

ornamental than a fruit crop. According to Popenoe (1920), strawberry guava was

originally transported from its native range in Brazil to China at an "early period",

presumably by the Portuguese. Seeds were taken to Europe in 1818 by two Englishmen,

Barr and Brookes, and described as originating from China (Bretschneider 1898). The

only reference hinting to European introduction of this plant in Hawaii is from Degener

(1932) mentioning that live plants of both forms of strawberry guava may have been

brought to the archipelago on board the "Blonde" in 1825. Similar introductions may

initially have been made by European or Portuguese ships resulting in the wide

distribution of the plant.

Distribution information obtained from literature records and herbarium specimens

have been compiled (Appendix A) in order to create a worldwide map of the distribution

of strawberry guava (Fig. 2-3). However, the distribution of strawberry guava is most

likely more extensive than mentioned in the literature. This plant is found in all 7 major

world biogeographical regions; in the Afrotropical region it is present in multiple coastal

countries of tropical Africa, and on Madagascar and surrounding islands. In the

Australasian region, strawberry guava is found in Australia and New Zealand and

surrounding island archipelagos. In the East Palearctic region it is found in southern

China and Taiwan. Recorded distribution is most common in the Neotropical region

where the plant is found throughout its native range of Brazil, surrounding South and

Central American countries and throughout the Caribbean. In the Oriental region,

strawberry guava has been reported on multiple islands in the Indian Ocean and the









Pacific and in Japan, Malaysia, and the Philippines. In the Western Palearctic region it is

found on the islands of Madeira, the Azores, the British Isles, the Cape Verde Islands,

various Mediterranean islands and, in central France. In the Nearctic region strawberry

guava is found in the southern portion of the continental United States and on Bermuda.



Worldwide Distribution of Psidium cattleianum



_h; -- The-- 6 .11
., l,,, ,L, -' ..4. U,'I',

riy '


X V n "o TT. Dilnti. T i .Rq iiq blic of Q- h d'u tiu
*iSI ,snoa .h- ,


stew Zern d





Legend o m 60 O awo So00 N
M Countries where Psidium cattleianum is Present W + E
s

Map Shape File: http. /www.cipotato.org/dia/data/MoreData.htr Map Created by: Frank Wessels 2005

Figure 2-3. Countries where strawberry guava has been reported (based on literature
references and herbarium specimens, see Appendix A).

Distribution in the United States

In 1884, approximately 3000 trees were planted as part of a commercial venture in

La Mesa, California, and the trees were reported as producing heavily a half century later

(Morton 1987). Additionally, a herbarium specimen was collected in 1995 from a

cultivated specimen at the Quail Botanical Gardens in San Diego County, California

(New York Botanical Garden [NYBG] 2005). Although strawberry guava was









introduced into southern California, no records suggest that the species is considered

invasive in the region. The California Invasive Plant Council (Cal-IPC) does not list the

species as being invasive in any part of the state (Cal-IPC 2000).

Although there are a few references confirming the presence of strawberry guava

in Texas, many books on the flora of Texas do not even include references to the family

Myrtaceae. Strawberry guava is listed in a state checklist of vascular plants; however,

there is no mention of locality or whether the specimen is cultivated or growing wild

(Jones et al. 1997). Strawberry guava can grow in US Department of Agriculture plant

zones 9-10, which suggests that the plant is capable of surviving in the southern parts of

Texas and along the Gulf Coast.

Oddly, one herbarium specimen shows the presence of strawberry guava in Boone

County, Missouri in 1974. This record is curious, because it is well out of the known

geographic range of the species in the United States. The Boone county specimen was

collected by D. B. Dunn at approximately at 39.02.00N, -92.20.00W at an elevation of

740 ft. (Missouri Botanical Garden [MBG] 2005). However, the collection data of the

specimen are rather vague regarding whether or not this specimen was cultivated.

The presence of strawberry guava in Hawaii and its subsequent effects on the

native ecosystems of the islands has been well documented in the literature. The weed is

considered the worst invasive plant in Hawaii and is present on almost all of the major

islands (Smith 1985). The plant forms dense stands in disturbed roadsides and in intact

forests where it effectively out-competes native species. Although the fruits are attractive

to birds and mammals, the feral pig (Sus scrofa L.) has been shown to be a major

dispersal agent of the plant in Hawaii (Diong 1982).









In Florida, strawberry guava was first listed as an ornamental in the 1887-1888

Catalog and Price List for Royal Palm Nurseries (Langeland and Hall 2000). By 1956, it

had escaped cultivation and was reported growing wild (Barrett 1956). Strawberry guava

is currently naturalized in the central and southern parts of the state. Voucher specimens

indicate the presence of the plant in 18 counties (Fig. 2-4) (Wunderlin and Hansen 2004).

However, the actual distribution is most likely more extensive, and needs to be confirmed

with additional voucher specimens.












Legend
SNot Present
-I | Present



Figure 2-4. Florida counties where vouchered specimens of strawberry guava were
collected (Wunderlin and Hansen 2004).

Beneficial Uses

The fruit of strawberry guava is said to be superior in taste to that of the common

guava, Psidium guajava L., and is preferred by many (Popenoe 1920). The fruits can be

eaten fresh, but are often contaminated by various fruit flies in Florida and Hawaii. Most

often the fruit is made into jelly, jam, butter, paste, punch, and sherbet (Morton 1987).

Although, the species is cultivated in many parts of the world, no distinct cultivars exist.

In Florida, strawberry guava is primarily grown as an ornamental or in hedges around









parking lots and other areas (Gilman and Watson 1994). In Brazil, the wood is

sometimes collected for firewood (Hodges 1988, Wagner et al. 1999).

There is little promise of commercial fruit production due to the delicate skin of the

ripe fruit, which would require careful shipping (Schroeder and Coit 1944). In addition,

Schroeder and Coit (1944) note that the shelf life of the fresh fruit is too short for a

commercial production, lasting only 3-4 days without refrigeration. The largest recorded

commercial planting was in La Mesa, California. A local farmer planted approximately

3000 trees on a 5 acre plot. In 1943, production was approximately 30 tons. The

majority of the fruit was processed into paste for sale; however, some fresh fruit was

marketed locally (Schroeder and Coit 1944). Despite the attempts at commercial

production, strawberry guava seems to have a more marketable potential as an

ornamental hedge or dooryard tree.

Invasive Properties

Plants that are invasive have a variety of characteristics that aid in their dominance

of new habitats. Many authors have tried to generalize these characteristics in order to

help predict which species may be potential problems (Noble 1989, Reichard and

Hamilton 1997). Although there are characteristics that many invasive plants share,

generally these are poor predictors of whether or not a species will be invasive.

Determining the invasiveness of a species based on biological attributes is probably too

simplistic. Sakai et al. (2001) argues that the invasiveness of a plant depends on a

combination of ecological, genetic, and evolutionary factors. A combination of factors

most likely aids strawberry guava in its ability to out compete native plants in areas

where it is introduced.









Strawberry guava has a high tolerance to environmental heterogeneity, it is

tolerant of moderate to highly acidic soils, and can tolerate the heavy litter fall present in

Hawaiian rainforests (Huenneke and Vitousek 1990). The plant also can grow in a wide

variety of soils including: rocky soil, clay, sandy loam, and wet forest soils (Diong 1983).

Strawberry guava is more cold tolerant than the common guava, and has been reported

withstanding temperatures as low as 22 degrees (Schroeder and Coit 1944). Strawberry

guava also is highly shade tolerant, a trait which can help seedlings grow up through the

understory and overtake native vegetation (Tunison 1991).

Another trait aiding to the invasiveness of strawberry guava is its ability to

reproduce both sexually and asexually. Huenneke and Vitousek (1990) found that

asexual plants produced by clonal root suckering had larger leaves and produced a greater

total leaf area than seedlings. This clonal behavior most likely contributes to the

formation of dense monotypic stands of strawberry guava which are common in invaded

habitats. These dense stands exclude native vegetation and may significantly alter native

plant and animal communities. Strawberry guava is a prolific fruiter with a high seed

count. Germination rates also are very high; Huenneke and Vitousek (1990) recorded

laboratory germination rates between 60-80% and field germination rates of

approximately 56%.

The fruits of strawberry guava are attractive to birds and other frugiverous

animals, which aid in the dispersal of the plant. In Hawaii, an interesting mutualistic

relationship between two invasive species occurs between strawberry guava and the feral

pig, Sus scrofa L. (Diong 1983). Diong (1983) found that strawberry guava was a

preferred food of the feral pig; during the fruiting season, strawberry guava represented









approximately 78.3% of the pig's stomach contents. Strawberry guava benefits from the

relationship by being carried to new habitats and deposited in nutrient rich pig droppings.

Frequently, the pig droppings were deposited on soil disturbed by the rooting activity of

the feral pig. Although seed viability was not affected by passage through the pig's gut,

gut treatment did result in earlier germination time compared to untreated seeds (Diong

1983).

In Hawaii, introduced agricultural livestock, such as goats, sheep, and cattle, also

have been reported feeding on strawberry guava (MacCaughey 1917). In addition, a

variety of birds have been reported as feeding on the fruits of strawberry guava, e.g. the

laced-necked dove (Streptopelia chinensis (Scopoli), mynah bird (Act i1,lithe1i e tristis L.),

rice bird (Lonchurapunctulata L.), house sparrow (Passer domesticus L.), the melodius

laughing thrush (Garrulax canorus Hwamei (Cheng), Japanese white eye (Zosterops

japonicus Temminck and Schlegel), and the red-billed leiothrix (Leiothrix lutea Maier &

Bowmaker), in addition to many others (Diong 1983, MacCaughey 1917). Other animals

reported to consume strawberry guava fruits include mongooses, bats (Diong 1983), and

squirrels (B. Overholt, University of Florida, IRREC pers. comm.).

Control Methods

Mechanical control

It is important for public land managers in charge of parks and preserves to remove

invasive species from their properties. In many cases, the most cost effective method of

removing small infestations is mechanical control (Tunison 1991). Young plants and

saplings originating from seed can be uprooted either by hand or with a weed wrench

(Tunison 1991). The uprooted plants must be disposed of, because they may re-root if

left on the ground. In extreme cases, mechanical control has been used to control large









infestations. In Florida, at Jonathan Dickinson State Park, the strawberry guava

infestation was so dense that it was necessary to bulldoze and restore a three acre portion

of the park (Langeland and Hall 2000).

Chemical control

The most common method of controlling woody weed species is by using chemical

herbicides, or a combination of mechanical and chemical control (cut stump, basal frill

treatments). Strawberry guava is sensitive to picloram (Tordon, Grazon, Dow

AgroSciences), dicamba (Banvel, Velsicol Chemical Corp.), glyphosate (Roundup,

Monsanto Co.; Accord, Rodeo, Dow AgroSciences), and triclopyr (Garlon, Dow

AgroSciences) (Tunison 1991). Pratt et al. (1994) tested the efficacy of 4 herbicides for

managing strawberry guava in Haleakala National Park, Hawaii. They found that soil

treatments with Velpar (hexazinone, 90% active ingredient, DuPont) and Spike 20P

(tebuthiuron, 20% active ingredient, Elanco now Dow AgroSciences) were ineffective.

The two most effective herbicides were 2 formulations of triclopyr produced by Dow

AgroSciences, Garlon 3A (0.36 kg/1) and Garlon 4 (0.48 kg/1) applied with the cut stump

method. These chemical control measures work reasonably well for managing

strawberry guava infestations within the maintained borders of parks and preserves.

However, chemical control methods are often non-selective, can be expensive, because

re-treatment is often necessary, and are not a long-term solution to the problem.

Biological control

A feasible long term control strategy for strawberry guava is biological control

(Smith 1985). Biological control is defined as the use of natural enemies to reduce the

numbers of pest organisms. Biological control is not an eradication technique but rather

a management strategy. There are many types of biological control, ranging from









augmentative releases of natural enemies to the traditional approach of classical

biological control. Classical biological control refers to the introduction of specialist

natural enemies from the native range of an adventive pest, with the intention of reducing

the pest population density. For the remainder of this manuscript, unless otherwise

mentioned, the terms biological control and biocontrol will be considered synonymous

with classical biological control.

The tropical climate and geographically isolated nature of the Hawaiian archipelago

has made it an ideal habitat for introduced exotic plant species to proliferate. In the early

1980s, the National Park Service and the University of Hawaii undertook a project to

prioritize the potential of invasive plant species for biological control (Gardner and Davis

1982). Initially, it was thought that strawberry guava was a poor candidate due to its

close relationship to the common guava, for which a small commercial market exists in

Florida and Hawaii (Gardner and Davis 1982). In addition, the family Myrtaceae

contains various genera which are commercially or ecologically important in both states;

for example Eucalyptus, Eugenia, Pimenta, and Syzygium. Prior to 1988, the only

mention of predators or potential pathogens for strawberry guava was a parasitic alga of

the genus Cephaleuros, which also attacks common guava (Marlatt 1980, Marlatt and

Alfieri 1981). In the mid 1980s, the National Park Service and the University of Hawaii

funded initial explorations for potential biological control agents of strawberry guava in

Brazil (Hodges 1988). The objective of this study was to determine the native

distribution of strawberry guava, to make local contacts for future collaboration, and to

determine kinds and relative impacts of predators and pathogens on strawberry guava.

No promising pathogens were discovered, although a wide variety of insects were found









to attack strawberry guava (Hodges 1988). Due to the wide distribution of the species in

Brazil, and the numerous insects found associated with the plant, it was recommended

that local entomologists be contracted to perform the initial survey of biological control

agents. As a result, the National Park Service and the University of Hawaii formed a

collaboration with the Parana Forestry Foundation and the Federal University of Parana,

Brazil. The Brazilian entomologists undertook the initial exploration for insects that

attacked strawberry guava within its native range.

Table 2-1. Potential biological control agents for strawberry guava (Wikler et al. 2000).
AGENT TAXONOMY CONTROL POTENTIAL
Dasineura gigantea Angelo (Diptera: Cecidomyiidae) Good- bud galling species
and Maia
Lamprosoma azureum (Coleoptera: Chrysomelidae) Poor- non-target effects
Germar
Unidentified Psyllid (Hemiptera: Psyllidae) Good- leaf galling species
Tectococcus ovatus Hempel (Hemiptera: Eriococcidae) Good- leaf galling species
Haplostegus epimelas (Hymenoptera: Pergidae) Poor- non-target effects
Konow
Sycophilia sp. (Hymenoptera: Eurytomidae) Good- seed galling species
Eurytoma sp. (Hymenoptera: Eurytomidae) Good- stem galling species


The Brazilian researchers identified seven potential biological control agents (Table

2-1) (Wikler et al. 2000). Two of these species were determined unsuitable because of

non-target effects. In initial studies, the sawfly Haplostegus epimelas Konow attacked

common guava, and the chrysomelid Lamprosoma azureum Germar attacked common

guava and other myrtaceous species (Wikler et al. 2000). Of the remaining five agents, it

was determined that the leaf galling eriococcid Tectococcus ovatus was the most

promising agent because of the type of damage inflicted and the ease of handling (Wikler

et al. 2000). A shipment of T. ovatus was then sent to the Institute of Pacific Islands

Forestry in Volcano, Hawaii to establish a laboratory colony for host specificity testing.









In 2001, the University of Florida initiated a biocontrol program against strawberry guava

in the state. Shipments of T ovatus were sent from the Hawaii colony to the FDACS

Division of Plant Industry, Florida Biological Control Laboratory in Gainesville, Florida,

to establish a colony for host specificity testing.

Tectococcus ovatus Hempel

Higher Classification

The higher taxonomic classification of the Order Hemiptera has historically been

one of great debate. The term higher classification is in reference to the ordinal and

subordinal taxonomic level. The Order Hemiptera often has been divided into two

separate orders: the Hemiptera and the Homoptera (Borror et al. 1989). This earlier

classification scheme was based solely on morphological characteristics. With the advent

of molecular systematics, many authors suggested that Homoptera was not a

monophyletic group (Campbell et al. 1995, Schuh and Slater 1995, Sorensen et al. 1995).

Recent molecular data supports this, showing that the order Homoptera is paraphyletic.

Currently, most taxonomists agree and combine the traditional Orders Homoptera and

Hemiptera into one order, the Hemiptera (sensu lato). This can be confusing, therefore

unless otherwise mentioned in this manuscript; Hemiptera will be used in the broad sense

(s.1.).

A debate also exists regarding the organization of the suborders. The Order

Hemiptera is usually split into 4 or 5 suborders; and different authors often suggest

different names for the suborders which further confuse the issue. To date, the most

complete treatment of the higher classification of the Hemiptera has been compiled by

Bourgoin and Campbell (2002). Bourgoin and Campbell (2002) built their phylogeny,

using a combination of morphological, molecular, and fossil data. They divide the









Hemiptera into 5 suborders; the Sternorrhyncha, Fulgoromorpha, Cicadomorpha,

Coleorrhyncha, and the Heteroptera (Fig. 2-5). Tectococcus ovatus belongs to the family

Eriococcidae. This family is grouped under the suborder Sternorrhyncha, which also

contains the psyllids and aphids. Most authors agree on the placement of the scales,

psyllids, and aphids into the Sternorrhyncha (Bourgoin and Campbell 2002, Campbell et

al. 1995, Schuh and Slater 1995, Sorensen et al. 1995).



Sternorrhyncha

Hemiptera -Fulgoromorpha

-Cicadomorpha

Coleorrhyncha


Heteroptera


Figure 2-5. Phylogeny proposed by Bourgoin and Campbell (2002) for the higher
classification of the Hemiptera based on morphological, molecular, and fossil
data.

Taxonomy

Members of the family Eriococcidae are commonly called felt scales or felted

scales. The family contains approximately 50 genera and 350 species (Hoy 1963).

Aside from the absence of paired anal plates, there are very few defining characteristics

of the family. In fact, the family seems to be comprised of a collection of unrelated

groups (Rung et al. 2005). Recent molecular data using small subunit rDNA supports

this conclusion, rendering the family paraphyletic (Cook et al. 2002). Although these

findings have serious nomenclatural implications, Cook et al. (2002) regard their data as









preliminary and suggest that the group remain intact until more extensive studies are

conducted. The currently accepted classification scheme for T. ovatus is as follows:

Class: Insecta

Order: Hemiptera

Suborder: Sternorrhyncha

Superfamily: Coccoidea

Family: Eriococcidae

Genus: Tectococcus

Species: Tectococcus ovatus Hempel

Life History

Gall description

Leaf galls of T. ovatus each contain one insect and are visible on both sides of the

leaf. There is one opening per gall and this is at the apical portion of the gall. The

opening of the gall is formed on the side of the leaf upon which the insect originally

initiates feeding (Vitorino et al. 2000). The galls are generally acuminate on both sides

of the leaf; occasionally the gall may be acuminate only on the side of the leaf with the

opening and convex on the other side. The inside of the gall is flat and covered with a

fine powdery wax (Vitorino et al. 2000). The size of the gall is variable and depends on

the developmental stage and sex of the insect. The galls are sexually dimorphic; the base

of the female gall is much wider than that of the male (Fig. 2-6).

Morphology

The genus Tectococcus is monotypic. Tectococcus ovatus was originally described

by the Brazilian entomologist Hempel in 1900. Ferris (1957) described and illustrated

the adult female and first instar "crawler" stage (Fig. 2-7). Adult males are typically not









described in detail because they are rarely collected in the field and are not used for

species identification. Tectococcus ovatus is a small species approximately 1.5 mm long,

ovate in form with the caudal end acuminate (Fig. 2-8). Derm is membranous and pink to

brown in color, dusted in a fine white powder (Vitorino et al. 2000). Legs are present and

well developed although the adult female is relatively sessile, spending her entire life

within the confines of a leaf gall. The antennae are six segmented with the first joint

being the longest. The major distinguishing feature of the species is the small, slightly

sclerotized and hairless anal plates (Ferris 1957, Hempel 1901). The males are slight and

narrow, either pink or light brown. Antennae are long and slender, approximately half of

the length of the body. Males have narrow legs and one pair of wings; they are capable

of weak flight.


A


Figure 2-6. Cross sectional view of the galls of T. ovatus A) Female B) Male.


































Figure 2-7. Ferris (1957) drawings of T ovatus A) Adult female B) First instar crawler.


A B



& s..
J I "i
t'1.: i


Figure 2-8. T. ovatus adult female A) Dorsal B) Ventral (scale bars represent 1 mm).









Biology

The first instar of T. ovatus is the mobile stage of the insect and they are commonly

called crawlers. Upon hatching, T. ovatus crawlers search the plant for suitable feeding

sites, the ideal site usually being new flush. Once a suitable site is found, the insect

begins feeding and becomes sessile. The plant responds by forming a gall around the

insect (Fig. 2-9 A, B, and C). Galls are typically formed on leaves, although they may

also form on floral buds, young branches and developing fruit (Vitorino et al. 2000). Gall

formation always begins on the same side of the leaf where the insect begins feeding.

Females are facultatively parthenogenetic, although there is at least one alternation of

generations per year (Vitorino et al. 2000). The female remains in the gall for her entire

life, whereas males are mobile in their adult stage. Once a female reaches the adult

stage, she begins producing eggs within the gall. The eggs are then extruded in a cottony

wax through the opening in the gall (Fig. 2-9 D). This wax probably aids in the dispersal

of the eggs via wind.

Nutritional Ecology

The majority of phytophogous insects consume large amounts of plant tissues or

fluids. To fulfill their nutritional requirements, these insects must move about their host,

feeding at multiple sites. Gall forming insects are unique in that they remain sessile, and

feed only on specialized nutritive cells that line the gall chamber (Dreger-Jauffret and

Shorthouse 1992). The gall former stimulates an abnormal growth process in its host,

from which it receives both nutrition and protection (Abrahamson and Weis 1987).

Because of this highly specialized relationship between the herbivore and its host, most

gall-inducing insects are highly host specific. The reduced risk of non-target effects









makes gall forming insect's ideal weed biological control agents (Harris and Shorthouse

1996).


Figure 2-9. Life stages of T. ovatus; A) gall initiation around first instar crawlers; B)
close up of leaf galls; C) view of galls on multiple leaves; D) eggs extruded in
cotton-like wax. (Photo credit: M. Tracy Johnson).

Plant galls can be divided into two basic groups, organoid and histioid (Rohfritsch

1992). Organoid galls differ slightly from the normal plant growth pattern while histioid

galls alter the basic growth patterns of the host. The galls of T. ovatus are histioid, or

more specifically they are categorized as prosoplasmic. Prosoplasmic refers to a histioid

gall which is highly organized and displays tissue differentiation (Rohfritsch 1992).

Galls of the Eriococcidae, however, typically have a less characteristic nutritive tissue

(Rohfritsch 1992).









The induction of abnormal cell growth by gall forming insects results in a

metabolic nutrient sink (Harris and Shorthouse 1996). In addition to depriving normal

plant cells of nutrients, high densities of T. ovatus leaf galls may inhibit photosynthesis.

Heavy infestations of T. ovatus result in premature leaf drop and in some cases complete

defoliation (Vitorino et al. 2000). There is anecdotal evidence that T ovatus infestations

also may reduce fruit production, reducing both the seed bank and potential fruit fly

breeding sites (Vitorino et al. 2000).

Recorded Host Range

When first described by Hempel (1900), the only reference to the host plant was

that it belonged to the family Myrtaceae. The majority of the older literature also is

vague regarding host species, mentioning only the family Myrtaceae (Da Costa Lima

1927, Hempel 1912, Lepage 1938, MacGillivray 1921). In a catalogue of Brazilian

insects published in 1936, the host records are more specific, although only common

names are cited (Da Costa Lima 1936). Da Costa Lima (1936) lists hosts as aracazeiro

(strawberry guava) and a plant known as "embira". Subsequent investigation revealed

that the name "embira" refers to two different species, in two different families,

Daphnopsis racemosa Griseb. (Thymelaeceae) and Rollinia salicifolia Schltdl.

(Annonaceae). This reference most likely led to the inclusion ofD. racemosa as a host

by Hoy (1963) in his catalogue of the Eriococcidae of the world. This reference is

probably erroneous and may be due to confusion between T. ovatus and its relative

Pseudotectococcus anonae Hempel (Johnson 2005). Whether the host record is

erroneous or not, members of the genera Daphnopsis and Rollinia from Florida and the

Caribbean would be appropriate for inclusion in a host specificity test plant list.






31


In field observations, T ovatus has been reported on other members of the genus

Psidium in Brazil. Psidium longipetiolatum Legrand and Psidium spathulatum Mattos

both have been reported as being attacked by T ovatus (Vitorino et al. 2000). These two

species are not present in the continental United States and therefore are in no danger of

being attacked by T. ovatus, should it be approved for release in Florida.














CHAPTER 3
LABORATORY BIOLOGY OF Tectococcus ovatus

Introduction

Tectococcus ovatus Hempel is a leaf galling scale insect (Eriococcidae) native to

the coastal regions of southeastern Brazil. The distribution of the insect is closely

correlated with the native range of its primary host plant, strawberry guava, Psidium

cattleianum Sabine (Myrtaceae); ranging from the Brazilian state of Espirito Santo in the

north to northern Uruguay in the south (Wikler 1995). Strawberry guava was originally

imported into Florida for the ornamental fruit trade in the late 1800s (Langeland and Hall

2000). However, it escaped cultivation and is now considered a highly invasive natural

areas weed in the state (IFAS 2005). In addition to invading native plant communities

and altering the natural ecological balance of plant and animal communities in Florida,

strawberry guava also is considered one of the major hosts of the Caribbean fruit fly,

Anastrepha suspense Loew (Tephritidae) (Nguyen et al. 1992, Swanson and Baranowski

1972).

Native to the West Indies, the Caribbean fruit fly is a highly polyphagous pest

species with nearly 100 recorded hosts (Weems et al. 2005). Populations of the

Caribbean fruit fly eventually became established in central and southern Florida in 1965

(Swanson and Baranowski 1972). This establishment was largely ignored until 1968,

when the fly was discovered in commercial grapefruit, Citrus xparadisi Macfad. (Greany

and Riherd 1993). To eliminate the spread of this pest, many domestic and foreign

markets initiated quarantines on fresh citrus shipments from Florida. Multiple









management strategies are currently being used to combat Caribbean fruit fly infestations

in the state. These strategies include the sterile insect technique, classical biological

control, and the development of the fruit fly-free protocol, an integrated approach to

certify citrus crops as fly-free (Baranowski et al. 1993). The fruit fly-free protocol

involves a combination of trapping, baiting, spraying, and the removal of major hosts

(including strawberry guava) from surrounding areas (FDACS 2005). Compliance with

this protocol can be problematic because the citrus grower is responsible for removal of

major hosts from adjacent properties, even if they do not own the property (FDACS

2005).

Controlling strawberry guava infestations mechanically and chemically are viable

options for easily accessible plants. However, these methods are not practical in

environmentally sensitive areas (e.g., in natural areas and state parks and preserves).

Classical biological control of strawberry guava is ideal because once released, biological

control agents are self sustaining, can locate less accessible plants, and do not require

landowner permission.

A colony of T. ovatus was established at the Florida Department of Agricultural

and Consumer Services (FDACS), Division of Plant Industry (DPI) Florida Biological

Control Laboratory (FBCL) in Gainesville, FL. Currently, this insect is under

investigation for potential release as a biological control agent for strawberry guava in

Florida. High infestations of T ovatus act as a nutrient sink. Diverting nutrients from

plant growth and reproduction can cause premature leaf drop, may reduce photosynthesis,

and inhibit fruit production, ultimately reducing fruit fly breeding sites (Vitorino et al.

2000).









Prior to the release of any classical biological control agent, it is necessary to

understand the biology and ecology of the agent. The purpose of this study was to make

detailed biological observations on T. ovatus in order to expand on the limited

information available in the literature regarding this species. An important factor in

understanding the developmental biology of an insect is determining the number of

instars and distinguishing these from the adult. The soft bodied nature of this insect

makes it difficult to determine the number of instars by traditional means. Because of

this, multivariate analyses of multiple leg measurements were conducted. The legs were

selected for these analyses because they are sclerotized and readily identifiable. The

majority of the life cycle of this species occurs inside of a protective plant gall.

Therefore, in order to better understand the life history of T. ovatus, a linear regression

was constructed to determine if a correlation exists between the size of the leaf gall and

the life stage of the insect (excluding the egg). This information may be useful for

studies where destructive sampling of specimens is not a feasible option, for example, the

construction of age-specific life tables.

Materials and Methods

Tectococcus ovatus specimens used in this study were obtained from the

laboratory colony maintained at the FDACS DPI Florida Biological Control Laboratory

in Gainesville, FL. All biological observations were made from insects reared in this

colony. The colony was maintained in acrylic cylinders, 46 cm tall and 15 cm in

diameter. These cylinders were placed over the host plants and the bottom was partially

buried in the soil to prevent T. ovatus from escaping. The acrylic cylinders were

ventilated by six holes 6 cm in diameter. The ventilation holes and the top of the cylinder

were covered with a fine mesh, with a screen size of 150 [i x 150 [i (Green.tek Inc.,









Edgerton, WI). The colony was maintained under natural light conditions in a quarantine

greenhouse, supplemented with fluorescent light (40 Watts) set on a 14:10 light: dark

photoperiod. Average colony temperature inside of the cylinders was 28.88 1.61 C

and the average humidity was 66.13 6.95 %. Insects were reared on the yellow fruiting

variety of strawberry guava (P. cattleianum var. lucidum) either grown from seeds

collected in the field or purchased from nurseries within Florida. The yellow variety was

selected because preliminary studies indicated that T. ovatus preferred this to the red

fruiting variety of strawberry guava (P. cattleianum var. cattleianum).

Five specimens of T. ovatus were dissected from leaf galls every other day for 30

days during development, resulting in a total of 75 specimens. This method of sampling

was chosen to ensure that every developmental stage was included in the analysis.

During the slide mounting process, 13 specimens were lost or damaged. All 13

specimens that had missing data were eliminated from this study, leaving a total of 62

specimens to be analyzed.

Before specimens were dissected from their protective plant galls, the width of the

insect gall was measured using micrometer calipers. These measurements were taken in

order to determine if a correlation exists between the gall size of T ovatus and the

various life stages after hatching. Due to variation between galls, the portion of the gall

measured was the diameter at the base of the apical portion (Fig. 3-1). Preliminary

investigation indicated that this portion of the gall had the least amount of variation. The

specimens were then slide mounted according to a protocol modified from Wilkey

(1992). Prior to slide mounting the specimens were cleared in 10% KOH for 24 to 48 h.

Specimens were then stained with #6379 double stain (BioQuip Products, Inc. Rancho









Dominguez, CA), and transferred into 95% EtOH (15 min) and then submerged in clove

oil (30 min) (Ward's Natural Science, Rochester, NY). Canada balsam (Fisher Scientific

Co., Pittsburgh, PA) was used as the slide mounting medium. The slides used in this

study were 75 x 25 mm Fisherbrand plain pre-cleaned slides (Fisher Scientific Co.,

Pittsburgh, PA) used with Fisherbrand (Fisher Scientific Co., Pittsburgh, PA) 12 mm

circular cover glass.






















Figure 3-1. Cross section depicting variation in female leaf galls of T. ovatus. Arrows
indicate the diameter of the apical base of the gall where measurements were
taken.

In order to determine the number of nymphal stages and distinguish them from

the adult, multiple leg measurements were recorded for each insect. The legs were

selected as the identifying feature because they are heavily sclerotized and readily

identifiable; T. ovatus is a soft bodied insect and slide mounting can alter the dimension

of non-sclerotized parts. Due to the discontinuous nature of insect development, the

number of instars and the adult can be determined by comparing variation of multiple leg









measurements. The length of the fused trochanter/femur segment, the tibia, and the

tarsus was recorded for one the pairs of each leg (prothorcacic, mesothoracic, and

metathoracic). In addition, the width of the fused trochanter/femur and tibia was recorded

at their widest points on each leg. This procedure resulted in a total of 15 measurements

for each specimen. The slide mounted specimens were digitally photographed with a

JVC model KY-F70B 3-CCD digital camera (JVC Americas Corp.) mated to a Leica

DMLB compound microscope (Leica Microsystems AG) with a Diagnostic Instruments

T-49C 0.45x c-mount coupler (Diagnostic Instruments, Sterling Heights, MI). The

Syncroscopy Auto-Montage software (Synoptics Ltd., Frederick, MD) was used to

measure the images at a resolution of 1360 x 1024 pixels. This system also was used to

measure the eggs of T. ovatus. Measurements were recorded digitally in order to reduce

error (the microscope only had to be calibrated once); this method also is less time

consuming than using an ocular micrometer.

Data were analyzed using the SAS statistical software (SAS Institute Inc., Cary,

NC). Due to the number of dimensions measured (15 observations per insect; n = 62), a

principal components analysis was performed using PROC PRINCOMP. This procedure

reduced the dimensions of the data by deriving a small number of linear combinations (in

this case 2 principal components) from the data. Next, a cluster analysis was performed

on the results of the principal components analysis using PROC FASTCLUS in order to

delineate distinct clusters of observations. Due to the discontinuous nature of insect

growth, these data clusters were used to differentiate the number of instars and separate

them from the adult stage. PROC FASTCLUS also calculated the mean and standard

deviation for all 15 morphometric parameters in each cluster. Finally, to obtain a better









graphical representation of the clusters, a canonical analysis was performed using PROC

CANDISC. This analysis transformed the data from the FASTCLUS analysis into two

canonical variables (Cani and Can2).

In order to correlate gall size to a particular life stage (excluding the egg), a

regression analysis was performed using Microsoft Excel (Microsoft Corporation,

Redmond, WA). The diameter at the base of the apical portion of the gall was correlated

with the length of the fused prothoracic trochanter/femur. The developmental instar can

then be determined by comparing the prothoracic trochanter/femur length measurement

with the mean length provided by the FASTCLUS analysis (see Table.3-2 in Results and

Discussion).

Results and Discussion

Biology

Tectococcus ovatus has a simple life cycle which has been previously described by

Vitorino et al. (2000). Eggs are deposited inside the gall of the female, and are then

extruded from the gall opening in a filamentous waxy secretion which may aid in their

dispersal by wind. The eggs are oval in shape and range in color from nearly white to a

light yellow (Fig. 3-2). Average egg length is 0.216 0.008 mm and average width is

0.115 0.006 mm (n = 20). Upon hatching, the mobile first instar or "crawler" disperses

on the plant in search of a suitable feeding site. Ideal feeding sites are young flushes of

leaf growth. Vitorino et al. (2000), however, mentions that galls also can form on floral

buds, young branches, or developing fruit. Once a suitable feeding site is found, feeding

elicits a plant response to form a gall around the sessile insect. The female will spend the

rest of her life within the confines of this protective gall. However, the winged male is

mobile as an adult (Fig. 3-3).
























4a



4.


.bl* .
'p


bB


E
E
C;
o


D



E
E
0


Figure 3-2. Developmental stages of female T ovatus; A) egg (abnormal coloration due
to preservation in 70% EtOH); B) first instar "crawler"; C) second instar; D)
adult (possible third instar not depicted).















Figure 3-3. T. ovatus adult male (scale bar represents 0.5 mm).


~rq 5Y



r r


f Vi%


;r
i"











Vitorino et al. (2000) describes reproduction as facultatively parthenogenetic with

at least one alternation of generations per year. Prior to this study, copulation has never

been observed. In this study, copulation was observed only once in the laboratory. Prior

to copulation, the male inspected the opening of the gall with his antennae, presumably

looking for a receptive female. The receptive female partially exposed her posterior end

from the gall. The male turned around and rubbed the posterior portion of his abdomen

around the anal area of the female in an apparently random writhing motion. Copulation

lasted for approximately 40 seconds. After copulation, the male continued searching the

leaf, possibly for other females, and the mated female re-entered her gall.

Separation of Nymphal and Adult Stages of Tectococcus ovatus

A traditional method of determining the number of instars of an insect is by

constructing a frequency histogram of a particular body measurement (Kishi 1971).

Typical measurements for this type of study are inter-ocular distance or some other head

capsule measurement; instars are identified as distinct peaks in the histogram (Daly

1985). This method works reasonably well in cases where the data produce discrete non-

overlapping peaks. To check if any instars were missed during visual inspection of the

histogram, the logarithms of the means for each mode are plotted against the presumed

number of instars. If all instars are included, a straight line should be observed (Daly

1985). This analysis is based on the prediction of Dyar (1890) that the head capsule size

increases by a constant geometric progression every molt, commonly known as Dyar's

law. However, this type of analysis only looks at one dimension of growth, and it has

been shown that the limited amount of dimensions may result in the misinterpretation of

the number of instars, particularly if the peaks are not discrete (Kishi 1971; Schmidt et al.









1977). In addition, Dyar's law does not necessarily hold true for every insect, or even

within a single family (Daly 1985, Gaines and Campbell 1935).

Describing morphological variation in biological organisms using a multivariate

morphometric analysis is a relatively old technique that has been widely applied (Daly

1985). The term multivariate morphometric analysis is used to describe any multivariate

statistical procedure used to describe relationships between measurements of a biological

organism. Blair et al. (1964) indicates that multivariate morphometric analyses have

potential for describing morphological variation in difficult groups such as the

Coccoidea. Blair et al. (1964) looked at the variation in a homogenous population of

Coccus hesperidum L. based upon the analysis of multiple measurements of the legs and

antennae. Similarly, Boratynski and Davies (1971) analyzed multiple morphometric

characters to describe taxonomic variation in male coccids. Insects do not grow

continuously but rather in discontinuous steps or instars. Therefore, it should be possible

to measure the average size increase of each instar of a particular species by using this

type of analysis.

By analyzing the 15 separate leg measurements per specimen, it was possible to

determine the number of life stages (excluding the egg). The results of the FASTCLUS

cluster analysis of the principal components indicate that there are three distinct clusters.

These clusters may represent two instars and the adult. However, the analysis also

identified a weak fourth cluster. This may indicate the presence of a supernumerary

instar in addition to the adult. The transformation produced from the canonical analysis

provides a better display of these results (Fig. 3-4). The first three clusters are clearly

visible; while the fourth cluster is shown by the five star shaped points on the right of the









graph. Typically, females in the family Eriococcidae have two instars in addition to the

adult (Stehr 1991). There are a few possible explanations for the potential presence of a

fourth cluster. There may be a wide range of size variation in the size of the adult of T

ovatus. In this case, the weak fourth cluster could be the result of outlying data from

measurements of extremely large adults. Outlying data are more difficult to identify

when comparing data sets with several dimensions such as this (15 observations per

specimen). The presence of a fourth cluster could also be due to variation in the initial

measurements. The measurements taken from the Auto-montage image were in two

dimensions; slight errors in the measurements could have occurred if the leg segment of

the slide mounted specimen was not perfectly horizontal. Both of these problems could

be solved by taking a larger sample size (n > 62). Another explanation is that there is a

supernumerary instar; this could also be elucidated by analyzing a greater sample size.

The occurrence of supernumerary molts in laboratory reared colonies is not an

uncommon observation (Chapman 1998).

Correlation of Gall Size to Nymphal and Adult Stages of Tectococcus ovatus

By comparing r2 values for multiple regressions of gall size vs. insect

measurements, the length of the prothoracic trochanter/femur segment was determined to

have the closest relationship with the gall width (Table 3-1). Figure 3-5 illustrates the

close correlation between these two variables; the best fit equation is y = 44.603x +

16.233 (r2 = 0.7126; df = 1, 65; p < 0.001).







43




Table 3-1. r2 values for multiple regressions of separate leg measurements of T. ovatus vs.
the gall size. Gall size is determined by the diameter of the base of the apical
portion of the gall.
Insect Measurement
(Length) r2 Value
Prothoracic Troch/Fem 0.7126
Mesothoracic Troch/Fem 0.7070
Metathoracic Troch/Fem 0.7114
Prothoracic Tibia 0.6718
Mesothoracic Tibia 0.6809
Metathoracic Tibia 0.7095


4-

3-





-


-3
-1 -
-2

-3-


-20


-10


***1 u j2 4+++3 m X 4


Figure 3-4. Two dimensional representation of the FASTCLUS cluster analysis. PROC
CANDISC was used to generate two canonical variables (CanI and Can 2) for
graphing the results of the cluster analysis.

The mean and standard deviation of the lengths for the prothoracic


trochanter/femur segments of each cluster were provided by the FASTCLUS cluster

analysis (Table 3-2). The mean and standard deviation are provided for each of the four


*+


-I-


++4-
+ +4
-4









clusters because the possibility of a supernumerary instar could not be ruled out. In

Table 3-2, the first and second clusters represent the first and second instar respectively.

Further sampling is necessary, but the third cluster may represent either a supernumerary

third instar or the adult stage. The fourth cluster represents the adults with three instars

or outlying data from measurements of extremely large adults.


160
u1 y = 44.603x + 16.233 Upper
r 140 2 Upper
SR2 = 0.7126 95% CI
2* 120 *

c 100
m Lower
S 80 95% CI
-,

E 60 o
S40 '- Linear
S, (Best Fit
o 20 Line)
I-
0
0 0.5 1 1.5 2 2.5
Gall Size (mm)


Figure 3-5. Linear regression analysis of the relationship between the diameter of the
apical base of the plant gall and the fused prothoracic trochanter/femur length
ofT. ovatus (r2 = 0.7126; df = 1, 65; p < 0.001).

Table 3-2. Mean and standard deviation of the length of the fused prothoracic
trochanter/femur segment for each developmental instar.
Cluster Mean Prothoracic Trochanter/ Standard Deviation
Femur Length (it)

1 (First Instar) 35.748 5.706
2 (Second Instar) 55.315 5.118
3 (Third Instar or Adult) 101.706 6.823
4 (Adult) 116.914 4.390






45


Acknowledgements

The author would like to thank Greg Hodges and Yen Dao for instruction in

preparing the slide mounted specimens. The multivariate statistical analyses conducted

for this study would not have been possible without the professional assistance of

Meghan Brennan and Kenneth Portier. In addition, thanks are due to Alejandro Arevalo

for his help with analyzing the regression data. This research was funded by the USDA

CSREES Tropical/Subtropical Agriculture Research program (T-STAR-Caribbean) grant

No. 01062227.














CHAPTER 4
HOST SPECIFICITY OF Tectococcus ovatus

Introduction

Strawberry guava, Psidium cattleianum Sabine, is a woody evergreen tree or shrub

native to the coastal regions of southeastern Brazil. Closely related to common guava

Psidium guajava L., strawberry guava was introduced to numerous countries worldwide

because of its small edible fruit, attractive foliage, and broad environmental tolerances

(Morton 1987). Strawberry guava was first introduced into Florida by the horticultural

trade in the late 1800s (Langeland and Hall 2000). The delicate nature and short shelf

life of the fresh fruit has inhibited the commercial potential of the plant (Schroeder and

Coit 1944). There are no horticultural cultivars of strawberry guava, although there are

two, possibly three varieties that are distinguished solely by the color and shape of the

fruit. Two varieties are present in Florida, a yellow fruiting variety, P. cattleianum var.

lucidum, and a red fruiting variety, P. cattleianum var. cattleianum (Wikler 2000b).

Strawberry guava is still commonly sold as an ornamental hedge and fruit tree in

the state. This is despite the fact that the plant has escaped cultivation and is invading

natural areas within the southern and central parts of Florida (Langeland and Hall 2000).

Factors that aid in the spread of strawberry guava are its ability to grow in low light

conditions, reproduce vegetatively, produce large amounts of fruit which are attractive to

birds and other animals, and lack of natural enemies (Vitorino et al. 2000, Huenneke and

Vitousek 1990). Because of this, the Florida Exotic Pest Plant Council and the

University of Florida, Institute of Food and Agricultural Sciences assessment of the status









of non-native plants in Florida's natural areas (IFAS Assesment) has categorized

strawberry guava as an exotic invasive species (FLEPPC 2003, IFAS 2005). In natural

areas, strawberry guava can out-compete native plant species and form dense mono-

specific stands, which alter native plant and animal assemblages (Tunison 1991).

In addition to the threat to native ecosystems, strawberry guava also is a major host

of the adventive Caribbean fruit fly, Anastrepha suspense Loew (Nguyen et al. 1992).

The Caribbean fruit fly, a native of the West Indies, is an extremely polyphagous species

with almost 100 recorded hosts (Weems et al. 2005). In 1968, the Caribbean fruit fly was

discovered in commercial grapefruit, which was previously thought not to be a host

(Greany and Riherd 1993). To eliminate the potential spread of Caribbean fruit fly in

fresh citrus shipments, the shipments were fumigated with ethylene dibromide. However,

ethylene dibromide was banned by the Environmental Protection Agency for this purpose

in 1984 (Nguyen et al. 1992). This led to the development of alternative methods to

control Caribbean fruit fly populations, such as the sterile insect technique, classical

biological control, and the Caribbean fruit fly-free protocol. Participation in the

Caribbean fruit fly-free protocol is necessary if a grower intends to export fresh fruit to

quarantine sensitive domestic and foreign markets (FDACS 2005). The protocol involves

a combination of trapping, baiting, spraying, and the establishment of a buffer zone

around the designated grove. This buffer zone consists of an area free of major hosts

extending 1.5 miles from the perimeter of the designated grove; major hosts are common

guava, strawberry guava, Surinam cherry (Eugenia uniflora L.), rose apple (Syzygium

jambos L.), and loquat (Eriobotryajaponica (Thunb.) Lindl.). The citrus grower is

responsible for removal of major hosts from the buffer zone, and it is their responsibility









to negotiate with property owners regarding removal (FDACS 2005). This can lead to

disputes that can hinder citrus growers from exporting their product. Controlling major

host plants with classical biological control may help solve this problem by reducing

plant populations without active involvement from the citrus grower or property owners.

The classical biological control program against strawberry guava began in Hawaii,

where the plant is considered the worst invasive weed in the archipelago (Smith 1985).

In 1991, the U.S. National Park Service and the University of Hawaii formed a

collaboration with the Federal University of Parana, Brazil to identify and evaluate

potential biological control agents (Wikler et al. 2000). Five potential agents were

identified and the leaf galling erioccocid, Tectococcus ovatus Hempel was determined to

be the most promising based on the type of damage inflicted and the ease of handling

(Wikler et al. 2000). Tectococcus ovatus is a scale insect that forms galls on the leaves,

stems, and fruit of strawberry guava. Feeding and subsequent gall formation act as a

nutrient sink depriving the plant of nutrients used for growth and reproduction. High

infestations of T ovatus can cause premature leaf drop, may inhibit fruit production, and

may reduce photosynthesis.

The damage caused by T ovatus directly effects the growth and sexual

reproduction of strawberry guava, which may ultimately reduce fruit fly breeding sites

(Vitorino et al. 2000). In addition, weakened plants will make removal easier in state

parks and preserves. However, prior to the release of any biological control agent, the

host range must be evaluated in order to ensure that the biocontrol agent will not harm

non-target species. The purpose of this study was to determine if T. ovatus is suitably

host specific in order to evaluate the potential for release in Florida.









Materials and Methods

The test plant list for this study was developed in accordance with the U.S. Dept. of

Agriculture, Animal and Plant Health Inspection Services Technical Advisory Group for

Biological Control Agents of Weeds (TAG) guidelines. The TAG guidelines are based

on the centrifugal-phylogenetic method developed by Wapshere (1974). The test plant

list is divided into 8 categories based on TAG guidelines and agricultural and economic

concerns in Florida. These 8 categories and the division of the test plant list into the

categories, as well as justification for inclusion are available in Appendix B. Alterations

made to the original test plant list are available in Appendix C.

Specimens for establishing the T. ovatus colony used for host specificity testing in

Florida were shipped to the Division of Plant Industry quarantine facility in Gainesville

from the Hawaii Volcanoes National Park Quarantine Facility in Hawaii (APHIS PPQ

526 permit 54024). Transport between the two states is more reliable and faster than

receiving shipments from Brazil. In addition, the colony of T ovatus in Hawaii was

already free of predators and parasitoids. The insects reared at the Hawaii quarantine

facility were obtained from an outdoor nursery colony established at the Federal

University of Parana (APHIS PPQ 526 permits 47452, 69049) (Johnson 2005). This

outdoor colony was obtained from field collected populations east of the city of Curitiba

(in the municipal districts of Piraquara, Sdo Jose dos Pinhais, and Colombo) (Lat -

25.51670, Long -49.1667) (Johnson 2005).

To transport the insects, strawberry guava leaves containing mature T. ovatus

galls were shipped in individual containers. The leaves were then placed on uninfested

caged strawberry guava plants, to allow emerging crawlers to establish on new plants.









Tectococcus ovatus colony production and host specificity experiments were

conducted at the Florida Dept. of Agriculture and Consumer Services, Division of Plant

Industry, Florida Biological Control Laboratory in Gainesville, Florida. Voucher

specimens of T ovatus were deposited in the Florida State Collection of Arthropods,

Gainesville, Florida.

Strawberry guava plants used to maintain the colony and as experimental controls

as well as all other test plants were maintained at the University of Florida, Department

of Entomology and Nematology, Gainesville, Florida. Plants used for these experiments

were either grown from seed, purchased, or collected from the field. If there was any

question regarding plant taxonomy or identification, a qualified botanist was consulted.

Test plants were not treated with systemic insecticides to eliminate the chance of these

chemicals altering the results. All control and colony plants were potted with Fafard

middleweight mix # 4 potting soil (Conrad Fafard Inc., Agawam, MA), test plants were

potted with soil mixtures appropriate for each species. Plants that required fertilization

were fertilized with Dynamite Plant Food (Florikan E.S.A. Corp., Sarasota, FL) 6 month

time-release pellets with a 13:13:13 (N:P:K).

No-choice host specificity tests were conducted because this type of test is rigorous

in nature (Heard 1997). Tests were replicated three times and the yellow fruiting form of

strawberry guava was used as a control (Heard 1997). When testing Rhexia lutea

(Melastomataceae) and Punica granatum (Punicaceae), only 2 replications were

conducted due to death of plants prior to testing. Control plants were set up at the same

time as the test plants. Being a leaf gall former, T ovatus requires new flush to produce a

gall. Therefore, prior to testing, test plants were pruned to induce the growth of new









flush. To break plant dormancy, Prunus angustifolia and P. persica were maintained in a

growth chamber with a 12:12 light: dark photoperiod at -1.11 OC for one month and then

placed outside under ambient conditions.


Figure 4-1. Test arena for host specificity experiments.

Twenty first instar nymphs or "crawlers" were placed on the new growth of each

test plant; no more than 5 insects were placed on one individual leaf. Test plants ranged

in height from 25 45 cm, and were planted in 3.8, 7.6, or 11.4 L pots. Insects were

transferred individually from a colony plant using fine forceps. Once the insects were


t F~L~ I~.L~-( IJC;~









transferred onto the test plant, an acrylic cylinder 46 cm tall and 15 cm in diameter was

placed over the plant and the bottom was partially buried in the soil to prevent the

crawlers from escaping. The acrylic cylinders were ventilated by six holes 6 cm in

diameter. The ventilation holes and the top of the cylinder were covered with a fine

mesh, with a screen size of 150 i x 150 i (Green.tek Inc., Edgerton, WI) (Fig 4-1). Once

the test arenas were assembled, the plants were placed in a quarantine greenhouse. Test

plants were exposed to both natural and artificial light conditions. Supplemental

fluorescent lighting in the greenhouse was set on a 14:10 light:dark photoperiod. The

average temperature inside test cylinders was 28.88 1.61 C and the average humidity

was 66.13 6.95 %. Tests were conducted for a duration of 2 weeks, after which the

plants were inspected for the presence of T. ovatus or gall development. If surviving T.

ovatus were found, then the tests were extended for another 2 weeks and subsequently re-

examined.

Results

In total, 57 species representing 21 families were tested. Tectococcus ovatus only

survived and formed viable galls on strawberry guava (Table 4-1). However, T. ovatus

survived longer than the 2 week test period on three species; strawberry guava, Brazilian

guava (Psidiumfi i' / Ii /1hiliil/lnin/ O. Berg), and Costa Rican guava (Psidium guineense

Sw.). Because of this, the Brazilian guava and Costa Rican guava tests were extended for

an additional 2 weeks. Tectococcus ovatus fed on Brazilian guava although no gall was

formed and all insects died within 4 weeks. Feeding on Costa Rican guava induced a

weak gall, which was poorly formed; the galls were cuplike in shape and did not fully

cover the insect, as a normal gall would. In total, 5 of these galls were formed and the

insects did not survive longer than 4 weeks on Costa Rican guava.









Table 4-1. Results of T. ovatus host specificity testing. A "+" indicates feeding damage
and gall development, whereas a "-" indicates a lack of feeding damage and
gall development.


Test Plant
Psidium cattleianum var. lucidum
Sabine
Psidium cattleianum var. cattleianum
Sabine
Psidium friedrichsthalianum O. Berg
Psidium guineense Sw.
Psidium guajava L.
Acca sellowiana (0. Berg) Burret
Eugenia axillaris (Sw.) Willd.
Eugeniafoetida Pers.
Eugenia uniflora L.
Myrciaria cauliflora (C. Martius) O.
Berg
Pimenta dioica (L.) Merr.
Pimenta racemosa (P. Mill.) J.W.
Moore
Syzygium malaccense (L.) Merr. &
Perry
Syzygium paniculatum Gaertner
Callistemon citrinus (Curtis) Staph
Callistemon viminale (Gaertn.) G.Don
ex Loudon
Eucalyptus camaldulensis Dehnhardt
Leptospermum scoparium J.R. & G.
Forst.
Melaleuca quinquenervia (Cav.) Blake
Cdlp i n//ihe pallens Griseb.
CatllJ)pintuhe' zuzygium (L.) Sw.
Eugenia confusa DC.
Eugenia rhombea Krug & Urban
Mosiera longipes (Berg) McVaugh
3 i Li i /m/he fragrans (Sw.) M Vaugh
Ammannia coccinea Rottb.
Cuphea hyssopifolia Kunth
Cuphea micropetala Humb., Bonpl. &
Kunth
Decodon verticillatus (L.) Ell.
Lagerstroemia indica L.
Lythrum alatum Pursh
Rhexia lutea Walt.
Rhexia mariana L.
Rhexia nashii Small


Family
Myrtaceae

Myrtaceae

Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae

Myrtaceae
Myrtaceae

Myrtaceae

Myrtaceae
Myrtaceae
Myrtaceae

Myrtaceae
Myrtaceae

Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae
Myrtaceae
Lythraceae
Lythraceae
Lythraceae


Results Replications
+ 50

+ 3


Lythraceae
Lythraceae
Lythraceae
Melastomataceae
Melastomataceae
Melastomataceae









Table 4-1. Continued.
Test Plant Family Results Replications
Tetrazygia bicolor (P. Mill.) Cogn. Melastomataceae 3
Rollinia mucosa (Jacq.) Baill. Annonaceae 3
Punica granatum L. Punicaceae 2
Conocarpus erectus L. Combretaceae 3
Chrysobalanus icaco L. Chrysobalanaceae 3
Nyssa sylvatica var. biflora Walt. Nyssaceae 3
Daphnopsis americana (P. Mill.) J.R. Thymelaeaceae 3
Ilex cassine L. Aquifoliaceae 3
Ilex x attenuata Ashe Aquifoliaceae 3
Delonix regia (Bojer ex Hook) Raf. Fabaceae 3
Quercus hemisphaerica Bartr. ex Fagaceae 3
Willd.
Persea americana P. Mill. Lauraceae 3
Ficus aurea Nutt. Moraceae 3
Myrica cerifera (L.) Small Myricaceae 3
Saccharum officinarum L. Poaceae 3
Eriobotryajaponica (Thunb.) Lindl. Rosaceae 3
Prunus angustifolia Marsh. Rosaceae 3
Prunuspersica (L.) Batsch Rosaceae 3
Pyrus x lecontei 'Hood' Rosaceae 3
Citrus limon (K.) Burm. F. Rutaceae 3
Citrus x paradisi Macfad. Rutaceae 3
Citrus sinensis (L.) Osbeck Rutaceae 3
Taxodium distichum (L.) L.C. Cupressaceae 3
Pinus elliottii Engelm. Pinaceae 3
Podocarpus macrophyllus (Thunb.) Podocarpaceae 3
Sweet
a T. ovatus survived longer than the 2 week test period; test was extended to 4 weeks, but
no damage or gall formation was observed.
b T. ovatus survived longer than the 2 week test period; test was extended to 4 weeks,
weak leaf gall formation was observed.

Discussion

The results of the host specificity tests show that T ovatus is highly host specific;

feeding and weak gall formation were only observed on two species closely related to the

target weed, Costa Rican guava and Brazilian guava. These results are not surprising

because T. ovatus has been reported to attack a close relative, Psidium spathulatum









Mattos in Brazil (Vitorino et al. 2000). This species is not native to North America or the

Caribbean and therefore is in no danger, should T ovatus be released.

Costa Rican guava is found throughout South and Central America (Morton 1987).

There is no commercial production but the fruit can be used to make jellies or fruit drinks

(Morton 1987). This guava is occasionally grown as a minor ornamental species in

Florida, although it is not commonly listed by most nurseries. Tectococcus ovatus was

able to feed on Costa Rican guava, and weak gall formation was observed. This

interaction may be explained by the conservative nature of no-choice testing because this

association has never been reported to occur in the wild. However, Costa Rican guava is

not commonly grown as an ornamental or fruit crop within the United States, therefore

damage inflicted by T. ovatus may not be of concern. Tectococcus ovatus also was

observed feeding on Brazilian guava, although feeding damage was not noticeable and no

gall formation was observed. Because feeding by T ovatus did not appear to have any

noticeable adverse effect on the plant, and all insects died within 4 weeks, this behavior

should not pose a risk to Brazilian guava if the insect were approved for release in

Florida.

Most importantly, T ovatus did not attack common guava, which is grown as a

fruit crop in south Florida. Originally, the biological control of strawberry guava was

thought to be impossible because of the close relationship between common guava and

strawberry guava (Wikler et al. 2000). These results have been confirmed in Brazil,

where the two species are found within the same range, and in host specificity tests in

Hawaii (Vitorino et al. 2000; Johnson 2005).









In a literature search of additional hosts of T ovatus, Daphnopsis racemosa Griseb.

(Thymelaceae) was listed as a host in a worldwide catalog of the family Eriococcidae

(Hoy 1963). The references from this catalog were obtained, and the reference to this

host plant association was traced back to a catalog of Brazilian insects (Da Costa Lima

1936). Da Costa Lima (1936) lists T ovatus as producing galls on the leaves of

strawberry guava and another plant called "embira". Subsequent investigation revealed

that the common name embira refers to two plants in two different families D. racemosa

and Rollinia salicifolia Schltdl. (Annonaceae). These particular species do not occur in

North America or the Caribbean and therefore are in no danger of being attacked by T

ovatus. However, there are members of the two genera present in the Caribbean,

including some endangered members of the genus Daphnopsis. Although this host

association is most likely erroneous, one representative of each genus (D. americana and

R. mucosa) was tested and were found not to be attacked by T. ovatus. An advantage of

choosing a gall forming insect as a biological control agent is they tend to have narrow

host ranges (Harris and Shorthouse, 1996). This is due to the complex co-evolutionary

relationship that gall forming insects have with their host plants. The results of this study

support this observation.

The most pressing issue that will need to be addressed prior to release of this insect

is the continued sale of strawberry guava as an ornamental in Florida. Despite evidence

that this species is invasive; the ornamental industry is reluctant to phase out strawberry

guava because of its economic value. The FDACS, DPI is the state agency charged with

implementing and enforcing laws regarding invasive plants. Their focus has historically

been on agricultural threats, most of which are not yet present in the state (FDACS 2004).









However, the IFAS assessment focuses on environmentally relevant invasive species.

The IFAS assessment is not a regulatory list; the primary goal of this assessment is to

direct research and extension at the University of Florida. A possible solution for

nurseries would be to substitute invasive species with native plants that have similar

desirable characteristics. A native plant substitution guide for Florida was developed by

FLEPPC (Ferriter 2003). Recommended substitutions were based on the aesthetic values

of the plants, and similarity of fruit characteristics. Three plants recommended as

substitutions for strawberry guava are Simpson's stopper, Iji i, nhe \fragrans (Sw.)

McVaugh, Guianese colicwood, Rapaneapunctata (Lam.) Lundell, and Jamaican caper,

Capparis cynophallophora L.

Based on the results of this study, T. ovatus is highly host specific and would make

a suitable biological control agent for the control of strawberry guava in Florida. The

non-target effects observed on Costa Rican guava and Brazilain guava were minimal, and

the insect was unable to complete its development on these guavas. However, the

continuing conflict with the nursery industry regarding the sale of guavas as ornamentals

in Florida needs to be resolved prior to the release of this organism.

Acknowledgments

The author would like to thank the botanists Mark A. Garland and Richard Weaver

for their professional assistance with nomenclature, identification, and collection of many

specimens used throughout this project. Generous plant donations were made by the

Chicago Botanical Garden and Ornamental Plants & Trees Inc. Thanks are also due to

Judy Gillmore for her assistance in procuring and maintaining test plants. This research

was funded by the USDA CSREES Tropical/Subtropical Agriculture Research program

(T-STAR-Caribbean) grant No. 01062227.














CHAPTER 5
DISCUSSION AND CONCLUSIONS

Prior to this study, little was known about the biology of T. ovatus. Literature

references were limited to mainly taxonomic descriptions (Ferris 1957) and catalog

citations (Hoy 1963). This is most likely due to the fact that T. ovatus is not an

economically important species and is limited in its geographical and host range.

Therefore, it is not surprising that prior to an outbreak of strawberry guava, the primary

host plant of this species, the main scientific value of T ovatus was taxonomic in nature.

The identification of strawberry guava as a major natural areas weed in Hawaii

and its association with another invasive species, the feral pig attracted attention to

strawberry guava and its natural enemies (Wikler et al. 2000). Initial investigations for

biological control were directed towards plant pathogens in hopes of developing a

bioherbicide. These initial explorations were not successful and the focus shifted to

highly specific phytophagous insects (Hodges 1988, Wikler 2000b, Wikler et al. 2000).

Previous studies on the biology of T. ovatus were preliminary in nature and

published in non-refereed proceedings (Wikler et al. 2000, Vitorino et al. 2000). These

two papers were the first published biological studies on T. ovatus. Wikler et al. (2000)

studied seven different natural enemies of strawberry guava. They identified T. ovatus as

being the most promising agent for biological control. Additionally, the paper included

general descriptions of the leaf gall, the male and female T. ovatus, a note on distribution,

records of two parasitoids, Metaphycusflavus Howard (Hymenoptera: Encyrtidae) and

Aprostocetus sp. Westwood (Hymenoptera: Eulophidae), and one predator, Hyperaspis









delicate Massuti and Vitorino (Wikler et al. 2000). Wikler et al. (2000) reported that T

ovatus is found more frequently on the red fruited variety of strawberry guava, whereas

Vitorino et al. (2000) stated that the coccid is more common on the yellow fruiting

variety.

Vitorino et al. (2000) provides a much more detailed account of biological

observations than Wikler et al. (2000). Vitorino et al. (2000) recorded mean diameters on

both sides of the leaf gall and height measurements. The gall sizes were divided into

three groups, based on the median for all of the measurements and one standard deviation

from the mean in both directions. Vitorino et al. (2000) also conducted tests to determine

the best method for transferring the first instar crawlers, and conducted preliminary host

specificity tests. Most biological observations were morphological in nature with a brief

description of life cycle. However, these two papers provided the basis for further

investigations into the biology of T ovatus.

Investigating the developmental biology of the female T. ovatus was chosen for a

couple of reasons. Female coccids are typically used in taxonomic descriptions of

species (Ferris 1957). This is because they are more persistent and usually sessile;

therefore, they are encountered more frequently in nature. Additionally, the life history

and physiology of female T. ovatus makes them more important for biological control.

Females are much larger because of their ability to produce progeny. Unlike males, they

must continue to feed in the adult stage in order to produce eggs. According to Stehr

(1991), female members of the Family Eriococcidae typically have two instars and one

adult stage. However, preliminary molecular investigations by Cook et al. (2002) suggest









that the family is not monophyletic. Therefore, researchers should be cautious of making

any general assumptions about the family.

Females of T ovatus spend the majority of their life inside a protective plant gall.

This makes investigating the development of the insect difficult without dissecting the

gall and disrupting the normal growth process. Therefore, an attempt was made to

correlate gall size to a particular nymphal stage or the adult, in order to make assumptions

about development without destructive sampling. However, before this could be

accomplished, the number of life stages (excluding the egg) had to be confirmed. This

turned out to be more difficult than anticipated. Typical methods of determining the

number of instars are by constructing a frequency histogram of a particular body

measurement (typically a head capsule measurement) (Daly 1985). However, this was

not possible for T ovatus because it is a soft bodied insect and slide mounting can alter

the shape of the soft integument, resulting in excessive variation in the measurements.

Measurements needed to be made of sclerotized structures that would not be altered by

the slide mounting process. The two sclerotized portions of T. ovatus are the mouthparts

and the legs; the legs were chosen because they are more distinct and easily identified

and measured. Multivariate morphometric analyses (principal components analysis,

cluster analysis, and canonical analysis) were conducted on 15 measurements of female

T ovatus leg. The results indicated the presence of two or possibly three instars. The

presence of supernumerary instars is not uncommon in laboratory colonies; however, the

possible third instar also could have been produced by outlying data points. It was

determined that increasing the sample size will be necessary to definitively determine the

correct number of developmental instars.









A regression analysis was performed to satisfy the original concept of correlating

instar to gall size. A significant correlation was found between the length of the fused

prothoracic trochanter/femur segment and the diameter at the base of the apical portion of

the leaf gall (r2 = 0.7126; doff. = 1, 65; p < 0.001). The length of the fused prothoracic

trochanter/femur can be used to correlate the life stage (excluding egg) with the gall size

(based on the morphometric analysis).

The original purpose of studying T. ovatus was to evaluate the potential of the

insect as a classical biological control agent of strawberry guava in Florida. The primary

concern of researchers attempting to evaluate classical biological control agents is the

safety of cultivated or socially important plants from attack by the agent (Wapshere

1979). The test plant list for this study was developed in accordance with the Technical

Advisory Group for Biological Control Agents of Weeds guidelines. These guidelines

are based on the centrifugal-phylogenetic method developed by Wapshere (1974). The

analysis of 57 species of plants representing 21 families resulted in only two relatively

minimal non-target effects. Tectococcus ovatus fed on two closely related Psidium

species, Brazilian guava and Costa Rican guava. Additionally, incomplete gall formation

was observed on the Costa Rican guava. Both of these tests were extended for 2

additional weeks, and no T ovatus specimens survived on either plant species. Due to

the limited value of these two guavas in Florida, these non-target effects were determined

to be negligible and not a concern if T ovatus were approved for release in Florida.

The most important result of the host range test was the lack of damage to common

guava, which is closely related to the target weed and commercially produced on a small

scale in Florida. Tectococcus ovatus did not attack common guava in this experiment and









these results have been confirmed by observations in Brazil and host specificity tests in

Hawaii (Vitorino et al. 2000, Johnson 2005).

These studies have elucidated a portion of the biology of T. ovatus, although the

continued evaluation of this insect in Florida is important. Further research on the

developmental biology of the insect is needed to establish the length or duration of each

instar. Additionally, data from these studies could be used to further correlate the gall

size of T ovatus with instar number. This correlation would make it possible to follow

multiple cohorts of individuals throughout their life cycle without disrupting the integrity

of the gall of the insect. The cumulative data would be useful for the constructing of

multiple age-specific life tables. These life tables could help researchers better anticipate

peaks in natural populations of T. ovatus, should the insect be released in Florida. These

tables also may assist in timing subsequent releases to coincide with natural population

spikes, increasing the chances of a successful release.

If T. ovatus is approved for released in Florida, studies also could be conducted on

the distribution rates of mobile life stages. This may aid researchers in predicting the

spread of T. ovatus under field conditions. The efficacy of T ovatus as a biological

control agent also could be better analyzed in the field. This could help researchers

determine if further studies on additional biological control agents for strawberry guava

should be pursued. This brings up the importance of evaluating the effect of generalist

predators and parasitioids on populations of T. ovatus in the field. Vitorino et al. (2000)

recorded a high parasitism rate of 49% within the native range of T ovatus.

Understanding how a biological control agent interacts within the new habitat is

fundamental to the continued progress of weed biological control as a science.









An important aspect of this study was the aim of reducing Caribbean fruit fly

populations by reducing breeding sites in strawberry guava. If T. ovatus is approved for

released in Florida, studies should be conducted on the subsequent effect on populations

of the Caribbean fruit fly. Field studies could be analyzed along with state fruit fly

trapping data collected before and after field releases of T. ovatus. A reduction of

Caribbean fruit fly populations that is correlated with reduction of breeding areas in

strawberry guava could justify the implementation of biological control programs against

other invasive hosts such as Surinam cherry (IFAS 2005).














APPENDIX A
WORLDWIDE DISTRIBUTION OF Psidium cattleianum

The following table, lists the worldwide distribution ofP. cattleianum. The table is

divided into seven sections based on biogeographic regions. The records are organized

by country or island chain and locations such as cities or particular islands are provided.












COUNTRY/ ISLAND LOCATION SOURCEa REFERENCE STATUS
CHAIN


AFROTROPICAL
Gana

Kenya
Madagascar




Mascarenes Islands
Mauritius

Reunion
Rwanda
Seychelles


Sierra Leone
South Africa

Tanzania


AUSTRALASIAN
Australia

Lord Howe Island
Micronesia
Melanesia


Wikler 1999


Nairobi
Antananarivo
Fianarantsoa
Toamasina
Toliara
Tamatave




Butare

Mahe
Silhouette Island

Natal, Durban
Natal, Durban
Amani
Lushoto


Lord Howe Island
Queensland


Wikler 1999
MBG
MBG
MBG
MBG
Wikler 1999
Weber 2003
MBG, Wikler 1999
Wyse-Jackson 1990
Wikler 1999
Wikler 1999
Wikler 1999
Gerlach 2004
Gerlach 2004
Wikler 1999
Wikler 1999
Henderson 1989
Wikler 1999
Wikler 1999


Wikler 1999
Wikler 1999
Wikler 1999
Weber 2003
Weber 2003


Invasive
Invasive
Invasive


Invasive
Invasive
Invasive

Invasive
Invasive


Introduced













COUNTRY/ ISLAND
CHAIN
New Caledonia
New Zealand
Norfolk Island
New Hebrides
Polynesia
Samoa
Solomon Islands

EAST PALEARCTIC
China


LOCATION

St. Louis


Pentecost Island

Upolu


SOURCEa

HS
LR
HS
HS
LR
HS
HS


Taiwan


NEARCTIC
Bermuda

Mexico
United States


Arizona
California

Florida
Missouri
Texas


REFERENCE"

MBG, Wikler 1999
Webb et al. 1988
Wikler 1999
Wikler 1999
Weber 2003
NYBG
Wikler 1999


Bretschneider 1898

Li and Huan 1979


Wikler 1999
Britton 1918
Weber 2003
ASU Herbarium
NYBG

MBG, NYBG, Wikler 1999
MBG
Jones et al. 1997


STATUS



Invasive

Invasive


Cultivated/ Possibly
Naturalized
Cultivated/ Possibly
Naturalized


Naturalized
Naturalized


Cultivated/ Possibly
Naturalized
Invasive


NEOTROPICAL
Argentina
Bahamas
Brazil


Wikler 2000
Morton 1987
NYBG, Wikler 1999


Bahia


Native


1












COUNTRY/ ISLAND
CHAIN
Brazil


Chile
Colombia
Costa Rica


Cuba
Dominican Republic
Galapagos Islands
Guatemala
Guayana
Honduras


Jamaica


LOCATION

Espirito Santo
Parana
Rio de Janeiro
Rio Grande do
Sul
Santa Catarina
Sao Paulo


Antioquia

Cartago

San Jose


SOURCEa

HS
HS
HS
HS

HS
HS
LR
HS
LR
LR

LR


Francisco
Morazan
Clarendon


Lesser Antilles


Guadeloupe
Martinique
Montserrat
Nevis


REFERENCE"

NYBG
NYBG
NYBG, Wikler 1999
NYBG, Wikler 1999

MBG, NYBG, Wikler 1999
MBG, NYBG, Wikler 1999
Weber 2003
MBG
McVaugh 1963
Standley 1937

Standley 1937

Roig and Mesa 1953
Wikler 2000
Weber 2003
McVaugh 1963
McVaugh 1969
MBG

MBG, NYBG
Fawcett and Rendle 1926
Wikler 2000


STATUS

Native
Native
Native
Native

Native
Native



Cultivated/ Possibly
Naturalized
Cultivated/ Possibly
Naturalized


Naturalized
Naturalized
Cultivated/ Possibly
Naturalized


Howard 1989
Howard 1989
Howard 1989
Howard 1989


1













COUNTRY/ ISLAND
CHAIN
Puerto Rico
Uruguay
Venezuela

ORIENTAL
Christmas Island
Fiji


Hong Kong
India
Japan
Malaysia

Philippines
Singapore
Sri Lanka

Tahiti
United States

WESTERN
PALEARCTIC
Azores

British Isles
Cape Verde Islands
France, Central
Mediterranean Isl.


LOCATION


SOURCEa


Taunovo
Viti Levu
Viti Levu


Bonin Islands
Sabah
Selangor


Seethaganguala


Hawaii


REFERENCE"

Liogier and Martorell 1982
Lombardo 1964
Pittier 1926


Wikler 1999
Wikler 1999
Wikler 1999
Greenwood 1944, 1949
Wikler 1999
Morton 1987
Wikler 1999
Wikler 1999
Wikler 1999
Morton 1987
Morton 1987
Wikler 1999
Fosberg 1971
NYBG, Wikler 1999
MBG, NYBG, Wikler 1999



Weber 2003


STATUS

Naturalized
Native


Invasive
Invasive
Invasive


Invasive


Introduced


Weber 2003
Weber 2003
Weber 2003
Weber 2003


1













COUNTRY/ ISLAND LOCATION SOURCEa REFERENCE STATUS
CHAIN
Madeira LR Lowe 1868
aHS denotes that the source is a herbarium specimen, while LR denotes a literature reference.
bMBG refers to the Missouri Botanical Garden and NYBG refers to the New York Botanical Garden















APPENDIX B
FINAL TEST PLANT LIST FOR HOST SPECIFICITY TESTING OF Tectococcus
ovatus

* Category 1 Genetic types of the target weed species (varieties, races, forms,
genotypes, apomicts, etc.) found in North America.

* Category 2 Species present in North America in the same genus as the target
weed, divided by subfamily (if applicable).

* Category 3 North American species in other genera in the same family as the
target weed, divided by subfamily (if applicable).

* Category 4 Threatened and endangered species in the same family as the target
weed divided by subgenus, genus, and subfamily

* Category 5 North American species in other families in the same order that have
some phylogenetic, morphological or biochemical similarities to the target weed.

* Category 6 North American species in other orders that have some
morphological, or biochemical similarities to the target weed

* Category 7 Any plant on which the biological control agent or its close relatives
(within the same genus) have been previously found or recorded to feed and/or
reproduce

* Category 8 Plants not closely related to weed, which have economic significance
and are grown in the same range as the weed in North America





AUTHOR COMMON NAME REASON FOR INCLUSION


Sabine

Sabine


strawberry guava

strawberry guava


PLANTS
CATEGORY 1
Family Myrtaceae
Subfamily Myrtoideae
Psidium cattleianum var.
cattleianum
Psidium cattleianum var. lucidium

CATEGORY 2
Family Myrtaceae
Subfamily Myrtoideae
Psidium friedrichsthalianum

Psidium guajava

Psidium guineense

CATEGORY 3
Family Myrtaceae
Subfamily Myrtoideae
Acca sellowiana
Eugenia axillaris

Eugeniafoetida

Eugenia uniflora

Myrciaria cauliflora


Costa Rican guava

common guava

Brazilian guava





feijoa, pineapple guava
white stopper

Spanish Stopper

Surinam cherry

jaboticaba


target weed

target weed


close related to target, grown as an minor
ornamental in FL
close related to target, grown as a minor
fruit cropl in south FL
closely related to target, possibly grown as
an ornamental in FL



commonly grown as an ornamental in FL
native Eugenia, also cultivated as an
ornamental
native Eugenia, also cultivated as an
ornamental
not native but commonly sold as
ornamental, preferred host of Caribfly
not native, ornamental fruit in south FL


O. Berg

L.

Sw.




(0. Berg) O. Berg
(Sw.) Willd.

Pers.

L.

(C. Martius) O.
Berg












PLANTS AUTHOR COMMON NAME REASON FOR INCLUSION


Pimenta dioica


(L.) Merr


Pimenta racemosa

Syzygium malaccense

Syzygium paniculatum
Subfamily Leptospermoidiae
Callistemon citrinus

Callistemon viminale

Eucalyptus camaldulensis
Leptospermum scoparium
Melaleuca quinquenervia

CATEGORY 4
Family Myrtaceae
Subfamily Myrtoideae
C, lp l,/1/ nlhe/ pallens
C, Ip'/i, %/,/e' zuzygium
Eugenia confusa
Eugenia rhombea
Mosiera longipes
1 f1 7I i hll'\ %fragrans


(P. Mill) J.W.
Moore
(L.) Merr. &
Perry
Gaertner

(Curtis) Staph

(Gaertn.) G. Don
ex Loudon
Dehnhardt
J.R. & G. Forst.
(Cav.) Blake


Griseb.
(L.) Sw.
DC.
Krug & Urban
(Berg) McVaugh
(Sw.) McVaugh


allspice


bay-rum tree

malay apple

Australian brush cherry

crimson bottlebrush

weeping bottlebrush

red river gum
broom teatree
melaleuca


spicewood
myrtle of the river
redberry stopper
spiceberry eugenia
mangroveberry
Simpson's stopper


not native, invasive in HI, minor
ornamental in FL
native in Caribbean, minor ornamental in
FL, extracts used in perfumes
common ornamental in FL

ornamental in FL

not native, but cultivated as an ornamental
in FL
not native, but cultivated as an ornamental
in FL
ornamental in US
invasive in HI, cultivated in FL
invasive in FL, had on hand


native threatened species
native endangered species
native endangered species
native endangered species
native threatened species
native threatened species


CATEGORY 5
Family Lythraceae
Ammannia coccinea


Rottb valley redstem native species


Rottb valley redstem


native species












PLANTS
Cuphea hyssopifolia
Cuphea micropetala

Decodon verticillatus
Lagerstroemia indica

Lythrum alatum

Family Melastomataceae
Rhexia lutea
Rhexia mariana
Rhexia nashii
Tetrazygia bicolor

Family Punicaceae
Punica granatum


Family Combretaceae
Conocarpus erectus


CATEGORY 6
Family Chrysobalanaceae
Chrysobalanus icaco

Family Nyssaceae
Nyssa sylvatica v. biflora


AUTHOR
Kunth
Humb., Bonpl. &
Kunth
(L.) Ell.
L.

Pursh


Walt.
L.
Small
(P. Mill.) Cogn


COMMON NAME
Mexican false heather
tall cigar plant

swamp loosestrife
crapemyrtle

winged Lythrum


yellow meadowbeauty
Maryland meadowbeauty
maid Marian
flordia cover ash


Pomegranite



button mangrove


icaco coco plum


Walt. swamp tupelo


REASON FOR INCLUSION
introduced species, common ornamental
cultivated species

native endangered species
introduced species, commercially
important
native species


native species
native species
native species
native threatened species


introduced, commercially important,
minor fruit crop


native, economically and environmentally
important



native species also sold as ornamental


native species also sold as ornamental













PLANTS AUTHOR COMMON NAME REASON FOR INCLUSION


CATEGORY 7
Family Annonaceae
Rollinia mucosa


Family Thymelaeaceae
Daphnopsis americana


(Jacq.) Baill


(P. Mill.) J.R.


wild sugar apple


bum nose


closely related to recorded host of T.
ovatus


closely related to recorded host of T.
ovatus


CATEGORY 8
Agriculturally Important Plants
Aquifoliaceae
Ilex cassine

Ilex x attenuata

Fabaceae
Delonix regia

Fagaceae
Quercus hemisphaerica
Lauraceae
Persea americana
Moraceae
Ficus aurea
Myricaceae
Myrica cerifera
Poaceae
Saccharum officinarum


L.

Ashe


dahoon holly

topal holly


(Bojer ex Hook)
Raf.

Bartr. Ex Willd.

P. Mill

Nutt.

(L.) Small

L.


royal poinciana


darlington oak

avocado

Florida strangler fig

wax myrtle

sugarcane


native, and common ornamental, as are
many Illex species
native, and common ornamental, as are
many Illex species

ornamental tree in S Florida


native, common hardwood tree

introduced crop tree, common in S Florida

native, common ornamental in S Florida

native ornamental

introduced, common crop













PLANTS
Rosaceae
Eriobotryajaponica
Prunus angustifolia
Prunus persica
Pyrus x lecontei 'Hood'
Rutaceae
Citrus limon
Citrus x paradise
Citrus sinensis
Cupressaceae
Taxodium distichum
Pinaceae
Pinus elliottii
Podocarpaceae
Podocarpus macrophyllus


AUTHOR COMMON NAME REASON FOR INCLUSION


(Thunb.) Lindl.
Marsh.
(L.) Batsch
Rehd.

(L.) Burm. F.
Macfad.
(L.) Osbeck

(L.) L.C.

Engelm.

(Thunb.) Sweet


loquat
chicksaw plum
peach
Hood pear

lemon
grapefruit
sweet orange

cypress

slash pine

southern yew


introduced, common ornamental/ fruit tree
native, ornamental
introduced crop tree
introduced, cultivated crop tree

introduced crop tree
introduced crop tree
introduced crop tree

native, common ornamental species

native, common ornamental species

introduced, common ornamental species















APPENDIX C
CHANGES TO THE FINAL TEST PLANT LIST FOR HOST SPECIFICITY
TESTING OF Tectococcus ovatus












Plants Removed From the Original Test Plant List
SCIENTIFIC NAME AUTHOR
Albiziajulibrissin Durazz
Cal ,pi uai(he' thomasiana Griseb.


Daphnopsis helleriana

Daphnopsis philippiana

Dirca palustris
Eucalyptus cinera
Eucalyptus grandis

Eugenia aggregata

Eugenia brasiliensis

Eugenia haematocarpa

Eugenia koolauensis

Eugenia reinwardtiana

Eugenia woodburyana

Pseudanamomis umbellulifera
Senna pendula


Syzygium cumini


Urban

Krug & Urban

L.
F. Muell. Ex Benth
W. Hill ex Maid

(Vell.) Kiaersk

Lam

Alain.

O. Deg.


Blume


Alain.


(Kunth) Kausel
(Humb. & Bonpl. ex
Willd.) Irwin & Barneby


(L.) Skeels


COMMON NAME
silktree
Thomas. Lidflower

Heller's cieneguillo

emajagua de sierra

eastern leatherwood
silver dollar tree
grand eucalyptus

aggregate eugenia

Brazil cherry

luquillo mountain
stopper
koolau eugenia

mountian stopper

Woodbury's stopper


monos plum
valamuerto


java plum


REASON FOR INCLUSION
Invasive according to IFAS Assessment
endangered in PR, unable to obtain, tested 2
native threatened Calypthranthes species
native to PR, could not obtain, tested D.
americana
native to PR, could not obtain, tested D.
americana
native to PR, could not obtain
introduced to Hawaii
introduced in FL, not common as
ornamental
not native, plenty of Eugenias represented in
test plant list
not native, plenty of Eugenias represented in
test plant list
endangered in PR, unable to obtain, tested
many native Eugenias
endangered in HI, in no danger from
proposed agent
native to Hawaii, in no danger from
proposed agent
endangered in PR, unable to obtain, tested
many native Eugenias
introduced, not common as ornamental
invasive in south Florida according to IFAS
Assessment

invasive, tested 2 Syzygium species, none
native to continental US












SCIENTIFIC NAME AUTHOR COMMON NAME REASON FOR INCLUSION
Syzygiumjambos (L.) Alston rose apple invasive, tested 2 Syzygium species, none
native to continental US
Syzygium samarangense (Blume) Merr. & Perry wax apple not native, tested 2 Syzygium species, none
native to continental US












Plants Substituted on the Original Test Plant List
SCIENTIFIC NAME AUTHOR
Ammannia robusta Heer & Regel


Cuphea aspera


Ficus benjamin

Lythrum curtissii

Lythrum flagellare

Prunus caroliniana

Pyrus communis

Rhexia parviflora

Rhexia salicifolia


Chapman


COMMON NAME
grand redstem

tropical waxweed


weeping fig


Fern.


Shuttlw. ex. Chapman

(P. Mill.) Ait

L.

Chapman

Kral & Bostick


Curtiss' loosestrife

Florida loosestrife

chrry laurel

pear

white
meadowbeauty
panhandle
meadowbeauty


REASON FOR INCLUSION
not present in release area, decided to
substitute more appropriate A. coccinea
highly endangered, collecting to test may
harm remaining population, testing 2
surrogate species
not native, decided to test a native Ficus
aurea
not grown as ornamental tested native
ornamental species L. alatum
not grown as ornamental tested native
ornamental species L. alatum
substituted closely related Prunus
angustifolia because of availability
tested a variety cultivated to grow in FL,
Pyrus x lecontei
tested 3 native surrogates R. lutea, R.
mariana, and R. nashii
tested 3 native surrogates R. lutea, R.
mariana, and R. nashii
















LIST OF REFERENCES


Abrahamson, W.G., Weis, A.E., 1987. Nutritional ecology of arthropod gall makers. In.
Slansky, F., Rodriguez, T.A., (eds.), Nutritional Ecology of Insects, Mites, Spiders
and Related Invertebrates John Wiley and Sons Inc., New York, 235-258.

ASU (Arizona State University) Vascular Plant Herbarium, May 2005. url.
http://lsvl.la.asu.edu/herbarium/.

Baranowski, R., Glenn, H., Sivinski, J., 1993. Biological control of the Caribbean fruit
fly, Anastrepha suspense (Loew). Florida Entomologist. 76, 245-250.

Barrett, M.F., 1956. Common Exotic Trees of South Florida (dicotyledons). University of
Florida Press, Gainesville, 414 pp.

Blair, C.A., Blackith, R.E., Boratynski, K.L., 1964. Variation in Coccus hesperidum L.
(Homoptera: Coccidae). Proceedings of the Royal Entomological Society of
London. 39, 129-134.

Boratynski, K., Davies, R.G., 1971. The taxonomic value of male Coccoidea
(Homoptera) with an application and evaluation of some numerical techniques.
Biological Journal of the Linnean Society. 3, 57-102.

Borror, D.J., Triplehom, C.A., Johnson, N.F., 1989. An Introduction to the Study of
Insects, Sixth Edition. Harcourt Brace Publishers, Orlando, 875 pp.

Bourgoin, T.H., Campbell, B.C., 2002. Inferring a phylogeny for Hemiptera: falling into
the "autapomorphic trap." Denisia. 176, 67-82.

Bretschneider, E., 1898. History of European Botanical Discoveries in China. Marston
and Co. Ltd., London, 1167 pp.

Britton, N.L., 1918. Flora of Bermuda. Scribner, New York, 585 pp.

Cal-IPC (California Invasive Plant Council), September 2005. Cal-IPC invasive plant
inventory. url. http://www.cal-ipc.org/pestplant list/

Campbell, B.C., Steffen-Campbell, J.D., Sorenson, J.T., Gill, R.J., 1995. Paraphyly of
Homoptera and Auchenorrhyncha inferred from 18s rDNA nucleotide sequences.
Systematic Entomology. 20, 175-194.









Chapman, R.F., 1998. The Insects: Structure and Function. 4th ed. Cambridge University
Press, Cambridge. 770 pp.

Cook, L.G., Gullan, P.J., Trueman, H.E. 2002. A preliminary phylogeny of the scale
insects (Hemiptera: Sternorrhyncha: Coccoidea) based on nuclear small-subunit
ribosomal DNA. Molecular Phylogenetics and Evolution. 25, 43-52.

Cronquist, A., 1981. An Integrated System of Classification of Flowering Plants.
Columbia University Press, New York, 1262 pp.

Da Costa Lima, A., 1927. Archives da Escola Superior de Agricultura e Medicina
Veterinaria. 8, 161.

Da Costa Lima, A., 1936. Terceiro Catalogo dos Insectos que Vivem nas Plantas do
Brasil. Directoria da Estatistica da Produccao, Seccao de Publicidade, Rio de
Janeiro, 460 pp.

Daly, H.V., 1985. Insect morphometrics. Annual Review of Entomology. 30, 415-438.

Degener, O., 1932. Flora Hawaiiensis. Honolulu, 273 pp.

Dehgan, B., 1998. Landscape Plants for Subtropical Climates. Univ. Press of Florida,
Gainesville, 638 pp.

Diong, C.H., 1983. Population ecology and management of the feral pig (Sus scrofa L.)
in Kipahulu Valley, Haleakala National Park, Maui, Hawaii. University of Hawaii,
Ph.D. Diss. Honolulu. 408 pp.

Dreger-Jauffret, F., Shorthouse, J.D., 1992. Diversity of gall inducing insects and their
galls. In. Shorthouse, J.D., Rohfritsch, O., (eds.), Biology of Insect-Induced Galls.
Oxford University Press, New York, 8-33.

Dyar, H.G., 1890. The number of molts of lepidopterous larvae. Psyche. 7, 189-191.

Ellshoff, Z.E., Gardner, D.E., Wikler, C., Smith, C.W., 1995. Annotated bibliography of
the genus Psidium with emphasis on P. cattleianum (strawberry guava) and P.
guajava (common guava), forest weeds in Hawai'i. Cooperative National Park
Resources Studies Unit, University of Hawai'i, Dept. of Botany, Honolulu.
Technical Report 95. 102 pp.

EXTOXNET (Extension Toxicology Network), February 2005. Pesticide information
profiles: ethylene dibromide (EDB), dibromoethane. url.
http://extoxnet.orst.edu/pips/edb.htm

Fawcett, W., Rendle, A.B., 1926. Flora of Jamaica. British Museum, London, 121 pp.









FDACS (Florida Department of Agricultural and Consumer Services), December 2004.
Minutes of the noxious weeds and invasive plant review committee meeting.
Division of Plant Industry. url.
http://www.doacs.state.fl.us/pi/enpp/botany/noxweed.html.

FDACS, June 2005. Caribbean fruit fly-free protocol. url.
http://www.doacs.state.fl.us/onestop/plt/cfffprotocol.html

Ferris, G.F., 1957. A review of the family Eriococcidae (Insecta: Coccoidea).
Microentomology. 22, 81-89.

Ferriter, A., 2003. Explore your alter-natives: a plant substitution guide for south Florida.
Wildland Weeds 6(4), insert.

FLEPPC (Florida Exotic Pest Plant Council), November 2003. List of invasive species.
url. http://www.fleppc.org/Plantlist/031ist.htm.

FNGLA (Florida Nursery, Growers, and Landscape Association), May 2005. State issue:
invasive plants. url. www.fngla.org/fngla-action/doc/Invasive%20Plants.pdf

Fosberg, F.R., 1941. Varieties of the strawberry guava. Proceedings of the Biological
Society of Washington. 54, 179-180.

Fosberg, F.R., 1971. Psidium (Myrtaceae) in Ceylon. Ceylon Journal of Science,
Biological Science. 9, 58-60.

Fox, A.M., Gordon, D.R., Dusky, J.A., Tyson, L., Stocker, R.K., 2004a. IFAS assessment
of the status of non-native plants in Florida's natural areas. Florida Cooperative
Extension Service, Institute of Food and Agricultural Sciences, University of
Florida. SS-AGR-225. 27 pp.

Fox, A.M., Gordon, D.R., Dusky, J.A., Tyson, L., Stocker, R.K., 2004b. The story behind
the IFAS assessment of non-native plants in Florida's natural areas. Florida
Cooperative Extension Service, Institute of Food and Agricultural Sciences,
University of Florida. SS-AGR-86. 6 pp.

Fox, A.M., 2005. Invasive plant lists of the southeast: a primer. Wildland Weeds. 8, 18-
20.

Frank, J.H., McCoy, E.D., 1995. Introduction to insect behavioral ecology: the good, the
bad, and the beautiful: non-indigenous species in Florida. Florida Entomologist. 78,
1-15.

Gaines, J.C., Campbell, F.L., 1935. Dyar's rule as related to the number of instars of the
corn ear worm, Heliothis obsoleta (Fab.), collected in the field. Annals of the
Entomological Society of America. 28, 445-461.









Gardner, D.E., Davis, C.J., 1982. The prospects for biological control of nonnative plants
in Hawaiian national parks. Cooperative National Park Resources Studies Unit,
University of Hawaii. Honolulu. Technical Report 45. 55 pp.

Gerlach, J., 2004. A 10-year study of changes in forest vegetation on Silhouette Island,
Seychelles. Journal for Nature Conservation. 12, 149-155.

Gilman, E.F., Watson, D.G., 1994. Psidium littorale, Cattley guava. Florida Cooperative
Extention Service, Institute of Food and Agricultural Sciences: University of
Florida. Fact Sheet ST-529. 3 pp.

Gomes, P., 1983. Fruticultura Brasileira. 9th Edicgo, Livraria Nobel, Sao Paulo. 448 pp.

Gordon, D.R., Thomas, K.P., 1997. Florida's invasion by nonindigenous plants: history,
screening, and regulation. In. Simberloff, D., Smith, D.C., Brown, T.C., (eds.)
Strangers in Paradise: Impact and Management of Nonindigenous Species in
Florida. Island Press, Washington D.C., 22-37.

Greany, P.D., Riherd, C., 1993. Caribbean fruit fly status, economic importance, and
control (Diptera: Tephritidae), Florida Entomologist. 76, 209-211.

Greenwood, W., 1944. Supplementary notes on the adventive and weed flora of the
leeward coasts of Fiji. Journal of the Arnold Arboretum. 25, 397-405.

Greenwood, W., 1949. Notes on some Fijian weeds and introduced plants. Journal of the
Arnold Arboretum. 30, 75-84.

Harris, P., Shorthouse, J.D., 1996. Effectiveness of gall inducers in weed biological
control. The Canadian Entomologist. 128, 1021-1055.

Heard, T., 1997. Host range testing of insects. In. Julien, M., White, G., Biological
Control of Weeds: Theory and Practical Application. ACIAR Monograph Series,
Canberra, Australia, 192 pp.

Hempel, A., 1900. As Coccidas Brasileiras. Revista do Museu Paulista. 4, 378-380.

Hempel, A., 1901. Descriptions of Brazilian Coccidae. Annals and Magazine of Natural
History. 7, 118-119.

Hempel, A., 1912. Catologos da Fauna Brazileira: Vol. 3. As Coccidas do Brazil. Museu
Paulista, Sao Paulo, 4v.

Henderson, L., 1989. Invasive alien woody plants of Natal and northeastern Orange Free
State (South Africa). Bothalia. 19, 237-262.

Hodges, C.S., 1988. Preliminary exploration for potential biological control agents for
Psidium cattleianum. Cooperative National Park Resources Studies Unit,
University of Hawaii, Dept. of Botany, Honolulu. Technical Report 66. 32 pp.









Howard, R.A., 1989. Flora of the Lesser Antilles: Leeward and Windward Island. Arnold
Arboretum, Harvard University. Jamaica Plain, 6v.

Hoy, J.M., 1963. A catalogue of the Eriococcidae (Homoptera: Coccoidea) of the world.
New Zealand Department of Scientific and Industrial Research Bulletin. 150, 1-
260.

Huenneke, L.F., Vitousek, P.M., 1990. Seedling and clonal recruitment of the invasive
tree Psidium cattleianum: implications for management of native Hawaiian forests.
Biological Conservation. 53, 199-211.

IFAS (University of Florida, Institute of Food and Agricultural Sciences), 2005.
Conclusion from the IFAS assessment of the status of non-native plants in Florida's
natural areas. Florida Cooperative Extension Service, Institute of Food and
Agricultural Sciences, University of Florida. 5 pp.

ISWG (Invasive Species Working Group), December 2003. Statewide invasive species
strategic plan for Florida. url.
http://www.dep.state.fl.us/lands/invaspec/2ndlevpgs/ISWG.htm

Johnson, M.T., 2005. Petition for field release of Tectococcus ovatus (Homoptera:
Eriococcidae) for classical biological control of strawberry guava, Psidium
cattleianum Sabine (Myrtaceae) in Hawaii. Institute of Pacific Islands Forestry,
USDA Forest Service Pacific Southwest Research Station. Volcano, HI.
Unpublished.

Jones, S.D., Wipff, J.K., Montgomery, P.M. 1997. Vascular Plants of Texas: a
Comprehensive Checklist Including Synonymy, Bibliography, and Index. The
University of Texas Press, Austin, 404 pp.

Kishi, Y., 1971. Reconsideration of the method to measure the larval instars by use of the
frequency distribution of head-capsule widths or lengths. The Canadian
Entomologist. 103, 1011-1015.

Langeland, K.A., Burks, K.C., (eds.), 1998. Identification & Biology of Non-native
Plants in Florida's Natural Areas. University of Florida, Gainesville, 165 pp.

Langeland, K., Hall, D., 2000. Exotic guavas in Florida: so delicious but-. Wildland
Weeds. 3, 4-11.

Langeland, K. 2002. Proper uses of FLEPPC's list of invasive species. Wildland Weeds.
5. 16-17.

Lepage, H.S., 1938. Catologo dos Coccideos do Brasil (Homoptera: Coccoidea). Revista
Do Museu Paulista. 23, 389-389.

Li, H.L., Huan, T.C. 1979. Flora of Taiwan: Vol. 6. Vascular Plants Checklist and Index.
Epoch Publishing Co., Taiwan, 6v.









Liogier, A.H., Martorell, L.F., 1982. Flora of Puerto Rico and Adjacent Islands: a
Systematic Synopsis. Editorial de la Universidad de Puerto Rico, San Juan.

Lombardo, A., 1964. Flora Arborea y Arborescente del Uruguay. Museo Nacional de
Historia Natural, Montevideo, Uruguay. 151 pp.

Lowe, R.T., 1868. A Manual of Flora of Madeira and the Adjacent Porto Santo and
Desertas. Jon van Voorst, London. 366 pp.

MacCaughey, V., 1917. The guavas of the Hawaiian islands. Bulletin of the Torrey
Botanical Club. 4, 513-524.

MacGillivray, A.D., 1921. The Coccidae: Tables for the Identification of the Subfamilies
and Some of the More Important Genera and Species Together with Discussions of
their Anatomy and Life History. Scarab Co., Urbana, 502 pp.

Marlatt, R.B., 1980. Susceptibility ofPsidium guajava selections to injury by
Cephaleuros sp. Plant Disease, 64, 1010-1011.

Marlatt, R.B., Alfieri, J.R., 1981. Hosts of a parasitic alga, Cephaleuros Kunze, in
Florida. Plant Disease. 65, 520-522.

MBG (Missouri Botanical Garden Herbarium), June 2005. url.
http://www.mobot.org/MOBOT/Research/herbarium.shtml

McVaugh, R., 1963. Flora of Guatemala, part vii. Fieldiana: Botany. 24, 283-405.

McVaugh, R., 1969. Myrtaceae. In. Maguire, B. (ed.) The Botany of the Guayana
Highland Part VIII. Memoirs of the New York Botanical Garden. 18, 55-286.

Merrill, E.D., Perry, L.M., 1938. The Myrtaceae of China. Journal of the Arnold
Arboretum. 19, 197-199.

Morton, J.F., 1987. Fruits of Warm Climates. Morton, J.F., Miami, 505 pp.

Nguyen, R.U., Poucher, C., Brazzel, J.R., 1992. Seasonal occurrence of Anastrepha
suspense (Diptera: Tephritidae) in Indian River County, Florida, 1984-1997.
Journal of Economic Entomology. 85, 813-820.

Noble, I.R., 1989. Attributes of invaders and the invading process: terrestrial and vascular
plants. In. Drake, J.A., Mooney, H.A., (eds.), Biological Invasions: A Global
Perspective. John Wiley and Sons Ltd., Chichester, 301-313.

NYBG (New York Botanical Garden Herbarium), June 2005. url.
http://www.nybg.org/bsci/herb/









OTA (Office of Technology Assessment, US Congress), 1993. Harmful non-indigenous
species in the United States. OTA-F-565, US Government Printing Office,
Washington, DC. 397 pp.

Pimentel, D., Lach, L., Zuniga, R., Morrison, D., 2000. Environmental and economic
costs of nonindigenous species in the United States. BioScience. 50, 53-65.

Pittier, H., 1926. Manual de las Plantas Usuales de Venezuela. Litografia del Comerico,
Caracas, Venezuela, 458 pp.

Popenoe, W., 1920. Manual of Tropical and Subtropical Fruits. Macmillan Co., New
York, 474 pp.

Pratt, L.W., Santos, G.L., Stone, C.P., 1994. A test of four herbicides for use on
strawberry guava (Psidium cattleianum Sabine) in Kipaiiulu valley, Haleakala
National Park. Cooperative National Park Resources Studies Unit, University of
Hawaii, Dept. of Botany. Honolulu. Technical Report 90. 24 pp.

Reichard, S.H., Hamilton, C.W., 1997. Predicting invasions of woody plants introduced
into North America. Conservation Biology. 11, 193-203.

Rohfritsch, O., 1992. Patterns in gall development. In. Shorthouse, J.D., Rohfritsch, O.,
(eds.), Biology of Insect-Induced Galls. Oxford University Press, New York, pp.
60-86.

Roig y Mesa, J.T., 1953. Diccionario Botanico de Nombres Vulgares Cubanos. Tomo II.
2da. Edicion Ampliada y Corregida. Seoane, Fernandez y Cia, Havana. 1140 pp.

Rung, A., Venable, G.L., Miller, D.R., Gill, R.J., September 2005. Scale families, family:
Eriococcidae. url. http://www.sel.barc.usda.gov/scalekeys/scalefamilies/Index.html.

Sabine, J., 1821. Account of a new Psidium in a letter to the secretary. by William
Cattley, Esq. F.H.S. Transactions of the Horticultural Society of London. 4, 315-
317.

Sakai, A.K., Allendorf, F.W., Holt, J.S., Lodge, D.M., Molofsky, J., With, K.A.,
Baughman, S., Cabin, R.J., Cohen, J.E., Ellstrand, N.C., McCauley, D.E., O'Neil,
P., Parker, I.M., Thompson, J.N., Weller, S.G., 2001. The population biology of
invasive species. Annual Review of Ecology and Systematics. 32, 305-332.

Schmidt, F.H., Campbell, R.K., Trotter, S.J., 1977. Errors in determining instar numbers
through head capsule measurements of a lepidopteran a laboratory study and
critique. Annals of the Entomological Society of America. 70, 750-756.

Schroeder, C.A., Coit, J.E., 1944. The cattley (commonly known as strawberry guava).
California Avocado Society Yearbook. 29, 44-47.









Schroeder, C.A., 1946. Priority of the species Psidium cattleianum Sabine. Journal of the
Arnold Arboretum. 27, 314-315.

Schuh, R.T., Slater, J.A., 1995. True Bugs of the World (Hemiptera: Heteroptera);
Classification and Natural History. Cornell University Press, Ithica, 336 pp.

Simberloff, D., 1997. The biology of invasions. In. Simberloff, D., Smith, D.C., Brown,
T.C., (eds.) Strangers in Paradise: Impact and Management of Nonindigenous
Species in Florida. Island Press, Washington D.C., 3-17.

Smith, C.W., 1985. Impact of alien plants on Hawai'i's native biota. In: Stone, C.P., Scott,
J.M., (eds.), Hawai'i's Terrestrial Ecosystems Preservation and Management
Cooperative National Park Resources Studies Unit, Honolulu. pp. 180-250.

Sorensen, J.T., Campbell, B.C., Gill, R.J., Steffen-Campbell, J.D., 1995. Non-monophyly
of Auchenorryncha ('Homoptera') based upon 18s rDNA phylogeny: eco-
evolutionary and cladistic implications within pre-Heteropterodea Hemiptera (s.l.)
and a proposal for new, monophyletic suborders. Pan-Pacific Entomologist. 71, 31-
60.

Standley, P.C., 1937. Flora of Costa Rica. Field Museum of Natural History, Chicago, 4v.

Stehr, F.W., 1991. Immature Insects. 2 vols. Kendall/ Hunt Publishing Co. Dubuque. 992
pp.

Swanson, R.W., Baranowski, R.M., 1972. Host range and infestation by the Caribbean
fruit fly, Anastrepha suspense (Diptera: Tephritidae), in south Florida. Proceedings
of the Florida State Horticultural Society. 85, 271-274.

Tunison, T., 1991. Element stewardship abstract for Psidium cattleianum. The Nature
Conservancy. url. http://theweeds.ucdavis.edu/esadocs/documents/psidcat.pdf, last
accessed January 2004.

Vitorino, M.D., Pedrosa-Macedo, J.H., Smith, C.W., 2000. The biology of Tectococcus
ovatus Hempel (Heteroptera: Eriococcidae) and its potential as a biocontrol agent
of Psidium cattleianum (Myrtaceae). Proceedings of the X International
Symposium on Biological Control of Weeds, 4-14 July, Bozeman, Montana. 651-
657.

Wagner, W.L., Herbst, D.R., Shomer, S.H., 1999. Manual of the Flowering Plants of
Hawai'i. Bishop Museum and University of Hawaii Press, Honolulu, 1919 pp.

Wapshere, A.J., 1974. A strategy for evaluating the safety of organisms for biological
weed control. Annnals of Applied Biology. 77, 201-211.

Wapshere, A.J., 1979. Recent progress in the biological control of weeds. EPPO
Bulletin. 9, 95-105.









Webb, C.J., Sykes, W.R., Garnock-Jones, P.J., 1988. Flora of New Zealand. Botany
Division DSIR, Christchurch, 867 pp.

Weber, E., 2003. Invasive Plant Species of the World: A Reference Guide to
Environmental Weeds. CABI Publishing, Cambridge, 548 pp.

Weems, H.V., Heppner, J.B., Fasulo, T.R., Nation, J.L., October 2005. Caribbean fruit
fly Anastrepha suspense. Featured creatures. University of Florida IFAS and
Florida Department of Agriculture and Consumer Services, Division of Plant
Industry. url. http://creatures.ifas.ufl.edu/fruit/tropical/caribbean fruit fly.htm

Wikler, C., 1995. Aspectos bioecologicos de Eurytoma sp. causador de galha-do-ramo do
aracazeiro. Universidade Federal do Parana, M.S. Thesis, Curitiba. 63 pp.

Wikler, C., 1999. Distribuicao geografica mundial de Psidium cattleianum Sabine e um
cecidogeno com possibilidades de utilizacao em control biologico. Universidade
Federal do Parana. Ph.D. Dissertation Curitiba. 135 pp.

Wikler, C., 2000a. Gall former as a potential biological control for strawberry guava -
Psidium cattleianum. Proceedings of the X International Symposium on Biological
Control of Weeds, 4-14 July, Bozeman, Montana, 667-671.

Wikler, C., 2000b. Psidium cattleianum, deliciously dangerous in Hawaii. Wildland
Weeds. 3, 5-10.

Wikler, C., Pedrosa-Macedo, J.H., Vitorino, M.D., Caxambu, M.G., Smith, C.W., 2000.
Strawberry guava (Psidium cattleianum) prospects for biological control.
Proceedings of the X International Symposium on Biological Control of Weeds, 4-
14 July, Bozeman, Montana, 659-665.

Wilkey, R.F., 1962. A simplified technique for clearing, staining, and permanently
mounting small arthropods. Annals of the Entomological Society of America. 55,
606.

Wirth, F.F., Davis, K.J., Wilson, S.B., 2003. Market economics of 14 potentially invasive
ornamental plant species in Florida. Department of Food and Resource Economics,
Economic Analysis Research Report, University of Florida. EIR 03-1. 29 pp.

Wunderlin, R.P., 1998. Guide to the Vascular Plants of Florida. University Press of
Florida, Gainesville, 806 pp.

Wunderlin, R.P., Hansen, B.F., December 2003. Atlas of Florida vascular plants. Institute
for Systematic Botany, Univ. of South Florida, Tampa. url.
http://www.plantatlas.usf edu.

Wyse-Jackson, P.S., 1990. Nesocodon mauritianus (Campanulaceae). Kew Magazine. 7,
113-117.















BIOGRAPHICAL SKETCH

Frank Wessels was born in 1980 in Indianapolis, IN, and moved to Minnesota

before finally settling in Columbus, OH, where he grew up. From a young age Frank

developed an interest in biology by exploring the inhabitants of a creek and small woods

near his house. This eventually led him to Tampa, FL, where he pursued a double major

of marine science and biology at the University of Tampa. In May of 2002, Frank

received his BS degree. Internships at the Department of Environmental Protection and

with Dow AgroSciences in Tampa piqued his interest in entomology. Frank decided to

pursue his master's degree in entomology at The University of Florida and has spent the

last two and a half years in Gainesville, FL.